<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Matt Pupa on Medium]]></title>
        <description><![CDATA[Stories by Matt Pupa on Medium]]></description>
        <link>https://medium.com/@mattpupa?source=rss-37652710b9a9------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*c4-xJDF8VwXQFxEujQP4kA.jpeg</url>
            <title>Stories by Matt Pupa on Medium</title>
            <link>https://medium.com/@mattpupa?source=rss-37652710b9a9------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 06:28:17 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@mattpupa/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Thinking Like a Product Manager vs a Data Analyst]]></title>
            <link>https://medium.com/@mattpupa/thinking-like-a-product-manager-vs-a-data-analyst-1f3757b4f535?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/1f3757b4f535</guid>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Mon, 06 Jul 2020 13:51:33 GMT</pubDate>
            <atom:updated>2021-12-06T01:42:33.440Z</atom:updated>
            <content:encoded><![CDATA[<h3>Not recognizing the difference can hurt your business.</h3><p>It also creates a disconnect that happens far too often. Insights from data don’t seem relevant. Time is wasted. Everyone is frustrated and the business suffers.</p><p>This scenario should be easily avoidable. Product managers have a great sense for a product and its users. They are usually analytical enough to understand data and metrics.</p><p>Data analysts have fantastic analytical skills, and can usually communicate well enough for a non technical audience to understand their insights.</p><p>If this is the case, why does the disconnect happen? <strong>It happens because the two are often coming from different perspectives.</strong></p><p>Consider this <a href="https://threadreaderapp.com/thread/1279062633489010688.html">tweetstorm</a> by <a href="https://medium.com/u/65a0a2db2464">Kaz Nejatian</a>, who talks about how a product manager should improve a water bottle. Kaz does a fantastic job breaking down how they should think about the bottle. This requires examining different users and what each of those users are looking for in a water bottle.</p><p>Thinking in this level of detail is important for product managers because they’re involved in the entire product lifecycle. They need to get a product from 0 to 1, 1 to 10, 10 to 100, etc. If they can think through all those scenarios that Kaz mentions and make smart decisions, it’s likely that the water bottle and the water bottle business will improve. Over time, this becomes harder to do.</p><p>This is usually when data analysts come into the picture. The product has matured and more company resources are now available. Data analysts are supposed to provide insights that will continue the growth/optimization of the product.</p><p>When data analysts do get involved, they come from a different perspective. Their thinking starts with the metrics of the business. In this case, they want to figure out and share what things related to water bottles will improve customer acquisition, revenue, or retention. Once they analyze data, they make recommendations to the Product Manager.</p><p><strong>This is where the disconnect starts.</strong></p><p>In many cases, there’s a slight difference in what matters to each person. Here’s one example…</p><p>According to Kaz, thinking about the type of end user for a water bottle (athlete, office worker, etc) isn’t the best way for a product manager to think about improving a water bottle. However, if the data clearly shows a difference in metrics between those types of end users, that’s something a data analyst will highlight in their analysis.</p><p>The analyst may make a recommendation to try to get other user types to copy the behavior of the best performing user type. However, this approach doesn’t match up with how the product manager has thought about a water bottle previously.</p><p>Since it’s a different way of thinking about a water bottle, the product manager may feel like they aren’t getting what they need. When this isn’t reconciled, the narrative can turn into “The product manager isn’t getting the data they need from the data analyst”.</p><p>It’s not that either person is wrong, it’s just hard to reconcile the two, especially when they are so grounded in how they each normally look at things. Regardless, this can stagnate product improvement and cause friction in the relationship.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CZcZRCy53tXVtGERx4f3dA.png" /></figure><p><strong>What’s the best way to solve this problem?</strong></p><p>I don’t have all of the answers, especially from a product manager perspective. From a data analyst perspective, I can say that bringing analysts into the product lifecycle as early as possible will help a lot.</p><p>I’ve been fortunate enough to be in that situation in the past and it was great. Not only did I work with a product manager to agree upon metrics we’d consistently look at for a new product, I was also able understand how developers would log the product data. I could make recommendations on how to store the data, so when I analyzed it and shared insights with the product manager, it would automatically be in the format we wanted it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7F0fGV5LalvraLNDXcZKqw.png" /></figure><p>I’m sure there are additional ways to tackle this. I’m hoping that as I get more into product, I’ll come up with some other helpful solutions that solve this problem.</p><p><strong>If you like this story, you can see what I’m up to with my new business…</strong></p><p><a href="https://www.madapup.com/">https://www.madapup.com/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1f3757b4f535" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Co-Created an Asynchronous Design Sprint]]></title>
            <link>https://medium.com/@mattpupa/i-co-created-an-asynchronous-design-sprint-d4152288b6c2?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/d4152288b6c2</guid>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Mon, 08 Jun 2020 17:08:12 GMT</pubDate>
            <atom:updated>2020-06-08T17:08:12.825Z</atom:updated>
            <content:encoded><![CDATA[<p>I’ve been participating in design sprints occasionally over the past five years. I was introduced to virtual design sprints in the Global Virtual Design Sprint last year. I recently started to wonder where they could go in the future. I was curious about how they could work with another interest of mine — asynchronous remote work.</p><p>I reached out to <a href="https://medium.com/u/7ec99f9441">Lee Duncan</a> who is the Enterprise Design Sprint Leader at IBM, and someone I’ve worked with in the last GVDS. He had the same curiosity that I had. We decided to team up and co-create the Manhattan Sprint Project.</p><p>Our goal for the project was to support the evolving needs of design sprints within our dynamically changing world. The way people work has become more digital, more remote, and more asynchronous. This shift has been accelerated even more by COVID-19. This evolution required challenging the inefficiencies around coordinating globally distributed schedules for meetings, dealing with internet connectivity issues, among other things.</p><p>We started out with the following principles about how the project should go…</p><p><strong>Edit the team. </strong>The make up of the team for something like this is very important. Not only did we want people who were experienced with design sprints, but we also wanted people who were experienced and comfortable with asynchronous work. Some personality types fit this better than others. We wanted to include people who we thought would benefit from this type of experiment.</p><p>We started out with 40 people interested in the project, which is much more than either of us anticipated. We decided to edit the team by requiring people to make a loom video within 48 hours of our project announcement explaining 1) why they were interesting in participating and 2) what was their experience with design sprints and asynchronous work. We ended up with 19 video submissions, which resulted in 3 teams of 7 people each (including Lee and I).</p><p><strong>A condensed design sprint process that focused more on the process than the actual outcome. </strong>We decided 3 days would be enough time to give us the insights we needed and would be enough work for us to manage. We focused on 4 main activities over the 3 days.</p><ol><li>How Might We (HMW) questions</li><li>Generating ideas and related outcomes to turn into prototype (if the team wanted to build something)</li><li>Producing lightning demos to get inspiration for a prototype</li><li>Producing solution sketches</li></ol><p><strong>We wanted to stick to deadlines. </strong>This was going to be our main way of measuring success for the first version of the manhattan sprint project. It was important to make people aware that they needed to be accountable to their team as well as the project overall. We had 2 deadlines each day. A deadline for completing the work (HMW, lightning demos, etc) as well as a deadline for voting on that work.</p><p><strong>We needed a tech stack that would make things run smoothly. </strong>Lee and I threw many ideas back and forth about what tools could enable the best possible outcome for the project. We ended up using slack, mural, and loom primarily. These worked out really well, and a lot of people enjoyed using loom for the first time. We also tried to implement Yac and Maze.design, but those didn’t really get used since their introductions were a little rushed.</p><p><strong>We tried to automate as much as we could. </strong>There’s the saying “You don’t want to see how the sausage gets made”. That’s pretty accurate for some of the automation attempts we had. We started off with a pretty extensive list of things we should try to do. We were thinking about automating slack reminders, automating voting when activities were completed, automating announcement of vote winners, automating user testing, and a few others.</p><p>Due to time constraints, we only were able to implement slack reminders and announcements on winning votes. Those were done by a mechanical turk approach. Although the notifications seemed automated to the teams, it was really me in the background typing those up a few minutes before they were sent out in slack.</p><p>I tried thinking about ways to do the other automations through no-code tools like Zapier, Airtable, and a few others. Since I didn’t own the slack account we were using, I realized those wouldn’t be possible. That was the biggest shortcoming of the project in my opinion, and something we need to improve on in the future.</p><p><strong>We wondered how to best enable asynchronous communication. </strong>In most design sprints, the facilitator is the one responsible for team building and leading the communication. We wanted to take that element out and see what would happen. We provided a few team building exercises in slack during the 3 days and some participants did come up with their own, but we noticed the team chemistry was lacking. That is another big improvement we need to make going forward.</p><p><strong>The challenge for the project was ‘How Might we redesign physical space for the world with COVID-19 and after?’</strong>. We chose this to push the boundaries and try to veer any potential solution away from a digital product. Sometimes design sprint solutions end up being a digital product no matter how ambitious the challenge starts off.</p><p>Throughout the 3 days, I followed along with each team as they went through all of the activities. I was really impressed with some of the work that came out of it. The ideas covered a lot of different concepts. The winning HMW questions were…</p><blockquote>HMW design spaces so that people can minimize their need to touch things?</blockquote><blockquote>HMW create multi-use spaces that can adept quickly, flexibly and easily to current needs?</blockquote><blockquote><strong>HMW rapidly find new uses for abandoned/unused spaces? (shopping malls, parking structures)</strong></blockquote><p>Below are a bunch of screen shots of lightning demos as well as solution sketches from the project.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/345/1*kJhcpxT-JCJ3jy_MFJ40jw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/414/1*Cv1mfY49kXGQbnUKRXzy3A.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/272/1*S6gNr8iSWhNA6NLbXDz-Bw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/355/1*rbIl8VIQPhAPi5jRPwbYNQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/334/1*c_xglvGfUoQ3UX02FU9y4A.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/351/1*TaHB-3L3qnQW18Rw2do6eA.png" /></figure><p>We learned a lot from the first iteration of the project. We tried to collect a balance of qualitative and quantitative insights to move forward with.</p><p><strong>We looked at a few metrics. </strong>I kept my eye on the deadlines to see how many people were getting their work done on time. Overall, I’d guess that the deadlines were hit 65–70% of the time. That was good enough for me for the first version iteration.</p><p>It’s hard to quantify, but I would love to know how many of those missed deadlines were due to real jobs/real life taking priority, how many were due to poor communication, and how many were due to working styles not matching that well with asynchronous work. Regardless of the reason, we want to improve this metric in the next iteration of the project.</p><p><strong>Feedback. </strong>After the 3 days, we sent out a retrospective for teams to write down any feedback they had. Here’s a screen shot below of one team’s retrospective.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aCRRUaVcNfUb6HTjJ2XbhA.png" /></figure><p>We also met with some team members in a zoom call, to give them a chance to give us their thoughts face to face. We got great feedback, and a few interesting ideas to incorporate going forward.</p><p><strong>Lee and I are already starting the planning for the next version of this. </strong>We will have more team building activities, better communication, better transparency, and better technology.</p><p>If you’re interested in hearing more about the project and what we’re thinking about going forward, you can watch the discussion that Lee and I had with <a href="https://medium.com/u/e224bb1c7dd3">Robert Skrobe</a> and <a href="https://medium.com/u/aae28dba08cf">Sandy Lam</a> on their podcast, Remotely Interesting.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fm61mTEFLYqk%3Fstart%3D970%26feature%3Doembed%26start%3D970&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dm61mTEFLYqk&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fm61mTEFLYqk%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/a257db766ae12ed01212136c38933d34/href">https://medium.com/media/a257db766ae12ed01212136c38933d34/href</a></iframe><p>It was a great discussion, and one that will continue over the next few months. I’m really excited to keep this effort going and see where it ends up.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d4152288b6c2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Maxima and Design Sprints]]></title>
            <link>https://medium.com/@mattpupa/maxima-and-design-sprints-cb120c559f66?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/cb120c559f66</guid>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[design-sprint]]></category>
            <category><![CDATA[experiment]]></category>
            <category><![CDATA[circulareconomy]]></category>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Fri, 05 Jun 2020 14:11:11 GMT</pubDate>
            <atom:updated>2020-06-05T14:11:11.403Z</atom:updated>
            <content:encoded><![CDATA[<p>One shortcoming of design sprints is that the resulting prototypes can sometimes be underwhelming. A team may start a sprint with a very ambitious idea, but by the end, they’re left with something that doesn’t seem to have much of an impact on the problem they planned to solve.</p><p>It’s important to know that this can still lead to a good outcome. A good way to think about this is borrowing the idea of maxima from statistics.</p><p>In a mathematical function, maxima include the largest value of the entire function (global maximum) and the largest value within a given range of the function (local maximum).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DRGSqrrmK4mF0MU77kseVg.png" /><figcaption><strong>Maxima in a Function</strong></figcaption></figure><p>Let’s loosely translate this concept to design. Our design problem can be the entire function and the different components of the problem<strong> </strong>can be the maxima based on their potential impact.</p><p>One design sprint team I worked on wanted to solve the problem of ‘redesigning coffee cup waste’. Something really ambitious like developing a biodegradable coffee cup made out of gourds could have been our global maximum. Growing awareness for a biodegradable coffee cup through an app could have been one of our local maxima. An app may be more feasible than the biodegradable cup, but it may not be as impactful.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*G1tMo1Tavf7SCU4Je1T19A.png" /><figcaption><strong>Redesigning Coffee Cup Waste</strong></figcaption></figure><p>When our team reached the prototyping phase, there was some feeling that we got away from our original ambition of solving this problem. Although our prototype may not have been the most impactful choice, does that mean it was a waste of a design sprint? I don’t think so. Here’s why…</p><p><strong>A local maximum can amplify the global maximum.</strong> DJ Khaled using snapchat is one example of this. Snapchat was very innovative for social media and communication when it came out. The global maximum of snapchat is the complex technology. A local maximum is DJ Khaled growing awareness for snapchat based on his usage of the platform. <a href="https://www.wired.com/2017/03/snap-ipo-dj-khaled/">DJ Khaled’s usage of snapchat was correlated with massive user growth as well as massive increases in company revenue</a>.</p><p><strong>Local maxima can be quicker and easier to execute. </strong>Smaller iterations that are executed more frequently can lead to more success than larger iterations that are more spread out and methodical.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*N37JRqYD7ENbXIKc_rfEhA.png" /><figcaption><strong>Shipping Cadence</strong></figcaption></figure><p>Here’s a soundbite of <a href="https://medium.com/u/67f5049293c7">Naval Ravikant</a> talking about how high shipping cadence is a great predictor of a company’s ability to achieve product-market fit.</p><h3></h3><p></p><p><strong>Local maxima can inform how well your team works together. </strong>This is more relevant for newly formed teams, but it’s still helpful. You can still discover your strengths, along with the skills that you need to add to you team. This will also (hopefully) build some team chemistry.</p><p>Even though it may seem frustrating or underwhelming in the moment, It’s always valuable to experiment with the local maxima in a design problem. It may not seem like it will have the biggest impact, but try it anyway. At worst, you’ll get learnings relatively quickly, with the option to experiment on a different maximum next. At best, you’ll realize that specific maximum has more impact than you thought.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cb120c559f66" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A Facilitation Brain Dump From the Global Virtual Design Sprint]]></title>
            <link>https://medium.com/@mattpupa/a-facilitation-brain-dump-from-the-global-virtual-design-sprint-332c01c4e86?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/332c01c4e86</guid>
            <category><![CDATA[design-thinking]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[design-sprint]]></category>
            <category><![CDATA[facilitation]]></category>
            <category><![CDATA[remote-working]]></category>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Wed, 03 Jun 2020 19:11:39 GMT</pubDate>
            <atom:updated>2020-06-03T19:11:39.920Z</atom:updated>
            <content:encoded><![CDATA[<p>In December 2019, I wrote about my experience participating in <a href="https://medium.com/u/e224bb1c7dd3">Robert Skrobe</a>’s Global Virtual Design Sprint (GVDS) for the first time. Although I had participated in design sprints before, this was a new experience for me because it was done remotely and it was also the first time I facilitated a sprint. <a href="https://medium.com/@mattpupa/designing-for-sustainability-with-the-global-virtual-design-sprint-a958ae3d130b">I wrote a recap of the activities, the timeline, and the things my team accomplished during the sprint.</a></p><p>Since the GVDS just occurred again in May, I wanted to write about my experience this time around. Now that I’m more comfortable with the process and don’t have to use as much energy to absorb everything, I’m in a better position to focus on the small details and nuances of facilitation and running a sprint team. I’m changing it up and doing a bit of a brain dump of things I noticed from the facilitation perspective. These include things I noticed in my teams that I need to improve on, as well as things I noticed from watching recordings of other teams’ meetings.</p><h4>Have a co-facilitator</h4><p>During a session, moving cards around and being responsible for interacting with the chat seem like small things that shouldn’t break the flow of a session too much. Unfortunately, they can. Have a co-facilitator that can take on these responsibilities for you.</p><h4><strong>Speak out-loud for a bit before you run a session</strong></h4><p>If you’re someone who naturally speaks quietly or quickly, this will get you in a better speaking rhythm for a session. I’m the worst version of this, and it’s amplified by a Boston mumble that makes me trip up my words enough as it is. I noticed at times it was like I was a mixture of a sloth and a hedgehog in the worst way. My energy was too relaxed but I was speaking too quickly. It’s something I need to be more aware of going forward.</p><h4>Don’t afraid to be strict on the exercises</h4><p>You may think you’re doing your team a favor giving them more time to fill out cards during an ideation session, but it’s a burden when they need to read through all those cards to check for duplicates and vote during an affinity diagramming session.</p><h4>Keep notecards concise</h4><p>That will make them easier to scan, read, and vote on.</p><h4>Remember to always summarize progress for the team</h4><p>At minimum, do this at the start of each day. Make sure people remember the context for the activity at hand. How does this relate to HMW, long term goal, sprint questions, etc. It’s shocking how easily results of an exercise can veer off from the original intent. If left unchecked, this can be a disaster for the results of your sprint.</p><h4>Be Flexible with homework if needed</h4><p>This mostly relates to the size of the team. If you’re going to have a team beyond the typical 7 person design sprint limit, think about requiring less lightning demos/solution sketches. If you still require 2–3 which is typical, going through all of those the next day will eat up a giant chunk of your meeting time.</p><h4>Keep digital whiteboards simple</h4><p>Mural boards became a giant pain towards the end of meetings when the boards got too full. It made it hard to interact, and hard to facilitate. Hopefully the technology improves so this isn’t a problem, but keep this in mind when thinking about team size and the amount of notecards you’re willing to use for an activity.</p><h4>Focus on one primary HMW</h4><p>The secondary HMW questions don’t really matter, and caused more confusion for the teams I was on.</p><h4>Be very careful about how you use pretotyping in a design sprint</h4><p>The idea of pretotyping is excellent, but it’s probably best served as a replacement for prototyping day. Using pretotyping as a replacement for discovery isn’t a good option for teams who are working on a problem they aren’t familiar with. Doing subject matter expert interviews and problem framing is a much better way to get the context you need for discovery. If you’re running an iteration sprint, or a sprint with a group that already knows the problem very well, pretotyping could work as a replacement for discovery.</p><h4>Be EXTREMELY clear on your target audience from problem framing</h4><p>Not having a clear target audience can cause confusion for the rest of your sprint and really stress out the team. If you decide to do a mapping exercise, which can be complicated to begin with, know that it’s OK to have multiple actors when mapping, but the experience should revolve around the target audience.</p><h4>Use a mapping exercise to understand the ‘current’ situation</h4><p>Using the map to try to come up with a future scenario seems to be too difficult, and causes more trouble than it’s worth. Another thing to note is that using pretotyping as discovery can make mapping much harder since it can cause a lack of understanding on what to map out.</p><h4>Learn how to take a step back during a sprint</h4><p>My brain sucks at multi tasking. When I’m so focused on facilitating a session and getting through it, and people feel lost during the sprint. I need to do a better job of processing that in the moment. After the sprint when people give detailed thoughts about what went wrong, I can try to process it and figure out where things went off track, and realize what I may have done to screw things up.</p><h4>Ask for feedback throughout the sprint</h4><p>…especially when people feel lost. If team members give you enough details, you can figure out where things got off track and fix it going forward.</p><h4>Watch others facilitate</h4><p>It’s is a great way to refine your skills, and you’ll notice small details that you can’t see when facilitating yourself. One quick example is that when someone was facilitating and sharing their screen on zoom, they weren’t zoomed in close enough on the mural board to easily read the notecards. This was something I never would have realized if I was facilitating myself.</p><h4>When speaking, make sure not to cut someone off as they finish what they are saying</h4><p>It can happen if you’re nervous and you usually won’t even realize it. Similar to this, avoid saying yes constantly as another person is speaking.</p><h4>Be self aware of your personality and facilitation style</h4><p>Know what personality types are complementary to you if you have the choice. I want people to do well and am good at moving people along. However, I struggle to multi task during a sprint, so my ability to do or say anything humorous is almost zero. It’s always great to have another person who can provide a little bit of humor throughout sprints.</p><h4>Use your facilitator powers when they’re needed</h4><p>Be more vocal about challenging assumptions. You don’t want to influence too much, but being more vocal can get teams out of a rut. You have multiple opportunities to do this, including 1) coming up with questions that could challenge the problem statement before SME script is made, 2) Making sure the HMW, sprint questions, and long term goal are aligned, and 3) Is the winning solution sketch ambitious enough.</p><h4>Be very aware of your tone of voice and energy throughout a session</h4><p>If people have questions, or need things repeated, keep the same tone as you speak. This can be more common if you’re working on a team with people who speak different primary languages. If your tone falls off or you start to sound disinterested in the conversation, that can be evident to other team members, and can be really off-putting. Some people do this without even realizing it, so make sure you watch recordings of yourself facilitating.</p><h4>The GVDS doesn’t get enough credit for being 3 weeks back to back for a facilitation learning tool</h4><p>Being able to get feedback/learnings on your performance and being able to turn that around the following week and make adjustments is super valuable. Reminds me of coaching back to back crossfit classes and making adjustments after each class, and seeing the results of those adjustments in the next class.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=332c01c4e86" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Droove — Documenting 10 Days of Building a No-Code Tool]]></title>
            <link>https://medium.com/@mattpupa/droove-documenting-10-days-of-building-a-no-code-tool-50f194040073?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/50f194040073</guid>
            <category><![CDATA[design]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[saas]]></category>
            <category><![CDATA[no-code]]></category>
            <category><![CDATA[visualization]]></category>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Fri, 31 Jan 2020 19:43:38 GMT</pubDate>
            <atom:updated>2020-01-31T19:43:38.532Z</atom:updated>
            <content:encoded><![CDATA[<h3>Droove — Documenting 10 Days of Building a No-Code Tool</h3><p><strong>Why I decided to build Droove</strong></p><p>I wanted to build a tool like Droove because I felt there was an opportunity in the digital workspace market. There were plenty of work management tools out there that helped teams stay organized, and track their work, but I thought most of them were missing a key aspect…visual interaction.</p><p>Most humans are visual creatures. Unfortunately, most companies don’t work that way. I wanted to build something that filled that gap. The need for something like this became clear to me in a few different times in my career.</p><p>The first was when I started consulting for a small remote team. I always wrote very clear emails and messages, and felt that I could communicate really well in written form. However, there were times when being able to draw and map things out on a whiteboard in person would have helped. The second was when I started working for very large company. I was on an analytics team and we had stakeholders all throughout the organization. I worked with product managers and engineers who were on different teams, had different messy reporting relationships to management, and had different goals for their teams. Not only that, but a giant re-structure happened while I was there, and it was never explained very well after it was finished. Although we used some of the popular tools to keep track of our work, I never felt like I had a clear idea of what teams were working on what, how things should be prioritized across teams, and what the true impact of my work was.</p><p>Even though I thought a tool like this could help a lot of people, I never thought I’d be able to build it. I knew a little bit of python, but definitely didn’t have the programming experience necessary, and I wasn’t going to look for outside help for something that may not come to fruition. It wasn’t until I started playing with no code tools that building something like this seemed more realistic.</p><p>I started exploring tools like Webflow, Zapier, and Airtable. I built my personal website with Webflow, and had used Zapier and Airtable for some small projects here and there. After getting comfortable with those, I knew I wanted to spend a few weeks pretty focused on Bubble. I quickly realized this tool had TONS of potential. I’d seen people build marketplace apps, websites, and a few other things with it. However, I hadn’t seen much with functionality in terms of dragging and moving (that’s where the name came from, for anyone who was wondering) elements around on an interactive workspace that could help tell stories and keep teams organized. At one point, I came across a plugin for dragging and dropping elements. When I found that, I knew I had to see if I could build this tool….entirely without code.The rest of this story is my experience each day of trying to build Droove.</p><p><strong>DAY 1–1/20/20</strong> ‍</p><p>Since I had some understanding of Bubble prior to starting the Droove build, the first thing I wanted to check out was the drag and drop functionality. I knew if that wasn’t as good as I hoped, this whole project was going to be over very quickly. Even though Bubble had some reference documentation, I just started playing with it.</p><p>I put a drag/drop group into my page, and then put a text element inside that group. When I clicked preview. I was happy to see that it worked. My assumption was that a drag and drop group allowed any element inside of it to be moved around. I made my drag and drop group the width and height of my screen, and put another element in there to test. When I hit preview, I realized my assumption was wrong. When clicked in preview mode, any elements inside that drag and drop group, were picked up at the same time, and moved around together. This wasn’t what I wanted, since I wanted to be able to move drag and move each element around on its own.</p><p>My next thought was that maybe you need to put a drag and drop group around each individual element that you want to move on the screen. I tested that next. I had two drag and drop groups on my page, with a text element inside each one. When I clicked preview this time, I was able to move the text elements individually on the screen. This was exactly the start I wanted to see. <strong>‍</strong></p><p><a href="https://drive.google.com/file/d/1HF7BJVwQz5NxhTMDqxK5HlJ_9qe_ScGq/view">startingwithdragdrop.mov</a></p><p>Now that I had that out of the way, the next part was building a workflow to get movable elements to appear on the screen at the click of a button. I wanted to do something cool for the homepage, while keeping the minimalist feel to it. I came up with this…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BFSpMqrAk7hACBFl.png" /></figure><p>The idea was that someone could get a little preview of how Droove worked by pressing the plus icon. A little text element would appear on the screen, and they could move it around. This took a little more work than I thought.</p><p>To get this to work, I knew that I needed to set up some type of data structure, set up several workflows, and use repeating groups along with drag and drop groups.</p><p>For the data structure, I set up a new type called ‘homepage demo’ with one additional text field. Since this was only going to be used on the homepage, and by anyone, I wasn’t worried about tracking the data (Maybe I should have been? What if these additions slowly accumulated over time and I ran into storage issues in Bubble? Wasn’t sure, but figured I could cross that bridge later).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/868/0*fVmzAsvXDkFJND2l.png" /></figure><p>Once I had the data structure set up, I added my ‘+’ icon to the homepage, and created a workflow to create a new record of ‘homepage demo text’ each time the ‘+’ icon was clicked.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/702/0*t7hvwhhgU7VzKLwa.png" /></figure><p>After that was set up, I went to the Droove homepage in preview mode, and clicked the ‘+’ icon 7 times. I went into the data page of bubble, and saw that the workflow was working, and that 7 records of ‘homepage demo text’ were created.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/961/0*FKbfkbOkDp5dhJzU.png" /></figure><p>Great to see…everything was moving pretty smoothly so far. The next step would be to now get those ‘Hello’ text elements to actually appear on the homepage when ‘+’ was clicked.</p><p>My original idea was that I would allow people to create unlimited ‘hello’ text elements as they wanted, and they could play with them freely on the homepage. I knew that I’d need to use a repeating group to produce the text elements, as well as a workflow to make them hidden on page load, but visible on a ‘+’ click.</p><p>The setup for the repeating group was important. It took awhile to figure out. Here’s the setup I used.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/507/0*Pg-6pjfCYGkenHpT.png" /></figure><p>After connecting these groups to the ‘homepage demo’ data structure, the idea was that each time the ‘+’ icon was clicked, a new inner group would be created within the repeating group, and then shown on the homepage. From there, someone would be able to click and move the text element around. Essentially, the red, green, and orange elements would stay in place on the page, but the blue and white elements would be clickable and draggable by the user. I was able to get this to work, but ran into 2 problems.</p><p>First, the text element could be moved around, but once it was moved into the inner group part of the layout, it visibly disappeared from the page. I thought this may have been from the order of the elements on the z-axis (front to back), but It was still happening after bringing the blue and white elements to the front over and over. My solution for this was to make the outer and repeating groups (red and green elements) the same size as the homepage itself, while keeping the inner group, drag drop group, and text element the original size and at the original X,Y coordinates on the page. Once I did that, the text element was able to be moved freely. Here’s a quick demo. <strong>‍</strong></p><p><a href="https://drive.google.com/file/d/1NuN6sHDB7noJYAP7G6t5j7J9YwHu_8--/view">homepagedemovideo.mov</a></p><p>This caused a different problem that I didn’t expect. Remember when I said that these groups would originally have to be hidden on page load, and then shown when the ‘+’ icon was clicked? Well that was happening, and now that the outer and repeating group were taking up the whole page, when they were shown, they were shown in front of the ‘+’ icon on the z-axis. This would mean that a user would only be able to create and move around 1 text element on the homepage. I thought about fixing this, but decided it wasn’t worth it. I figured users would get the point whether they created 1 element, or 5<strong>. </strong>I had more important things to worry about. Making this decision also temporarily eliminated the 2nd problem I mentioned with the repeating groups, but it was one I knew I’d face when I got to the UI.</p><p>DAY 2–1/21/20</p><p>My 2nd day of building Droove was focused on the main UI which was the visual workspace users would see when they signed up / logged in. I reproduced the repeating group setup I used before, but I knew this would get more complicated since I needed to build in more functionality.</p><p>The first thing was setting up a more robust data structure. I wanted to simplify the terms for the elements users were adding to their workspace. Rather than call it a ‘thing’ or an ‘element’, I decided to call it a ‘droove’. I figured that would cement the idea that a ‘droove’ could be created, and could be turned into almost anything, ranging from a simple note, an image, a file, all the way to a task that tracked work, like a traditional work management tool. As a first version, I set up a droove data type with fields of ‘droove color’ and ‘droove type’.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*fi926z0D779CSiPM.png" /></figure><p>I put in entries for 3 new records, giving the drooves colors of red, blue, and green, and types of note, note and note. This didn’t mean anything yet, so don’t worry about that too much. Just remember that I created 3 records. During day 1, I mentioned that I had a 2nd problem with the repeating group setup that I was going to need to fix. Well, now was the time to try and fix it.</p><p>In my head, I was picturing that each time a new droove was added to the workspace, it would appear on the screen at the same X,Y coordinates, and users could move it around from there. However, what actually happened is that each new droove that was created in my repeating group, would appear below the previous Droove. This had an effect of new drooves appearing to be hidden on the workplace, unless you scrolled down on the repeating group. I knew I needed to come up with a way to fix this. <strong>‍</strong></p><p><a href="https://drive.google.com/file/d/17yJC2DucqVQTWrtwSHvDvzitvc-PAMCC/view">newdrooverepeatinggroupissue.mov</a></p><p>Just an FYI, if you’re trying to compare the colors in screenshot from day 1 to the colors in this video, they don’t match up. I realized I probably should have kept them consistent, but I didn’t realize that until too late. Oops.</p><p>So this was the first big challenge I came across. How do I get new drooves to appear on the screen in the exact same spot each time?</p><p>My first thought was that maybe I needed another data structure for drooves. The first one that I already had would be for droove creation, and the second would be for droove activity. If there was a workflow trigger to move a droove from the droove creation data structure to the droove activity data structure, this may give me the visual effect I want on the repeating group elements. In programming, I believe this is called the fifo (first in, first out) approach.</p><p>After thinking about it, I realized this approach wouldn’t work. Although it may create the effect I want as far as new drooves from the droove creation data structure showing up on the same X,Y coordinates on the screen, if the droove was moved to a droove activity data structure, I’d still need a way to display the droove that was now in the droove activity data structure. In simpler terms, If a new droove that was created was dragged somewhere on the screen, it would still have to appear at some other X,Y coordinates when it was moved to droove activity.</p><p>A second thought popped into my head. What if there was a workflow that could be built that would automatically scroll to new drooves that were created? Similar to an assembly line, new drooves would be created, but the action (droove appears on screen), always happens at the same spot. In the workflow, there’s an element action to scroll to a certain element, with a pixel offset. I didn’t know what that was, so I played around with it.</p><p>After experimenting for a bit, I realized this wasn’t going to work. This scrolls the whole page to a certain element. You can’t choose to scroll within an element itself.</p><p>I wondered if there was a way to have a new record in a data structure show up dynamically somewhere on the screen. If so, maybe I could use the fifo approach I mentioned above?</p><p>Aha! I had an epiphany. I think I could use a workflow to hide the inner group element, and collapse its height, once a user clicked on the new droove that appeared on the screen. This should cause the next droove to be created to show up in the same spot.</p><p>This didn’t work either. I ran into weird workflow rules about which elements you could hide, based on which elements triggered the workflow. I just needed a way for the data in the repeating group in the groove creation data structure to be deleted, without deleting the new droove that I was creating. <strong>‍</strong></p><p><strong>DAY 3–1/22/20</strong> ‍</p><p>I was still stuck on the new droove problem from day 2. Fortunately, I had another epiphany while brushing my teeth the night before. I realized I hadn’t put a sort on the repeating groups. If I put a descending sort order on the repeating group of drooves by their creation date, the newest drooves should always appear on the same spot. I tried it, and I was correct! Another problem fixed, but of course, this caused 2 more.</p><p>Since I only wanted 1 new droove to appear on the screen at a time, I set up my repeating group to show 1 row and 1 column. This setup, along with having the repeating group take up the whole screen, was a problem. Each time a new droove was created, the current one was moved down, so the whole UI that the user was seeing on their screen was being replaced each time, which would cause any previous drooves to disappear…not good. I wasn’t sure what to do so I took a break for a few hours, hoping I’d come back with fresh eyes.</p><p>When I came back, I realized the 1 row 1 column repeating group problem wasn’t a hard problem to solve. Initially I was thinking that I’d want the drooves to appear in the middle of the screen, but having them appear in the top left corner was fine. I changed the repeating group to be 5 rows and 5 columns. Since new drooves would now appear in the top left corner and could now move freely around the screen, this would work for me.</p><p>The other problem I had to fix now was that when I dragged a droove that was just created to the different part of the repeating group on the screen, the next droove created after it would appear at that same spot, and not the top left corner like I intended. In fact, this was happening with all the drooves. If I moved any drooves around, when a new droove was created, all the drooves would move to the last location of the droove created right before it.</p><p>I realize this is confusing, so here’s a video showing what I’m trying to explain. I’ve numbered the new drooves coming in, so you can see the order they are created. <strong>‍</strong></p><p><a href="https://drive.google.com/file/d/1PUtJL_Oz0TT6c5MyBAkF_VTDfUxSPGV8/view">repeatinggroupsadddeletewithdrag.mov</a></p><p>You can see that each new droove that’s created is appearing on the screen in the same spot in the top left corner of the repeating group. That’s exactly what I wanted to solve on day 2, so that issue was now fixed. However, when I dragged a droove into a different spot, the new drooves created after it would appear at the dragged location of the one before it. That was a problem because it meant all the drooves on the screen would be starting at a different spot each time a new droove was added.</p><p>I wondered if I could set up a workflow that would move a droove to the location of the previous droove? I also wondered what would happen If I deleted drooves? How would this affect the movement? Did I need to set up a way to dynamically track x and y coordinates to know the location of drooves? Was that even possible? There were a lot of thoughts spinning in my head, and it was tough to keep track of all the details. I needed a little bit of a break. I took a few hours off.</p><p>After some google searching and checking out the bubble forum, I found a plugin that would track X and Y coordinates for an element. It took a little while to figure out, but it allows a bubble creator to link a tracker to 4 elements at a time. I wondered how I’d handle this if there were more than 4 drooves on the screen, but that was a question for later.</p><p>Through a lot of trial and error with a simple setup I made for testing, I figured out how to use the tracker and I was able to store an element’s X and Y coordinates in a sandbox data structure. I was happy about it, but I still needed to apply this logic to my repeating groups element that was showing all of the drooves, and that was connected to my actual droove data structure. This was a nightmare.</p><p>Because of how bubble lets you connect to a data structure in a workflow, I was only able to try and figure this out with the first or last item in my droove data structure. If I ended up having 10 drooves on the screen at once, I had no idea how I’d be able to use the tracker to appropriately track each individual droove. Maybe there was a way, but I felt like that was a limitation in Bubble I hadn’t figured out yet. I figured that was better than nothing for now. Unfortunately, that’s what I was able to figure out…nothing.</p><p>I ran into a lot of issues here with how the tracker interacted with the element hierarchy I had set up. I tried setting up the tracker outside of the outer group, within the outer group, within the repeating group, and at every other level of the hierarchy. Regardless of what I tried, I couldn’t get the X,Y coordinates to save in my droove data structure when I was using the tracker in this more complicated setup.</p><p>I spent a long time playing around with different ideas, whether that was continuing to change the position of the tracker, changing workflows, changing elements within workflows, but nothing seemed to work. I’d get errors from the plugin in preview mode. This was especially frustrating because it was a plugin from someone other than Bubble, so I had no documentation to help me fix this. After putting my mind into a pretzel, I decided it was enough for the day, and that I should step away. I was hoping a break would help me figure this one out. <strong>‍</strong></p><p><strong>DAY 4–1/23/20</strong> ‍</p><p>Before I opened up Bubble, I wanted to take a step back. I knew I was too deep into the weeds at the moment. Up to this point, I’d been using the repeating groups approach because that was my understanding on what was required to attach data to a data structure, and also show it on the screen…but what if there was another way?</p><p>When I was playing on my sandbox page, it was much simpler and I had more of the functionality I wanted. I could drag my few elements around the screen freely, without having to worry about the repeating group complexity. I assumed that this approach wouldn’t scale, because there wasn’t a way to link these elements to an actual data structure. I started to question whether that was a correct assumption. I got out a pen and some paper.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*hE3N5Vi9zmdw6zgY.jpeg" /></figure><p>I started sketching out the current approach I was taking (left), and an approach that I could maybe take (right). Since the benefit of the current approach was scale, and the benefit of the maybe approach was functionality, I decided I should try the maybe approach. If you don’t have a working product, being able to scale it doesn’t matter. I decided that I’d pre populate a droove data structure with 4 drooves, create 4 droove elements on my screen, and try to join (or link, for the non database folks) them with some type of ID.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*izq_LFLj00r4SjRV.png" /></figure><p>On my screen I had a pink element, which was the tracker, add and delete buttons that I’d try to incorporate later, and 4 drooves that I could move around on the screen. For the data structure, I set up a table with 4 records for 101 to 104, with droove number, and X and Y positions as the fields. X and Y positions were unpopulated, so I had to set up workflows to populate them when the drooves were moved around.</p><p>The data was easy enough to set up, but the workflows were not. Even though I wanted to have a 1 to 1 relationship (sorry for another database term) between a droove and a record in the table, the workflow wouldn’t do what I wanted it to do.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*niXM_9JkK-QEaUSZ.png" /></figure><p>Anytime a droove was clicked and dragged, the workflow was supposed to update the new X,Y coordinates. This was done by pulling the updated X,Y coordinates via the plugin, for step 1, and step 2 was to update the record in the table for the matching droove number (101, 102, 103, 104).</p><p>I think the workflow struggled with the ‘1 to 1’ and ‘1 to many’ relationship between the droove and the data. Although I was trying to link 1 droove to 1 record by filtering the list of drooves on droove number (1 to 1), the workflow was recognizing it as me trying to update a list of records(albeit filtered) based on 1 droove (1 to many). I played around with the workflow, but nothing was working. I realized my best bet was to create a table for each droove with 1 record each. Very annoying, but I’d see how it would run.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sdaLEGxRCU5qxFWJ.png" /></figure><p>Although it was very time consuming to set this up manually, it did work as intended. I could move any of the droove around on the screen, and when I looked at the data, the X and Y coordinates were updating correctly.</p><p><a href="https://drive.google.com/file/d/1HXWDXRIRMK92JTnm_24FCzU7A51VpQst/view">drooveXYupdates.mov</a></p><p>In addition to knowing the X and Y coordinates of the drooves for the repeating group approach I worked on the first 3 days, I also expected it to be important for users who were logging out and logging back into the Droove platform. I wanted to make sure users could move drooves around in their workspace, and they would stay in place each time they logged into the product. I figured I’d test that next.</p><p>For this, I had to build sign up and login functionality. Thankfully I had done this before, so I set up a pop up on the homepage that collected email, passwords, and could determine if people were signing up or logging in to an account they created previously.</p><p>Once that was done, it was time to see if having the X and Y coordinates would do anything on its own. As the video shows, it did not. When I moved my drooves around, logged out, and logged back in, they were in their original places, not the places I moved them to. <strong>‍</strong></p><p><a href="https://drive.google.com/file/d/1oVCzP7naclqGkXhu_ixCodbP5MfN83r5/view">droovelogoutlogin.mov</a></p><p>This was going to be something else I’d have to solve. Was it possible to keep those elements in place? If so, how? Maybe I needed to use custom states? Something else? All I knew was that it was midnight, and I was done with another day of working on this. <strong>‍</strong></p><p><strong>DAY 5–1/24/20</strong> ‍</p><p>Day 5 was a good day. I was getting pretty discouraged that this may not end up where I want it to. The new approach I was taking seemed way too clunky, and the original approach with the repeating groups seemed too ambitious.</p><p>I spent a lot of the morning and afternoon researching bubble X and Y coordinates on forums, slack groups, and google searches. At one point, I came across a post on the Bubble forum by Emmanuel, who is one of the founders of Bubble. It was the post announcing the new draggable elements plugin back in 2016. In the post, there was some conversation about storing X and Y coordinates, for people trying to do things similar to what I was trying to do now. At one point in the thread, he mentioned that Bubble added a way to save the X and Y coordinates.</p><p>This caught my attention because I thought the only way I was able to do it was with the plugin I installed. I went back to the workflow, and created a new one without the plugin, and was stunned when I got one to work. Not only did it work, but I got it to work in the repeating groups approach I initially used. This was a HUGE step forward, which I was super excited about. It was also one of those moments where taking a long break helped. It stunned me how quickly I figured this out…one of those moments everyone has where you think ‘How did I not get this before?’. Better late than never.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sfrxRp1IZXXU9XqA.png" /></figure><p>With my new workflow error free, I tested it out. I moved some drooves around in the repeating group while in preview mode, and went to look at my data. My X and Y position fields were now populated.</p><p>In addition to that, I also figured out how to get drooves to stay in the same position when a user left the page, and came back later. I went back to my sandbox for this for simplicity, but I set up a workflow for the page load that would move the draggable elements to their X and Y position that was now being saved in the droove data structure. It definitely wasn’t as smooth as I hoped, but it worked. I was willing to take it, for now. I was hoping I could figure this out in the repeating groups as well. If I could, then I’d be able to move on to more of the work management part of Droove. <strong>‍</strong></p><p><strong>DAY 6–1/25/20</strong> ‍</p><p>I didn’t spend much time on Droove this day, but I did get a little bit done. I wanted an easy way to delete drooves, and I think I figured it out. Along with storing X and Y coordinates, the draggable element plugin also has something called a drop area. This is an element that can be put on a page, and when an element is dragged over it, it will trigger a workflow. The simplest way to think about this is probably inserting a key (the element) into a lock (the drop area), and having a door open (the workflow).</p><p>My intention was to have this drop area trigger the deletion of whatever droove was dropped into it. For simplicity’s sake, I didn’t want this drop area to be on the screen the whole time. I only wanted it to show up if someone had clicked and was pressing on a droove. I was able to set up conditional logic for the drop area to only show it when that click and press happened.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*_Zd4Ea09RA2YA0dt.png" /></figure><p>The next step was setting up the workflow to trigger the actual deletion of the droove. This was pretty straightforward, thankfully. After I set this up, I made sure that my draggable droove was able to be dropped. I tested it out in preview mode, and it worked.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*01B2ntbyShW8HExd.png" /></figure><p>It seems to be a running theme as I’ve written this whole process so far, but 2 new issues came out of this. First, I could no longer drag drooves freely around the screen, unless I was dropping them in the drop area. If I tried dropping them somewhere else, they moved back to their original position. Second, the deletion of the droove seemed to be more related to the position of the repeating group vs. the droove itself. Here’s a video to explain what I mean. <strong>‍</strong></p><p><a href="https://drive.google.com/file/d/1wf_xquWLa-zuzmO1GTLLnt95OggNTVn_/view">dropareadeletedroove.mov</a></p><p>In the video above, you can see that when I drag droove 15 around the screen, I move it towards the bottom left of the repeating group. When I let go of my mouse, it automatically moves back to where I initially clicked it from. It seems that having a drop area may override the functionality to drag elements freely. Eventually, I drop it in the drop area, and droove 15 is deleted.</p><p>Next you can see that droove 16 is the only droove in the repeating group. If I click the ‘add’ button, a new droove appears, but droove 16 disappears. If I add a few more, droove 16 eventually reappears. This would be a problem if my drop area causes drooves that still exist, to temporarily disappear. <strong>‍</strong></p><p><strong>DAY 7 1/27/20</strong> ‍</p><p>I took a day off from working on Droove. At this point, there were 4 problems that I needed to solve.</p><ol><li>How to move drooves around and place them freely, even while using a drop area</li><li>How to make sure drooves don’t temporarily disappear when new drooves are added (also related to drop area).</li><li>If a droove is added/deleted, how to make sure the drooves are moved to the correct spot automatically</li><li>Can I move the drooves to their last saved X and Y position whenever the page is loaded instantaneously (This was currently on a delay, and very visible to the user)</li></ol><p>I probably had more in my mind, but these were the ones that I knew I had. I didn’t want to get burned out on building this, so I decided to shift my focus to more of the work management part of the tool. This would include a page for reporting on drooves, how to comment on a droove, how to change the appearance of a droove, and so much more.</p><p>I wanted to keep this part of the tool as simple as possible…at least to start. I wasn’t sure exactly where to start so I drew out some sketches. One sketch was for how an individual droove might work, and how to manipulate it based on what users want it to be. The idea is that a droove could be text box for notes, an image, a file, a task, or maybe something else.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*aZWL6nu7ImiAPP9c.jpeg" /></figure><p>The rough idea I had was that a droove would start as a drag/drop element, with text/image/file groups hidden or behind it. Based on the droove type that user selected, the appropriate group would then be shown, and users could interact with it accordingly.</p><p>The other thing I started sketching was the general layout of the UI. To start, I figured I needed a page for the team/account management, a page for reporting, a page with the drooves to show current tasks, and maybe a page for stakeholders to submit a droove.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sbv8_FvgLdQIE2gn.jpeg" /></figure><p>I made a few visuals of what these could look like, just so I could have something to work towards. I decided I would work on the team page first.</p><p>The team page needed to have a few different features, including the ability to add team members, delete team members, and add an avatar for the team. Setting permissions was also an option that made sense, but I decided I’d leave that as a possible addition for the future. I kept this page really simple to start, and created the necessary workflows. To get the add/delete team mate functionality working, I created a few pop ups that would appear on the page if a user chose to add or delete team members.</p><p><a href="https://drive.google.com/file/d/111Y4sTLBnxH33FM1uzSc_SScSyJY3TBi/view">drooveteampage.mov</a></p><p>Once I had the team page done, I moved on to the task page. For the most part, the task page was a clone of the team page. The only adjustments I needed to make were to have a task repeating group showing instead of a team member repeating group, as well as updates to the pop ups. The add task pop ups needed additional fields, including the stakeholder, the category, any files related to the task, and the team member that was assigned to it. This was pretty easy to do, so I was on to the next thing.</p><p>Before working on the data structure for the repeating groups in these pages, I wanted to build a navigation for when a user was logged in. I liked the very simple left navigation layout that youtube uses, so I decided to mimic that. I created a long gray bar running along the left side of the page, and made it a reusable element with different page options. Even after setting my ‘floats relative to’ in the appearance pop up, it wouldn’t stay in a fixed location when I scrolled in preview mode.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/946/0*qaAxFqtqEvz33IJN.png" /></figure><p>I played around with this, and figured out that once a floating element was converted to reusable, it no longer would float on the page. Maybe there was a way to fix this, but I wasn’t able to easily find it. I decided to start over on my nav bar.</p><p>My next attempt did work. I knew that a floating element that wasn’t reusable would stay in a fixed location on page scroll, so I built a blank nav bar on the left side of each of my pages. Once I had that done, I set it to float relative to the ‘Top’ and ‘Left’, and confirmed that it worked in preview mode.</p><p>For the reusable element, I created a smaller group with the different page options. Once that was done, I placed that element inside the floating nav bar I had on each page. Once again, I was able to confirm this stayed in place when I scrolled down the page, and also that the navigation actually worked between pages.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/835/0*9yfVBdTUrTzrkeir.png" /></figure><p>Now that the navigation was done, I had to think a little bit more about the data structure. I wanted Droove to be able to be used by multiple teams, so I needed to think more comprehensively than I was before when I was mainly building things for myself.</p><p>I knew I had the pre-built User table in Bubble, but I had to think about the logic a little bit. Should the User table be only for Users who sign up for their team? What about team members they add to their team? Should those additions go in a new table, or could they just be a list field in the User table? Depending on which choice I made, how would that affect functionality? What if users on different team names have the same user name? How do I make sure this gets distinguished when their data is shown? There were a lot of questions I hadn’t dealt with before. I decided to just start building, and adjust any issues I ran into along the way.</p><p>There was definitely a point of trial and error, but I eventually figured it out. I originally had a ‘teammates’ field in the pre-built User table that I set up as a list. I thought I’d be able to use a repeating group to show all the teammates for the current user on the team page. Unfortunately, I couldn’t get that to work. After doing some research, it seems like repeating groups can show different records in a table, but not different elements of a list within a record. This also created a problem because at scale, there wasn’t going to be an easy way to guarantee that all the teammates of the current user that had signed up for Droove would end up in that teammates field.</p><p>I ended up with the pre-built User table, and a new Team table. I needed to figure a way to join them, so I set a field in the User table called ‘User Team Name’. I set the field type as a ‘Team’, which would take the data from the Team table. This is known as a composite field type, which was new to me. I don’t feel like Bubble has the best documentation around it, because I had a hard time understanding how it worked at first. My original goal was to have a user create a team when they signed up, and a workflow would 1) create a new record in the Team table with that team name, and 2) Modify the user record in the User table to have the created team name as their User Team Name field. Step 1 worked, but step 2 did not. I tried a few different things via a workflow, but I get kept getting type inconsistency errors between the data types of ‘team’ and ‘text’. Eventually I was able to get a workflow working, but instead of putting the team name in the ‘User Team Name’ field in the User table, it put the unique ID for the team instead. I figured this was close enough, so I changed my ‘User’ table to have a field of ‘Team ID’.</p><p>I went back to my team page, made some adjustments to my repeating group, and was now accurately seeing the team members I expected to see. <strong>‍</strong></p><p><strong>DAY 8 1/28/20</strong> ‍</p><p>I didn’t do much work in Bubble. I took a day to go back and try to find answers the questions I had at the start of day 7…</p><ol><li>How to move drooves around and place them freely, even while using a drop area</li><li>How to make sure drooves don’t temporarily disappear when new drooves are added (also related to drop area).</li><li>If a droove is added/deleted, how to make sure the drooves are moved to the correct spot automatically</li><li>Can I move the drooves to their last saved X and Y position whenever the page is loaded instantaneously (This was currently on a delay, and very visible to the user)</li></ol><p>The reason these were so important, is because they were directly related to what makes droove different…having the ability of visual interaction. After a few hours of research, I came to the unfortunate realization that I wouldn’t be able to build out Droove as much as I wanted. I was hoping that wouldn’t be the case, but I knew I may run into limitations with Bubble when I started this project.</p><p>I was at an interesting crossroad at this point, with a few options. I could stop working on Droove altogether, and move on to something else, which would give me more time back, but I’d have no user insights from building it. I could continue with a prototype where I could have a few draggable elements, but it would be more of a ‘duct tape and band aid’ approach which probably wouldn’t be the best experience for a user. I could also build Droove out as just a super simple work management tool without the visual elements, and see if there was maybe a niche market that would find it valuable. I had a decision to make.</p><p><strong>DAY 9 1/29/20</strong> ‍</p><p>I decided I’d go with the ‘duct tape and band aid’ approach. Even though I wasn’t going to be able to build the product I was hoping to, I figured doing this would at least allow me to show something for my efforts, and hopefully glean some insights. I also decided that I would put a deadline on this…14 days of work. <br>‍</p><p>At this point, the repeating groups approach was officially done. It just wouldn’t scale how I wanted it to. I was back to my less scalable, very clunky, approach.</p><p>For this to work, I needed to build the drooves. From my sketch on day 7, I had an idea of putting a draggable group on the page. Within that draggable group would be a multiline input group that users could take notes on, a picture uploader, and a file uploader, and a button that would create a new task. These elements would be hidden when the page was loaded. When users initially clicked on a droove, a pop up would show asking them what type of droove they wanted, whether it was a note, a picture, or a file. Based on what they chose, a workflow would start that would transform the droove into what they wanted. From there, it could be moved around. <strong>‍</strong></p><p><a href="https://drive.google.com/file/d/1mJ3P_C24jflO-h34L6WNkF_1z5Xk8FTq/view">droovenewdroove.mov</a></p><p>As you can see, this didn’t work how I wanted. I was able to create a note, but the drag functionality wasn’t as good as I hoped. <strong>‍</strong></p><p><strong>DAY 10 1/30/20</strong> ‍</p><p>Even though I said I was going to stick with a duct tape and band aid approach to Droove, I decided to stop working on it. The more I thought about the issues I was trying to overcome, the less confident I felt that I’d be able to build a prototype worth sharing.</p><p>The visual workspace just wasn’t going to get to where I wanted it to be. Even though I could have continued with a super simplistic work management tool without the visual workspace, I felt like this was a saturated market already. I knew about tools like google sheets and trello, but I also found other tools like taco app. These were tools that were on the more simplistic side of work management, and had a freemium pricing model. The fact of the matter was that these tools could do what I wanted to build, if not more, for free.</p><p>It would have been nice to have a prototype to show for my work, but hopefully writing this experience is enough that either I, or someone else finds it useful in some way. I think this will be a good reference for me to go back to if I come across similar issues in the future.</p><p>As I wrap this up, I have a few parting thoughts…</p><ol><li>Even though Bubble has a lot of functionality, it still does have its limitations. I feel like I chose a pretty ambitious product idea for a no-code solution, but I’m happy about that. I feel like a learned a lot about the nuances of the tool, that I may not have learned, if I decided to build something more traditional that had already been done before.</li><li>Should I only try to build things that I’ve validated in some way with other users? In the case of work management, I know there is a desire to have more visual tools. Tools like Mural, Notion, as well as Ideapaint, prove the fact visual is good. I’ve also seen plenty of evidence that supports it. However, I never explicitly asked anyone if they’d be interested in a tool like this before I started. I was OK with that in this case, because it was a way for me to learn Bubble, while trying to build something I was interested in. Now that I’m familiar with Bubble, that probably will be more of a requirement for projects I take on going forward.</li><li>Is it worth building something in no-code if it’s not financially viable? Again, I could have built Droove out as work management tool, and that would probably be a pretty cool no-code accomplishment, but I don’t think it would have a chance of making money. There’s got to be a balance between building something in no-code because you can, vs it actually making/saving money.</li><li>Should I livestream videos instead of writing in the future? This is something I struggle with. One one hand, live streaming definitely seems better from a content perspective, as it’s easier to explain in real time what’s going on. However, I enjoy writing because I think it’s a better skill for me. Also, going back between writing and doing, helps me think through problems, and come up with new ideas. I don’t know how this would translate if I was live streaming my work.</li></ol><p><em>Originally published at </em><a href="https://www.mattpupa.com/blog/droove-documenting-10-days-of-building-a-no-code-prototype"><em>https://www.mattpupa.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=50f194040073" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Designing for Sustainability with the Global Virtual Design Sprint]]></title>
            <link>https://medium.com/@mattpupa/designing-for-sustainability-with-the-global-virtual-design-sprint-a958ae3d130b?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/a958ae3d130b</guid>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Tue, 10 Dec 2019 14:37:47 GMT</pubDate>
            <atom:updated>2019-12-10T20:32:01.045Z</atom:updated>
            <content:encoded><![CDATA[<p>Earlier this fall, I started researching ways to get more involved in design sprints. I’ve always been drawn to sprints because of how quickly concepts can be tested, and how they can combine business acumen with creativity. I was especially hoping to get experience facilitating a sprint. I had experience facilitating people from my time as a crossfit coach, but I wanted to transition that skill into sprints. On twitter, I came across the <a href="https://medium.com/dallas-design-sprints/announcing-the-next-global-virtual-design-sprint-november-4-22-2019-4a635fb63e40">Global Virtual Design Sprint</a> organized by <a href="https://medium.com/u/e224bb1c7dd3">Robert Skrobe</a>. It’s an online design sprint event with people all over the world who choose to work together on different challenges relevant to the world today. This was also appealing to me because it was a completely virtual experience. I worked remotely in my last role, as well as in a stint doing some consulting for a small non profit startup, but getting more experience on how to work better virtually is something I’m always attracted to.</p><p>The GVDS used <a href="https://medium.com/u/bb76103e9f36">Mural</a> as its primary working space, where all of the participants interacted. I’m a very visual person, so I couldn’t have been happier about this. I’ve been in meetings where people are walking through things without any visual aids, make mistakes, backtrack, and use acronyms that aren’t known, and it usually ends in confusion. As I scanned the mural early on to see what the participants did for work and where they were from, I reached out to a few folks who incorporated design sprints into a corporate role. I’ve always been interested in using sprints more in my own career, so I was eager to get advice on how to do that. I was fortunate enough to reach out to <a href="https://medium.com/u/7ec99f9441">Lee Duncan</a>, who is a design sprint leader at IBM. He gave me some great tips for my own career, and separately, he brought a great challenge idea to the GVDS…Designing for the circular economy with a focus on mobilizing awareness and action for regenerative products and services.</p><p>All of the GVDS teams were primarily organized by challenge topic as well as geographical location, with a few exceptions. My team had 7 primary team members located throughout the US, Europe, and Australia, as well as a few other GVDS participants observing our sprint via slack and zoom. The team was made up of <a href="https://www.linkedin.com/in/matthewpupa">myself</a>, <a href="https://www.linkedin.com/in/lee-duncan-design/">Lee Duncan</a>, <a href="https://www.linkedin.com/in/fatimakamali/">Fatima Kamali</a>, <a href="https://www.linkedin.com/in/joshuafryszer/">Josh Fryszer</a>, <a href="https://www.linkedin.com/in/fatima-bambo-ja%C3%AFtay-20227563/">Fatima Bambo-Jaïtay</a>, <a href="https://www.linkedin.com/in/karakus-mehmet/">Mehmet Karakus</a>, and <a href="https://www.linkedin.com/in/mary-beth-hill/">Marybeth </a>Hill. We had a great mix of experience in design, research, facilitation, and business, and it really helped us throughout the sprint. Even though we had some very experienced facilitators on our team in Lee and Mehmet, they took on more advisory roles, which allowed me to step up and facilitate a large portion of the sprint. I’m very thankful for that, since it has given me a lot of confidence running sprints in the future.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*esa1TMdyNmcp-TPI1MYt1g.png" /></figure><p>The GVDS ran like a traditional design sprint, having activites around framing the challenge, interviewing subject matter experts, developing lightning demos and ‘How Might We’ questions, solution sketching, story boarding, developing a prototype, and user testing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jDrS-vkJwlNiS7VyqU7azg.png" /></figure><h3>Pre Sprint Activities</h3><p>Before the design sprint week officially began, we did some contextual work in order to frame the problem, so we could understand the true problem we were trying to solve for. We used affinity diagramming to figure out WHO has the problem, WHAT is the nature of the problem, WHERE/WHEN does the problem occur, and WHY is the problem worth solving. Once we had those answers, we transformed them into a singular problem statement. Looking at the yellow post it below — we decided to focus our efforts on consumers who purchase products, but aren’t aware of how the supply chain process of a product being created, can have a negative impact on the environment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*L-BTgToyJOX4ragoBhSpwQ.png" /></figure><p>We then moved on to the subject matter expert interviews to get additional insight for our design sprint week. We interviewed consumers who either wanted to purchase with more of a sustainability mindset, or those who were already purchasing with a sustainable mindset. We got a lot of great insight from the interviews, with some common themes being ‘convenience is most important to me’, as well as ‘I’m willing to pay slightly more for sustainable products, but not a lot more’. With those insights learned, we were ready to move on to our design sprint week.</p><h3>Day 1</h3><p>The purpose of the first day of the sprint was to develop our ‘How Might We’ questions (HMW), our sprint questions, and our long term goal (LTG). This involved A LOT of affinity diagramming. For that, we took all the feedback that we heard from the interviews, and put each insight onto a separate post-it in mural. Once we had them all written down, we grouped all the similar post-its into clusters. Once the clusters were completed, we voted on the post-it we thought was most relevant for the exercise we were doing. We repeated this throughout the day and kept building off the results of previous diagramming.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*97lMkvlJmt-pzUlR4AiiCQ.png" /></figure><p>After the first day, it was very clear that our team had good chemistry working together. We were very collaborative in our thinking, open to others’ opinions, and moved through all the activities smoothly. We were ready for day 2.</p><h3>Day 2</h3><p>The second day of the sprint involved lightning demos, mapping the user journey, and choosing target areas to focus on within the user journey. The lightning demos were a way for our team to get some ideas flowing, and to help our creativity come out. We all took time on our own to research ideas/services/concepts that existed out in the world, that were somewhat related to the challenge we had. We each presented our demos, while the rest of the team took notes on things that caught their attention. I really enjoyed this exercise because it’s one that can connect many dots that you may not think can be connected. I’m always fascinated by the idea of seeing one idea or way of doing something, and then applying that concept to something else that will help it thrive.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DfqnMrKJ5llGwf_PsbyAHw.png" /></figure><p>Mapping the user journey was one of the more involved exercises of the entire design sprint. This was where we really had to think about the users, what goal they have relative to our sprint questions and long term goal, and then map out the steps for them to reach that goal. It really breaks the journey down into steps that you may not think of initially. We wrote user journeys for online consumers, offline consumers, and small businesses. In the end, we voted on the online consumer.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZgNkgj65K4D-SnmBFFO7DA.png" /></figure><p>The last piece was looking at our user journey, and identifying the target areas that we felt most aligned with our HMW questions. Within our user journey, we chose the ‘typing in search’ criteria step as our main target, because we felt it aligned with our HMW about making sustainable options more convenient. We also chose ‘search results’ and ‘checking reviews’ as supporting targets because we felt they aligned with our HMW questions about easy access to sustainability data and sparking a movement for sustainable production.</p><h3>Day 3</h3><p>The third day felt like the turning point in our sprint. In any sprint, people can be a little skeptical in the beginning because things can be a little ambiguous, and not tangible yet. I believe Lee used a term called ‘The valley of despair’ for this. Day 3 was about getting out of that valley, and to a point where we felt like we would have a tangible prototype that we could share with users. The activities that helped us get to a tangible prototype were solution sketching and storyboarding.</p><p>For the solution sketches, we built off of the user journey and the targets defined in day 2, and each person on the team sketched out their own idea of what a prototype could be to answer the HMW questions. We had quite a variety, ranging from an desktop browser extension, to sustainability filters embedded in an online shopping experience, all the way to using blockchain to make sure retailers prove the sources for their products and production. All of the sketches could have been made into a prototype, but in the end, we voted on a Sustainability Browser Extension. Two of the most attractive elements about this sketch were that they focused on both convenience for the user, as well as making sustainability data available.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/788/1*anPoxiy7DAMSidlLNa_NGA.png" /></figure><p>The last step we had before the actual prototype was built was putting the storyboard together. For this, we took the browser extension concept, and wrote all of the steps that a user would go through while using the extension during an online shopping session. We had to keep our targets from day 2 in mind, so we made sure to align the storyboard with the HMW of making sure finding sustainable products could be convenient. We wrote storyboards individually, and the team voted on the best one at the end. This exercise was a great one for our team. As the facilitator, I probably let it go on a little too long, but we did have a great discussion around what made sense for a prototype and what didn’t, as well as what elements from other solution sketches could be used for what we wanted to build. That required challenging others’ opinions within the team, but we were able to do it respectfully and thoughtfully. It was just another example of how well this team worked together. After the storyboard was completed, our designers worked hard to have our prototype ready for day 4.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/792/1*Jzt8v5DpiAdqrdavn7CExQ.png" /></figure><h3>Day 4</h3><p>The fourth day of the sprint was a nice little break after the intense third day. We felt more certain about the progress we had made to this point, because we now had a tangible prototype to show for our efforts. Our team used <a href="https://medium.com/u/bf1152b11387">Figma</a> to put a 15 slide shopping experience together for our user tests. Again, I was really impressed with what our designers turned around after just one day. Both Fatimas, Marybeth, and Mehmet did an awesome job. We decided to name our prototype ‘Sustain.ly’. In other cases, it may have been more appropriate to spend more time to focus on a proper name. Being transparent here, we decided on that name pretty quickly, but we were all happy with it. When we met online on day 4, we reviewed our prototype and took notes on things we wanted to change before we had users go through it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1Cf2dUzdygA_E5cuHjcPNQ.png" /></figure><p>The other offline exercise that was part of day 4 was developing a guide and script for our researchers to use when they were conducting the user tests. Josh, who was a researcher on our team, took on this responsibility, and developed a great guide that helped us get really good feedback from our users. He also facilitated the sessions for day 4 and 5, since they were more focused on design and research. Now that we had our prototype in a good place, and had a proper interview guide, we were very excited to get users to try it out and give us their feedback.</p><h3>Day 5</h3><p>Since we had some scheduling issues, we cheated on our design sprint a little bit. We pushed the sprint into two weeks, so we actually held the user tests over the weekend in between weeks. We had 8 user tests set up, which is a lot, but we had plenty of help with Mehmet, Marybeth, Josh, and Fatima all conducting the tests. Watching their different styles was great for me to see, especially because it’s a skill I’m new to, and need to develop as I get more involved in design. We learned so much from the user tests, and some things that made us smack our own heads and say ‘How did we not realize that?!?’ when building the prototype. It can be amazing which blind spots are revealed when others are looking at something you’ve built. As we were watching the user tests, we took notes on post-its and grouped them into positive/negative/informational/insightful statements.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zcbCAnIjwuw_qJH8_RzTqA.png" /></figure><p>Once all the user tests were done, it was time for us to reflect on the design sprint as a whole. We had 3 specific questions we had to answer to help us understand how we did. We used a voting session to do this.</p><blockquote>Did the prototype answer the sprint questions?</blockquote><blockquote>Will the prototype properly address the long term goal?</blockquote><blockquote>Do you think we addressed the challenge we took on as a team?</blockquote><p>When we looked back on our sprint, our answers were mixed. The borders on the images below show to winning vote. We voted on each sprint question individually.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EREYhhxodTkB4dJki10U-Q.png" /></figure><p>I was really happy this exercise was done at the end of the sprint. It made us look at our work critically, and think about what we could have done differently. Could we have stayed better aligned to our HMW questions, sprint questions, and long term goal as we progressed through the sprint? Did we think ambitiously enough to try to tackle our long term goal? Although I think our team did a great job on the sprint overall, I want to make sure that I keep these results in mind for the next sprints I work on, and maybe use them to adjust how I do plan to do things.</p><p>Because our team felt we did a great job overall, and we worked so well together, we do have plans to continue this work going forward. That may mean building a minimum viable experiment (MVE), presenting the sprint results to companies with similar interests, or building on this idea again in the next GVDS. I’m not sure of the exact path, but I’m excited to keep this work going.</p><h3>GVDS Showcase</h3><p>Now that our design sprint was officially over, it was time for us to showcase our work. We met with Robert and we walked through our entire week, highlighted our sprint questions, the prototype, the user test feedback, whether we validated our challenge, and our personal highlights for the sprint. I mentioned it earlier, but being able to facilitate a sprint like this when I didn’t have any previous experience doing it meant a lot to me, and has given me a lot of confidence to do it more going forward.</p><p>The GVDS showcase was a great way for others to see what we accomplished, and we were able to watch other teams’ showcases as well. There were so many challenges that were worked on, with some really cool prototypes coming out of them. I know Robert and the powers that be, are working hard to make the next GVDS even better, and I can’t wait to participate and see what happens.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a958ae3d130b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Using Analytics to Build, Test, and Grow a New SaaS Offering (Updated)]]></title>
            <link>https://medium.com/@mattpupa/using-analytics-to-build-test-and-grow-a-new-saas-offering-updated-4f2e30153da1?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/4f2e30153da1</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[customer-retention]]></category>
            <category><![CDATA[saas]]></category>
            <category><![CDATA[ab-testing]]></category>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Thu, 04 Jul 2019 16:37:26 GMT</pubDate>
            <atom:updated>2019-07-04T16:37:26.902Z</atom:updated>
            <content:encoded><![CDATA[<p>I wouldn’t say I fit the description of an indie hacker, since my formal background is in analytics, and I’ve mainly worked at established publicly traded companies throughout my career. However, I consider myself pretty lucky because I’ve had the opportunity to work on a start up initiative within one of those companies. It was the best of both worlds. I was able to be part of a team that built a SaaS product from 0 to over 500K customers, with minimal risk.</p><p>A large part of why the team was successful was because of how we used analytics to guide our decision making. The first part of this article will cover what we did to build, test, and grow the SaaS product. The second part will cover two additional analytics projects I originally created for separate company inititaives, but were used to help grow the SaaS product.</p><h3>Context</h3><p>Around the start of 2013, the senior leadership team of my company decided to make a big change to our business model. Our company switched from selling stand-alone software products, to selling an all-in-one platform with a good/better/best pricing model. Depending on which package you were in, you would have access to certain products/features. Overall, we believed this change would be beneficial for both our company, and for our customers.</p><p>In order to bring this idea to life, senior leadership put together an internal start-up team that was made up of about twenty five people. There was some representation from different departments in the company including marketing, analytics, billing, sales, support, and a few others . I was one of three analytics members on the team. Although we collaborated on everything analytics related, each of us had areas of work that we were most involved in. Among other things, we were tasked with providing some input into product features, leading quick and iterative A/B testing, building easy to consume reporting and analyses, and giving updates to C-level executives. For this post, I’m going to skip “other things” and “providing input into product features” mainly because they were pretty basic or were primarily done by the other two analysts. I’m going to give a little more detail on the three areas I was involved in and contributed to the most.</p><h3>A/B Testing</h3><p>The first thing to figure out for testing was how much of a sample size we would need. We based this on two things. 1) To get to, or as close as we could to, statistical significance with our testing. That would give us confidence in the results we were seeing. 2) If this completely tanked ($0 revenue), that we wouldn’t harm the overall business. Keep in mind that although we were treating this like a start-up, we were still part of a public company with revenue targets.</p><p>Like most SaaS businesses, we have a funnel that narrows down from website visitors &gt; free trialers &gt; paying customers &gt; cancelled customers. The metrics we traditionally used for this funnel are Visitor:Trial rate (V:T), Trial:Pay rate (T:P), Average Revenue Per User/Customer (ARPU), and Cancel rate. After playing around with different conservative scenarios based on historical trends, we figured out that if we took 10% of our overall website traffic to put into this new experience, we should have enough sample to test with. We also worked with the financial analyst on the project to confirm that 10% wouldn’t put our revenue targets in jeopardy.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fXcLQ5-a4yQQRsLTS5RHkw.jpeg" /></figure><p>Along with the 10% of website traffic that we were putting into this new experience, we held out another 10% of our old business experience as a control group. The control group was held out of any promotions, offers, or any other special activity that we may have been offering at the time. That would keep the control group clean for the best analysis. <strong>For this project, we added a new metric that was our primary measure of success. It was called Average Revenue Per Visitor (ARP-V). </strong>It combined the four previous metrics into one that gave a simple view into whether or not the new experience as a whole was working better than our old experience. An example of ARP-V is below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*-_XcZkGG8FgysQBTbO7Yqg.jpeg" /></figure><p>If you follow the funnel example from start to finish, you’ll see that the control group has a much higher T:P rate than the test group. However, the test group has a higher V:T rate, a higher ARPU, and a lower cancel rate. When you divide the total revenue by the starting visitor number, you get more revenue from each visitor in the test group than you do in the control group.</p><p>The last part of the testing was to figure out how long we would have to run each test. As I mentioned before, the goal for testing was to test, learn, and iterate quickly. We wanted to make as many tweaks as possible in the shortest amount of time. After playing around with more scenarios of how some of these early tests might go, we came up with about 14 days for each test iteration.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*Kx-bITyhsbadZC1s0BbWUg.jpeg" /></figure><p>After running a test for 14 days, we would take any learnings we could (click tracking data, feedback from the sales reps, etc), to make tweaks to our website, the product UI, pricing, and more. In total, we went through about twenty iterations of testing. Originally, we had a goal of 5% lift in ARP-V. By the end of our testing iterations, we were able to achieve close to 20% lift in ARP-V.</p><h3>Reporting and Analysis</h3><p>Once we had our testing plan in place, we needed to be able to see how our new experience was performing, so we could make appropriate recommendations. To do this, we built some advanced excel reporting that was updated daily. In addition to the funnel metrics mentioned earlier, this reporting showed product engagement metrics like % of trailers starting our product flow, % of trialers uploading their contacts into our system, and the % of customers logging into the product in the last 90 days. Each of these detailed metrics were indicators of some of the funnel metrics, so if funnel metrics weren’t performing well in the new experience, we know which engagement metrics to look at. Based on what we saw, we would know if it was worth tweaking the product for the next iteration of testing, or if the sales reps may need to adjust how the pitch the new experience. Below is a sketch of what some of the visitor and trial reporting looked like in excel.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*SHmxW_IYuXF_uHfzec92AQ.jpeg" /></figure><p>That’s still a little vague, so I’ll walk through a specific example of how we used the reporting, along with an ad-hoc analysis we did occasionally, to discover insight about trial performance in the new experience.</p><p>One of the biggest metrics we struggled with was Trial:Pay (T:P) rate. Whether it was the new UI, pricing, or sales positioning, we always had a decline, or a gap, between the new experience and the old experience. The reporting showed this was consistent no matter which test iteration we were looking at. We brainstormed with the product team to figure out if there was a change we could make to the UI that would close the T:P gap.</p><p>As I mentioned before, we have a few product engagement metrics that we know are indicators of a higher T:P rate. One of them, which is a little intuitive, is successfully getting through our product flow. We also looked at T:P rates of trialers in other segments including 1) starting the product flow, 2) clicking a button on the UI homepage, 3) having a sales interaction only, no UI, and 4) no interaction at all. Based on the % of trialers falling into each of these segments, combined with the T:P rates of these segments, we were able to figure out which segment was the biggest cause of the overall T:P gap we had (basically a weighted averaged).</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*JaYjeiaoxrAd0yJS1KXirQ.jpeg" /></figure><p>It was the segment of trialers who were starting the flow. The T:P rate for this group was similar between the new experience and old experience, however, there were less people starting the flow in the new experience. If we could get more trialers in the new experience to get into the product flow, our overall T:P gap should close.</p><p>After collaborating with the product team on different changes we could make for the next test iteration, we agreed on the change below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*ExCdXZKLDOPvH5vH2W-8tg.jpeg" /></figure><p><strong>Our hypothesis was that the initial UI homepage that trialers saw was too busy with some of the menus and additional features we added in to the new experience, and trialers were getting overwhelmed. </strong>We simplified the UI by giving trialers the ability to only click the button for the product flow. Once they went into the product flow, they would see the additional menus whenever they went back to the homepage. This change was successful, and did close the overall T:P gap we had by a few percentage points.</p><h3>Updates to Senior Leadership</h3><p>A very important aspect of this initiative was how we communicated with the rest of the team, the executives, and the company as a whole. We had a different communication based on whether it was a daily, weekly, month, or ad-hoc update.</p><p>Every day, we’d send out an email to the start up team on the results of the latest test iteration. This mostly included the funnel metrics, and whether or not we were seeing statistically significant differences. If there was anything special to note, we’d include that as well.</p><p>Once a week, we would send an email out to start up team as well as the executives. This would include the funnel metrics for the latest test iteration, as well as insights we found, and updates on the changes we were going to make like the UI change I highlighted above.</p><p><strong>Whenever we were giving updates to the team, we always used a Fact/Meaning/Action structure.</strong> This was the simplest and most impactful way we were able to get our message across and make progress. The Facts are the actual results of the test. In other words, what the data is telling you. In the UI change, the <strong>facts</strong> were that we were seeing a 30% increase in trialers starting the product flow, a 10% increase in trialers finishing the product flow, and the T:P gap close 2%. All of these results were statistically significant. This <strong>meant </strong>that trailers were more successful in the new experience, and more likely to buy based on ease of use over the appearance of many features. The <strong>action </strong>was that we kept the simplified UI as the default in the following test iterations.</p><h3>SaaS Product Recap</h3><p>Due to the 20% lift we saw in ARP-V, our executive team made the decision to launch the new experience to 100% of our web traffic after just 7 months of testing.</p><p>That wouldn’t have been possible if the 25 person start up team didn’t work so well together. I think it was a great example of what can happen when you put great employees together, and empower them to do great things.</p><h3>Additional Project 1: What if we don’t have statistical significance?</h3><p>In a few cases where A/B test results were to close to be statistically significant, but not quite there, we supplemented them with an A/B testing forecast that I had previously built. That gave us a little more confidence in the recommendations we were making. This is a great tool that can be used for many bootstrapped SaaS products, especially when the user base is still small and sample sizes in tests are an issue. Here’s the how and why of the testing forecast being built originally…</p><p>Awhile back, the conversion team at my company came to our analytics team about doing testing on our website. We used to have a trial sign up form that was very long, that required 8–10 data points including name, email, industry, etc. This was overkill if you compare it to a lot of digital products today that want to get people into their UI as fast as possible.</p><p>The conversion team wanted to test out a much shorter trailer form, and see what that would do to our conversion rate. We were happy to help them out, but there was one big challenge…they were hoping to do a series of 5–6 tests over the next 3 months.</p><p>That was extremely difficult to do with our business model. Not only did we usually need to let tests run for a few weeks to get enough sample size, we also offered a 60 day trial period. To get the complete test results and have them be statistically significant, it could take us almost 3 months to get a read on 1 test, never mind 5 or 6. In order to solve this problem, I build a testing forecast that would help us make decisions on test results in 3–4 weeks.</p><p>To provide some background on the metrics being measured, the conversion team typically ran tests that affected our website sign ups, which we measured by Visitor to Trial Rate (V:T), as well as tests that affected our customer conversion rates (Trial to Pay Rate (T:P). For this forecast, the primary measurement combined the two into Visitor to Pay Rate (V:P). This would offset differences between the two metrics. For this shorter trial form test, although our hypothesis was that a shorter trial form would lead to a higher V:T rate, we were concerned that T:P rate may decrease because the trialers may not be as invested. V:P let us see the net effect of that, and would help us declare a winner.</p><p>As far as the forecast went, The V:T part was pretty straight forward. We would let the test run for a few weeks and track V:T rates for the test group(s) and the control. I included inputs that would allow people to add X number of additional days the test could run for, with Y total visitors being added each day. The forecast would then take those input values, appropriately split them between the test/control groups, and apply the historical V:T rates that we had seen so far. This forecasted visitor data, would be added to the historical visitor data, and we would have a much bigger sample size.</p><p>The T:P part of the model was a little more in depth. Now that we had our Total visitor and V:T data (Historical + forecasted), the model next needed to trend out what the T:P data would look like. In order to do this, I captured T:P rate snapshots at 7 day increments up to 56 days (Not a perfect 60, but close enough for my purposes). Because most of our conversions came in the first few days of our trial period, our conversion rate was a logarithmic curve. The logarithmic formula was then applied to the few snapshots we had, so we were able to predict what our 56 day T:P rate would be with only a few weeks of data. Although this was great for the forecast, we would be happy to see the actual data come in as people aged through their 60 day trial. The more actual data we had, the more accurate the prediction became.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EGTwRWOERK3YBd0Y9muBxA.png" /></figure><p>At this point, we had our total visitor data that I mentioned before, as well as a predicted 56 day T:P rate. The last piece of the forecast was adding inputs for future T:P performance. As another option, the forecast had adjustment inputs to be either more conservative or more aggressive. If the conversion team thought T:P might be better than what we’d seen so far, they had the option to increase forecasted T:P by 10%. If they wanted to be more conservative, they could also decrease T:P performance by 10%.</p><p>To recap, the total visitor data (historical + forecasted), the forecasted T:P data, and the T:P adjustment data (optional) was combined together to come up with a forecasted V:P rate, and statistical significance reading. Although using a forecast like this isn’t always going to be 100% accurate, it definitely saved us time, gave our conversion team more comfort in making decisions on tests, and allowed us to test more in a shorter time span than we previously could.</p><p>Although the testing forecast wasn’t specifically built for the new SaaS product initiative I was working on, we did use it multiple times throughout the testing process, and it helped us a lot as we made decisions on the product.</p><h3>Additional Project 2: How to Measure Customer Retention Using Cohorts</h3><p>Analyzing customer cohorts for retention purposes has become very common in SaaS businesses. Many analytics platforms, like Google Analytics, have this type of view included in their offering. However, there was a time when this wasn’t as much as a focus for companies. Here’s the backstory on my experience building this cohort analysis…</p><p>When I first started working at my company, the business was still in “growth” mode. Our goal was to consistently grow our customer base, by bringing in more new customers each month. At a certain point, that growth started to slow down. At the time, one of the leaders of the finance organization recommended that we should focus more on customer retention. We had a good amount of new customers coming in each month, but we had barely any strategy for keeping customer successful and satisfied throughout their tenure with us.</p><p>The company started to put more focus on customer retention shortly after that. One important part of this was getting an idea of what retention looked like historically. We knew that our average customer tenure was 34 months, but that was it. We didn’t know if there was a certain month of tenure, or a certain customer milestone where retention would drastically decline. To build a story of historical retention, I built a retention cohort analysis.</p><p>Each row included a cohort, which is the number of customers that were new to our business each month. Each column represented a different month of tenure for the customers in that specific cohort (first month, 2nd month, etc). If you look at the spreadsheet below, you can see that you can get a sense of how customers are staying over time. Data like this can reveal any glaring declines in retention after a certain tenure, or after a certain point in time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Fs3jMA8sBEQgWLjKR3futA.png" /></figure><p>After putting the analysis together, the next step was to find insights from the data. In our case, retention had been very consistent. We never saw any significant declines after a certain tenure or after a certain cohort month. Our retention rate would decline a few percentage points each month, and it was pretty consistent from cohort to cohort. The only times it dipped were certain cohorts where we offered a month end promotion for X number of months. After X months, we’d see a significantly larger decline than normal.</p><p>Now that we had our historical story, we wanted to use this to track the new cohorts coming in each month. For the most part, we made note of changes the company made, and would follow retention performance after that. For example, if we made a change to our pricing in October 2014, we would note that, and if performance changed, we would have a good hypothesis as to why. Keep in mind that we would always A/B test something like that, but this cohort analysis is a good compliment, and it’s useful for senior leadership who just want the high level company metrics, without going into too much detail. Knowing whether an A/B test is successful is very important, but it doesn’t show the impact on the overall business.</p><p>If your business model is a month-to-month subscription model, this analysis is pretty simple to build. As long as you can track when customers started paying for your business, and have records of monthly invoices/payments, then you can put an analysis like this together. If you’re subscription model requires payments in 6/12/etc month contracts, then this may be harder to build. It’s harder to pinpoint exactly where customers decided to stop paying. My suggestion here would be to figure out what customer engagement metrics are indicators of retention. Whether it’s customer logins, product engagement, etc, you could build a similar analysis with that data, and use that as a proxy.</p><p>Just like the A/B testing forecast, this wasn&#39;t initially built for the SaaS product I worked on. However, we did start looking at retention cohorts as the customer base for the product grew. It helped us figure out what some of our customer retention efforts would look like, including retention testing, engagement programs, etc.</p><p><strong>DISCLAIMER: For confidentiality reasons, the actual metrics values included in this post have been adjusted. However, the story is still an accurate portrayel of the project, and what occured.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4f2e30153da1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A/B Testing Forecast: A Solution for Statistical Significance and Small Sample Sizes]]></title>
            <link>https://medium.com/@mattpupa/a-b-testing-forecast-a-solution-for-statistical-significance-and-small-sample-sizes-abf4d8bc99ea?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/abf4d8bc99ea</guid>
            <category><![CDATA[a-b-testing]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[conversion-optimization]]></category>
            <category><![CDATA[product]]></category>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Tue, 17 Oct 2017 20:24:06 GMT</pubDate>
            <atom:updated>2017-10-18T03:43:11.885Z</atom:updated>
            <content:encoded><![CDATA[<p>Awhile back, the conversion team at my company came to our analytics team about doing testing on our website. We used to have a trial sign up form that was very long, that required 8–10 data points including name, email, industry, etc. This was overkill if you compare it to a lot of digital products today that want to get people into their UI as fast as possible.</p><p>The conversion team wanted to test out a much shorter trailer form, and see what that would do to our conversion rate. We were happy to help them out, but there was one big challenge…they were hoping to do a series of 5–6 tests over the next 3 months.</p><p>That was extremely difficult to do with our business model. Not only did we usually need to let tests run for a few weeks to get enough sample size, we also offered a 60 day trial period. To get the complete test results and have them be statistically significant, it could take us almost 3 months to get a read on 1 test, never mind 5 or 6. In order to solve this problem, I build a testing forecast that would help us make decisions on test results in 3–4 weeks.</p><p>To provide some background on the metrics being measured, the conversion team typically ran tests that affected our website sign ups, which we measured by Visitor to Trial Rate (V:T), as well as tests that affected our customer conversion rates (Trial to Pay Rate (T:P). For this forecast, the primary measurement combined the two into Visitor to Pay Rate (V:P). This would offset differences between the two metrics. For this shorter trial form test, although our hypothesis was that a shorter trial form would lead to a higher V:T rate, we were concerned that T:P rate may decrease because the trialers may not be as invested. V:P let us see the net effect of that, and would help us declare a winner.</p><p>As far as the forecast went, The V:T part was pretty straight forward. We would let the test run for a few weeks and track V:T rates for the test group(s) and the control. I included inputs that would allow people to add X number of additional days the test could run for, with Y total visitors being added each day. The forecast would then take those input values, appropriately split them between the test/control groups, and apply the historical V:T rates that we had seen so far. This forecasted visitor data, would be added to the historical visitor data, and we would have a much bigger sample size.</p><p>The T:P part of the model was a little more in depth. Now that we had our Total visitor and V:T data (Historical + forecasted), the model next needed to trend out what the T:P data would look like. In order to do this, I captured T:P rate snapshots at 7 day increments up to 56 days (Not a perfect 60, but close enough for my purposes). Because most of our conversions came in the first few days of our trial period, our conversion rate was a logarithmic curve. The logarithmic formula was then applied to the few snapshots we had, so we were able to predict what our 56 day T:P rate would be with only a few weeks of data. Although this was great for the forecast, we would be happy to see the actual data come in as people aged through their 60 day trial. The more actual data we had, the more accurate the prediction became.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EGTwRWOERK3YBd0Y9muBxA.png" /></figure><p>At this point, we had our total visitor data that I mentioned before, as well as a predicted 56 day T:P rate. The last piece of the forecast was adding inputs for future T:P performance. As another option, the forecast had adjustment inputs to be either more conservative or more aggressive. If the conversion team thought T:P might be better than what we’d seen so far, they had the option to increase forecasted T:P by 10%. If they wanted to be more conservative, they could also decrease T:P performance by 10%.</p><p>To recap, the total visitor data (historical + forecasted), the forecasted T:P data, and the T:P adjustment data (optional) was combined together to come up with a forecasted V:P rate, and statistical significance reading. Although using a forecast like this isn’t always going to be 100% accurate, it definitely saved us time, gave our conversion team more comfort in making decisions on tests, and allowed us to test more in a shorter time span than we previously could.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=abf4d8bc99ea" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My Resume]]></title>
            <link>https://medium.com/@mattpupa/my-resume-29f802188b4c?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/29f802188b4c</guid>
            <category><![CDATA[analytics]]></category>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Tue, 11 Jul 2017 18:43:59 GMT</pubDate>
            <atom:updated>2017-09-30T02:16:28.557Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>SUMMARY</strong></p><p>Analytics Professional with 6+ years experience in various cross functional settings. I’m looking for an opportunity to expand beyond Analytics, and to help build and execute great ideas and successful business strategies.</p><p><strong>PROFESSIONAL EXPERIENCE</strong></p><p>Constant Contact Inc, Waltham, MA May 2010 — February 2017</p><p><em>Serve as a consultant to internal stakeholders by transforming large amounts of business data into insights that are simple, actionable, and effective</em></p><p><strong>Senior Analyst, Customer Analytics</strong></p><ul><li>Analytics Lead for Constant Contact Toolkit testing &amp; migration, including product changes, trial parameter changes, and additional offerings</li><li>Created forecasts and sized revenue opportunity, as a part time financial analyst for a new business initiative focused on improving digital advertising for small businesses</li><li>Analytics ambassador for Innovation team and Constant Contact Innoloft, supporting innovation design sprints and Innoloft start-up testing</li></ul><p><strong>Analyst, Customer Analytics</strong></p><ul><li>Selected to be an analyst on Constant Contact Toolkit start-up initiative that focused on redefining Constant Contact’s product offering and positioning in the small business market</li><li>Implemented iteration-based A/B testing that enabled quick decision making and supported continuous product improvements</li><li>Performed ad-hoc analysis to better understand levers of customer conversion and retention</li><li>Met with CEO and senior management twice a week to present findings and make recommendations for product/service improvements</li><li>Worked with data warehouse team to make sure reports and models were prepared for full launch to new customers</li></ul><p><strong>ANALYTICS CONSULTING FOR </strong><a href="https://socialseam.org/"><strong>SOCIAL SEAM</strong></a><strong>, April 2017 — Present</strong></p><ul><li>Work directly with lead designer to implement google analytics click event tracking on website elements</li><li>Built several custom google analytics reports to give CEO insight into customer conversion customer activity</li></ul><p><strong>CROSSFIT LEVEL 1 TRAINER, </strong><a href="http://www.mountainstrength.com/"><strong>Mountain Strength Crossfit</strong></a><strong>, January 2017 — Present</strong></p><ul><li>Completed CrossFit Level 1 certification course in January 2017</li><li>Run and organize classes/events, conduct private training for new members, and provide encouragement and feedback to members while helping them achieve their health goals</li></ul><p><strong>EDUCATION</strong></p><p><strong>Bentley University</strong>, Waltham, MA</p><p>Bachelor of Science, Finance May 2010</p><p>Minor in Computer Information Systems</p><p><strong>TECHNICAL SKILLS</strong></p><p>Microsoft office, SQL, R, Google Analytics, D3.js, Processing, TOAD, Aginity, Tableau, Hive, Hadoop</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=29f802188b4c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Investigating the Financial Value of Customer Satisfaction]]></title>
            <link>https://medium.com/@mattpupa/investigating-the-financial-value-of-customer-satisfaction-4479be92ac8b?source=rss-37652710b9a9------2</link>
            <guid isPermaLink="false">https://medium.com/p/4479be92ac8b</guid>
            <category><![CDATA[marketing]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[customer-lifetime-value]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[customer-experience]]></category>
            <dc:creator><![CDATA[Matt Pupa]]></dc:creator>
            <pubDate>Sat, 11 Feb 2017 23:07:39 GMT</pubDate>
            <atom:updated>2017-02-11T23:07:39.122Z</atom:updated>
            <content:encoded><![CDATA[<p>A few years ago, one leader at my company was really into customer satisfaction. We had our own internal measure for customer satisfaction, and for the sake of simplicity, let’s say it’s a scale of 1–5 with 1 being the least satisfied, and 5 being the most satisfied. Although we knew what trends looked like historically, we never know what the financial value of each score was. Also, we didn’t know what the financial impact was of someone moving scores. That’s what this leader wanted to know.</p><p>In order to do this analysis, I first needed to get the data. I worked with the person who sends out these customer satisfaction surveys to get responses from tens of thousands of customers who we surveyed over the years. After a little bit of cleaning, I had each customer ID, the date they were surveyed, and the score they gave us on the 1–5 scale.</p><p>The next step was to profile these customers by score to figure out if there was anything that stood out. I looked at different demographic data points, engagement, financial data, and a few other things. The one thing that stood out right away was the financial data. My hypothesis was that customers who were more satisfied with us would be willing (and would actually) pay us more money on a monthly basis. I expected to see average revenue per customer increase as the customer satisfaction score increased. However, I saw the complete opposite. Average revenue decreased as customer satisfaction increased. In some cases, that is perfectly logical if customers are very price sensitive. However, we’ve learned from previous surveys that that’s not a primary concern for our customer base. The next step was to dig into the financial data further to figure out what was happening.</p><p>At the time of doing this analysis, the money customers paid us depended on two things, 1) how many products they used, and 2) How many of their contacts they were communicating with through our products. I first looked at the average number of products and product mix by satisfaction score. Those were pretty similar, so I ruled out that the issue was related to a specific product. I then looked at average revenue by score at 3 different points in time, 1) the customers’ first month with us, 2) the month right before the customer was surveyed for their satisfaction score, and 3) the most recent month at the time of the analysis. The graph below helped reveal what was happening.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*s68Nogsz5j-eRxMax9hXdQ.jpeg" /></figure><p>When looking at 1st month average revenue by customer satisfaction score, the decline in revenue was much flatter as scores increased. The big decreases in revenue only occured when looking at the month before people were surveyed, and the most recent month before the analysis was done. There was something happening early on in the customer experience that was a problem.</p><p>Like I mentioned before, part of what determines how much revenue we get from our customers is how many of their contacts they communicated to with our products. I made the same graph as above, but I replaced average revenue data with data on the # of contacts. The graphed looked exactly the same. The decline in the 1st month of being a customer wasn’t that steep, but the decline was very steep as customers were surveyed.</p><p>The next step was figuring out why some customers seemed to have a hard time bringing a high volume of contacts into our product. I talked to our marketing and product teams to see if they had anything specific related to this. I thought that maybe we had a poor product flow, or confusing/lack of content on how to add contacts. Those teams didn’t have anything that would explain the story, but I did get pointed to our compliance team. Without going into too much detail, these customers were being flagged due to suspicion of spam, and their accounts were frozen. In many cases, this action was legitimate and the right thing to do. However, there were cases where customers were mistakenly flagged, which left a really bad taste in their mouths. Now that we had a better understanding of how financials were related to customer satisfaction, the next step was to forecast what the impact could be if the compliance team tightened up their flagging process.</p><p>One of the other data points that I looked early on in the analysis was average customer tenure by score. As I expected, customers with higher satisfaction scores stayed with us much longer. For a basic forecast, the math here is pretty simple. Assuming we could get some of the customers giving us low scores (1 and 2) to higher scores (4, 5) by changing the flagging process, the revenue would still be the same. However, the customers might stay with us longer. The financial impact calculation is…</p><p><strong>Average Revenue of low score X (Average months tenure of a high scoring customer — Average months tenure of a low scoring customer)</strong></p><p>If a customer with a low score was paying us $50 per month on average, but will now stay 3 months longer, we would get an incremental $150 for that customer over their lifetime.</p><p>After presenting the forecast data to some of my team, the head of compliance, and the leader who was originally asking about customer satisfaction, we ended up not making the change to the compliance process. It turns out, the volume of customers giving us extremely low scores wasn’t high enough to make an impact. Even though we would have seen a financial benefit, the cost to change the process would have outweighed the benefit. Although it’s unfortunate, analyst recommendations don’t always get carried out. In this case, the reasoning made sense as to why. I would have liked to follow up on this analysis more recently, but priorities for the company have since changed.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4479be92ac8b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>