<?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[Whistle Product &amp; Engineering - Medium]]></title>
        <description><![CDATA[Insights from the people creating the smartest pet products used by pet owners everywhere. - Medium]]></description>
        <link>https://medium.com/whistle-product-engineering-blog?source=rss----e6aa629c6ed2---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Whistle Product &amp;amp; Engineering - Medium</title>
            <link>https://medium.com/whistle-product-engineering-blog?source=rss----e6aa629c6ed2---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 06:41:53 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/whistle-product-engineering-blog" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Hypothesis Driven Design: Be the Francis Bacon of Your Product Team]]></title>
            <link>https://medium.com/whistle-product-engineering-blog/hypothesis-driven-design-be-the-francis-bacon-of-your-product-team-1aef48b8daea?source=rss----e6aa629c6ed2---4</link>
            <guid isPermaLink="false">https://medium.com/p/1aef48b8daea</guid>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[user-research]]></category>
            <category><![CDATA[design]]></category>
            <dc:creator><![CDATA[Malia Eugenio]]></dc:creator>
            <pubDate>Wed, 11 May 2016 19:04:40 GMT</pubDate>
            <atom:updated>2016-05-11T19:02:22.622Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*J42QaswY4-a4OQIiPwOUGA.jpeg" /><figcaption>Photo Credit: Maddie the Coonhound</figcaption></figure><p>by <a href="https://twitter.com/maleug">Malia Eugenio</a></p><p>There is power in knowing what you don’t know. So much inherent value lives within the gaps of knowledge that we all have when we’re designing products — the hard part is being brave enough to confront them and work through their complexities. One of the best ways to flesh out what you know vs. what you don’t know, is to form hypotheses about what you and your product team <em>think you might know.</em></p><blockquote>As designers, we’ve all heard the proverbial rule that we must have reasons behind our design decisions. However, good design doesn’t just start and end with a reason. It starts with a <strong>hypothesis.</strong></blockquote><p>In fact, if you want to design great products, you should have multiple hypotheses for everything you design— and you should seek to debunk all of them.</p><h4>How a Design Hypothesis is Born</h4><p>Thanks to <a href="https://www.khanacademy.org/humanities/monarchy-enlightenment/baroque-art1/beginners-guide-baroque1/a/francis-bacon-and-the-scientific-revolution">Francis Bacon</a> and his scientific posse, we have textbook methods for asking and answering questions by <strong>making observations</strong> and <strong>practicing experimentation</strong>. This exact kind of discovery found in observation and experimentation should be applied to design thinking.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_CG2bTJ9QVaM1xm3T-HEDg.png" /><figcaption>A strong hypothesis makes a strong experiment. In the above chart, the Deductive Approach places the hypothesis early on in the process.</figcaption></figure><p><strong>Deductive Design Thinking</strong></p><p>To borrow from <a href="https://en.wikipedia.org/wiki/Deductive_reasoning">Aristotle’s deductive reasoning</a>, taking a deductive approach is the best way to identify a general theory. This way, you can start by forming a premise or theory, then get more specific as you refine your theory through experimentation later on.</p><p>Deductive reasoning is known as the “<a href="http://www.socialresearchmethods.net/kb/dedind.php">top-down approach</a>” for a reason. You can’t experiment until you’ve identified what you need to test — this is where the hypothesis comes in. As you build a hypothesis, deduced from a general theory, you can more easily define your subject for experimentation.</p><p>A strong hypothesis makes a strong experiment, and is likely to produce highly actionable (positive or negative) results. Conversely, experimentation lacking a hypothesis can put you at risk of spending time and energy on the wrong stuff.</p><p>Deductive design thinking is great for concept testing and feature validation, by its nature of starting with broad premise and distilling it down into a more specific set of potential solutions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CfMwCE_XcInhnpG5nNCvmQ.png" /><figcaption>By carefully identifying what you think you know, you can develop a strong hypothesis to test later on as shown in the Inductive Approach above.</figcaption></figure><p><strong>Inductive Design Thinking</strong></p><p>The inductive approach is opposite. Here, it’s all about starting with very specific observations. Within these observations, you can identify patterns and trends that are the perfect segway into a set of tentative hypotheses.</p><p>Once you’ve opened your observations up to more broad thinking and identified what you <em>think</em> you know, you’ll be able to land on some general theories to test later on.</p><p>Because of inductive reasoning, conclusions are allowed to be false. This is actually somewhat of a silver lining, as identifying where you’re wrong will only bring you closer to being right.</p><p>Inductive design thinking is best for developing product ideas and validating concepts at a higher level. The tentative hypotheses induced from this approach can help you and your team define your research direction. When you confront your theories head-on with testable hypotheses, you can easily figure out what assumptions you have about product or users that are wildly incorrect.</p><h4>The Golden Rule: Ask Why Before Asking How</h4><p>It’s not only important to form hypotheses in the research phase, but you must also <a href="https://medium.com/@mwambach1/hypotheses-driven-ux-design-c75fbf3ce7cc#.1o4kx05ds">let your hypothesis guide you through the design process</a>. A strong hypothesis is the glue that ties a problem to a solution. Whether you’re solving for user stories or a visual style, you must always design with a hypothesis in mind.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HvvhY8911cnfZa9wBnUx7Q.png" /></figure><p>Generating a hypothesis doesn’t stop at identifying an assumption and trying to find evidence to deem it as “true”. A good (false) hypothesis is where the learning happens. You should be trying to prove that hypothesis wrong. By testing things <em>against</em> your hypothesis, you can uncover learnings faster.</p><blockquote>At Whistle, our product team focuses on hypothesis “buckets” to further identify themes within the educated guesses we are making. This allows us to be more agile in our design process.</blockquote><p><strong>Attitude Related Hypothesis: </strong>Understanding intention, need, and desire</p><p><strong>Behavior Related Hypothesis: </strong>Understanding motivation and interaction</p><p><strong>Feature Related Hypothesis: </strong>Understanding use cases and potential solutions (tying needs and desires to <em>potential</em> solutions, before actually committing to solutions)</p><h4>Hypothesis as a Communication Tool</h4><p>Spread the word and get your team on board!</p><p>A design hypothesis will have the most impact on your team’s designs if everyone is aware of it. When you evangelize your hypothesis, as well as the road leading to it, you can ensure that your whole team is keeping it in mind throughout the design process.</p><p>Interested in making cool products for people and their pets? <a href="http://www.whistle.com/careers/">Whistle is hiring!</a> You can also reach out to me at malia@whistle.com</p><p><a href="http://www.maliaeugenio.com/">www.maliaeugenio.com</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1aef48b8daea" width="1" height="1" alt=""><hr><p><a href="https://medium.com/whistle-product-engineering-blog/hypothesis-driven-design-be-the-francis-bacon-of-your-product-team-1aef48b8daea">Hypothesis Driven Design: Be the Francis Bacon of Your Product Team</a> was originally published in <a href="https://medium.com/whistle-product-engineering-blog">Whistle Product &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Building a Magic On-Boarding]]></title>
            <link>https://medium.com/whistle-product-engineering-blog/building-a-magic-on-boarding-410edd24e48c?source=rss----e6aa629c6ed2---4</link>
            <guid isPermaLink="false">https://medium.com/p/410edd24e48c</guid>
            <category><![CDATA[design]]></category>
            <category><![CDATA[product-design]]></category>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[onboarding]]></category>
            <dc:creator><![CDATA[Arman Nobari]]></dc:creator>
            <pubDate>Fri, 06 May 2016 21:23:27 GMT</pubDate>
            <atom:updated>2016-12-01T05:04:43.821Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wjzma_Z1aVo0AJqHh2rH0g.jpeg" /></figure><p>We’ve all been there. You download an app, launch it, and it asks for half a dozen permissions before you even get the chance to create an account. Or perhaps they throw you into an empty-state without explaining how to use the product? These first-impression fumbles are the elements of a blundered on-boarding process. This is what we’re going to avoid. <em>We can do better.</em></p><blockquote>The following assumes the product in question is functioning, is usable, and is reliable. Without a foundation composed of those three elements, the experience will ultimately crumble.</blockquote><h4>The First Impression</h4><p>Delighting your users doesn’t just happen, and it’s so much more than a buzzword. As a goal, users should be delighted the moment they open an application. This doesn’t mean the interface should be <em>Disney on Ice, </em>with every element dancing around;<em> </em>but it should exceed their expectations and offer them a warm welcome to the product.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/270/1*4bLSnEeiNol4-Jo1O9pQrA.jpeg" /><figcaption>Instagram is all about photos, so their first impression had better be about photos. Even the buttons take a step back with their alpha channel and let the photo shine (pun intended).</figcaption></figure><p>A great first impression as a core product element is to remove all unnecessary friction that stands between a new user signing up and their first “WOW!” moment, and having an elegant in the path getting there.</p><blockquote>Adding a progress bar can increase conversions up to 40%.</blockquote><h4>Award early wins</h4><p>There’s a <a href="http://www.jstor.org/stable/10.1086/500480?seq=1#page_scan_tab_contents">variety</a> <a href="http://ijbssnet.com/journals/Vol_6_No_8_1_August_2015/8.pdf">of</a> <a href="http://home.uchicago.edu/ourminsky/Goal-Gradient_Illusionary_Goal_Progress.pdf">studies</a> that suggest the same thing: progress towards a goal increases likelihood and speed of engagement towards completing the goal. By giving users early victories — a perceived “huge” reward for minimal input effort — they’re more likely to stick through harder on-boarding steps.</p><p>Let’s take, for example, the basics of creating a new account with a service: setting a username and password.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/952/1*X7zcZ4IpnkZCHFVYlS5vKQ.png" /></figure><p>By giving an arbitrary “33% completion” for adding these two easy, mandatory items, we can imply the user is already a third of the way there. Having a progress bar gives the user context as to where they are in the new account process, and makes it more likely they’ll complete the remaining steps in the setup. Because of this, it should come as no surprise that adding a progress bar can <a href="http://firstround.com/review/this-is-how-you-design-your-app-for-maximum-growth/">increase conversions up to 40%</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/412/1*HJRmH4OE_eVvLcByaQfLjg.jpeg" /><figcaption>LinkedIn’s “profile completeness” meter shows various titles instead of a percentage. Why settle for “Average” when importing your email contacts list would make you an “Expert”?</figcaption></figure><p>Without a progress bar, users would have no end goal to focus their effort towards. LinkedIn and Twitter are excellent examples of products that use progress bars as a persuasive utility for users during the setup process. Delighting users with visually satisfying progress can make each step of the process feel that much more worthwhile.</p><blockquote>In a nutshell, putting resources into something makes you more likely to see that “something” through to the end.</blockquote><h4>Play to the sunk-cost fallacy</h4><p>The basic idea of a sunk-cost is that investments (time, effort, etc) can not be recovered once incurred. If you‘ve ever eaten food just so you didn’t “throw away that money” despite being full, then you’ve experienced the sunk-cost fallacy in action. In a nutshell, putting resources into something makes you more likely to see that “something” through to the end.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/526/1*Ds9UQjO5XSL6ioZd0yiGDg.png" /><figcaption>Duolingo prompting a new user to create an account (“profile”) <strong>after</strong> compleing a first quick lesson.</figcaption></figure><p>Duolingo does this <em>beautifully</em>. Their on-boarding process (documented by <a href="https://medium.com/u/ae1d1c7cf0b6">Samuel Hulick</a> <a href="https://www.useronboard.com/how-duolingo-onboards-new-users/?slide=68">here</a>) places account creation at the end of a fun tour that feels like a lesson in a language of your choosing. <em>Of course</em> you’re going to create an account — you don’t want that first language lesson to go to waste!</p><h4>Prime your requests</h4><p>No one likes strangers showing up, unannounced, asking for sensitive information — so why do apps think it’s alright to greet potential new users with a boatload of permissions requests before offering an introduction?</p><p>By <em>priming</em> users for a permission request (to send push notifications, or to access to a camera, for example), you can drastically increase the likelihood of getting that <em>ever valuable </em>permission token, and you can enhance user experience all at the same time.</p><p>On an experiential level, <em>“priming”</em> means to explain that a permission request is needed. The app has a user trigger the system-level permission modal by pressing a button themselves, so there’s no surprise in the request — after all, they asked for it. In the example below, a user presses “Notify Me” to trigger a system-level request for permission to send push notifications.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/413/1*E-5fzaov8rpDvPh55PAEXw.png" /><figcaption>Cluster.co asking users for Push Notification permissions</figcaption></figure><p>In redesigning their on-boarding flows, <a href="https://cluster.co/">Cluster</a> saw a <a href="http://techcrunch.com/2014/04/04/the-right-way-to-ask-users-for-ios-permissions/">100% completion rate</a> by having a user bring up the notifications prompt by using a “Notify me” button. There’s only one chance to ask for each of these permission tokens; if a user declines, there’s a chance they’ll need to muck about in their system settings for your product to work properly.</p><h4>Educate, don’t abandon</h4><p>An on-boarding process doesn’t end once they’ve created and account and given you all the permissions you need. Users need to be educated to all the gestures, menus, and essential functions of your product — it’s unlikely they’ll seek this information out on their own.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/270/1*Fp51MptAy05d6wuZ1TSXOA.jpeg" /><figcaption>Google Earth educating a new user on the essential gestures for using the product — with a button to allow existing users to skip the tutorial session.</figcaption></figure><p>New users are just that — they’re new. They don’t necessarily know how to do anything with your product, so teaching them new features with a guiding hand is essential. Giving existing users a way to skip the tutorial is also an easy win which will scale as product adoption grows.</p><h4>In Conclusion</h4><p>Treat users like you’d treat a friend. Guide them through new experiences, make them feel empowered, and make their time and effort invested <strong><em>feel</em></strong> worth the while.</p><p><a href="http://www.armannobari.com/">www.armannobari.com</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=410edd24e48c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/whistle-product-engineering-blog/building-a-magic-on-boarding-410edd24e48c">Building a Magic On-Boarding</a> was originally published in <a href="https://medium.com/whistle-product-engineering-blog">Whistle Product &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When the Happy Path is GPS to a Lost Pet]]></title>
            <link>https://medium.com/whistle-product-engineering-blog/when-the-happy-path-is-gps-to-a-lost-peet-2c42c30e6662?source=rss----e6aa629c6ed2---4</link>
            <guid isPermaLink="false">https://medium.com/p/2c42c30e6662</guid>
            <category><![CDATA[design]]></category>
            <category><![CDATA[product-design]]></category>
            <category><![CDATA[design-thinking]]></category>
            <dc:creator><![CDATA[Arman Nobari]]></dc:creator>
            <pubDate>Wed, 20 Apr 2016 20:52:15 GMT</pubDate>
            <atom:updated>2016-04-20T20:56:55.530Z</atom:updated>
            <content:encoded><![CDATA[<h4>or: “Designing for the worst possible scenario”</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ubyhJXZKsYOyQjELN_cZxA.jpeg" /></figure><p>One of many untraditional things about building product at Whistle is that we hope many of our users never have to use our most prominent feature. In the same way that the makers of smoke detectors hope people never have to use their product, we — as the makers of the Whistle GPS Pet Tracker — hope that people never lose their pets. The unfortunate truth is that over 10 million pets get lost every year. As pet owners ourselves, we’re building, hacking, and designing with those lost pets — and their likely stressed owners — in mind.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*TUaGhvnf2XWdnupbRbbSMA.jpeg" /><figcaption>The current generation Whistle GPS Pet Tracker</figcaption></figure><p>The Whistle GPS Pet Tracker is a small on-collar device that helps pet owners find lost pets with GPS and keep tabs on their best friend’s overall wellness. And it’s a trip to build and iterate a product with so many emotional aspects.</p><p>Our users, in a lost-pet scenario, are likely frantic, worried, and don’t have time to navigate menus or decipher tooltips. A lot of us on the team — myself included — know firsthand what that’s like to be in that awful moment. While it might have been terrifying when it happened, knowing what an experience like that feels like is an important reference point in building a product to solve it.</p><h4>Context</h4><p>Aside from knowing who your target users are, it’s important to know how, when, and why users are interacting with your service. Without knowing those things, there’s no foundation for building an emotional product. Because we know it’s a high-stress, urgent matter to find a lost pet, we can cut a lot of UI out and get straight to the point: <em>“Here is your pet’s location”</em>.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F163601460&amp;url=https%3A%2F%2Fvimeo.com%2F163601460&amp;image=http%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F567014843_295x166.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=vimeo" width="400" height="300" frameborder="0" scrolling="no"><a href="https://medium.com/media/596aa488a25d484e27adf5424cd3e1f7/href">https://medium.com/media/596aa488a25d484e27adf5424cd3e1f7/href</a></iframe><h4>Clarity</h4><p>Naturally, humans try to fit the world into patterns, and a lost-pet scenario is not the time to introduce a new one. We could’ve had some crazy map design, with all sorts of gesture-based inputs and explosive visual transitions, but those things take learning and increase cognitive load. By working on a visceral level, we want our users to feel like our interface resembles things they already understand — maps, directions, minutes — as opposed to new branded UI. The goal of entire experience is to feel familiar, and comforting. This is something we’re constantly refining, day in and day out.</p><h4>Relevancy</h4><p>There’s a <em>ton</em> of tech crammed into our device. I mean, it’s crazy. Just because we have all this data on hand doesn’t mean our users want to see it all at once, or during certain situations. For example, we’ve heard of people using our device to find their pet lost in canyons and in caves. Their pet’s feeding schedule isn’t relevant in situations like that, so we throttle back what we present to them.</p><p>The entire Whistle product experience boils down to <strong>trust</strong>. Some of our users rely entirely on our push notifications to tell if their pet has left home, without using the full app for — and that’s completely okay. If that’s what they want, and they trust us to be accurate, we want to empower them to use the product in that way.</p><h4>Reflection</h4><p>If someone is just checking up on their pet with no cause for alarm, we want them to feel that Whistle has their back — because we do. The device and app can do quite a lot, but there’s a time and a place for everything. By keeping the features focused and separate — with times for action and times for reflection — the visual noise is reduced per-screen, and there’s always clear path for users to take.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*vFWWOl3uJ0MsAHaEAIAexQ.gif" /><figcaption>Space Doge, our Chief Petfinding Officer</figcaption></figure><h4>Future</h4><p>While the device and app can do all of these things currently, <em>we can do it so much better</em>. We can innovate on the GPS front, getting more accurate location readings faster. Information can always be more streamlined and should preempt the questions or needs of our users. Solving for these challenges is precisely what keeps us up at night. Product design and development is an ongoing iterative process, and with our users’ pets in mind, we’re sprinting forward building the next generation of connected pet devices.</p><p>Want to be a part of the team giving owners pet-finding superpowers? We’re hiring! Check out <a href="http://www.whistle.com/careers">whistle.com/careers</a> or drop me an email at arman@whistle.com</p><p><a href="http://www.armannobari.com/">www.armannobari.com</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2c42c30e6662" width="1" height="1" alt=""><hr><p><a href="https://medium.com/whistle-product-engineering-blog/when-the-happy-path-is-gps-to-a-lost-peet-2c42c30e6662">When the Happy Path is GPS to a Lost Pet</a> was originally published in <a href="https://medium.com/whistle-product-engineering-blog">Whistle Product &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Pets vs. Cattle Analogy at Whistle]]></title>
            <link>https://medium.com/whistle-product-engineering-blog/the-pets-vs-cattle-analogy-at-whistle-70f8c4b5f4ba?source=rss----e6aa629c6ed2---4</link>
            <guid isPermaLink="false">https://medium.com/p/70f8c4b5f4ba</guid>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[devops]]></category>
            <dc:creator><![CDATA[Garret Pick]]></dc:creator>
            <pubDate>Sat, 19 Mar 2016 00:37:19 GMT</pubDate>
            <atom:updated>2016-03-19T00:37:19.656Z</atom:updated>
            <content:encoded><![CDATA[<blockquote><em>My friend says we’re like the dinosaurs</em></blockquote><blockquote><em>Only we are doing ourselves in</em><br><em>Much faster than they ever did</em></blockquote><blockquote><em>We’ll make great pets, we’ll make great pets</em></blockquote><blockquote>— Perry Farrell 1993</blockquote><p>Here at Whistle, we love our pets and they are the reason we do what we do. Bringing pets to the office is allowed and even encouraged. In fact, it’s not uncommon to have a few dogs under and on the conference table during meetings.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5zUjzl7VZK1dQcqO7qM_zw.jpeg" /><figcaption>Whistle Software Team Scrum — the author is in the lower right corner</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/746/1*5K5rMq1VY_54Bw7B-2hDsg.jpeg" /><figcaption>Vitas taking care of business and sporting his Whistle device</figcaption></figure><p>In the <a href="https://en.wikipedia.org/wiki/DevOps">DevOps</a> world, there is a fairly well-known mantra that you should treat your servers as cattle and not pets. In a nutshell, the idea is that your infrastructure should be as disposable and easy to replace as possible.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/922/1*JVnDLMs8GDHwcAV-2Dnf0A.jpeg" /></figure><p><em>*As someone who was born and raised in the Midwest, it is important for me to note that farmers certainly care for their cattle and the analogy should not be taken to extremes. If a cow dies, it is still a significant event for a farmer and if a server dies, an engineer may still wake themselves up in the middle of the night with an automated notification.</em></p><h3>Immutable Infrastructure</h3><p>The analogy is helpful for introducing the concept of immutable infrastructure, which is a philosophy that the DevOps team at Whistle has come to embrace. Previously, the team worked with a variety of environments including managed hosting on physical and virtual hosts, and eventually completely virtual hosting in the cloud. Making the transition to immutable infrastructure was not easy, but well worth it in the long run. It was also a gradual process, some of which I’ll describe here.</p><h3>Infrastructure Evolution</h3><p>Prior to joining Whistle, I worked with a backend comprised of dozens of servers. Some were deployed as virtual machines, while others ran directly on physical hardware. Deployments (upgrades, introducing new features, etc) were always a scary, painful event. Often, they involved dozens of manual steps including modifying configurations directly on production machines. Development teams threw their releases over the wall to the operations team and had little collaboration after that. Deployments would work in test and staging environments, but fail in production. In short, the DevOps culture did not exist. I infamously once heard a manager quip that “we don’t do agile”.</p><p>At Whistle, the operations team had adopted some of the same practices used by the development team. Configurations were placed under source control (<a href="https://git-scm.com/">git</a>) and configuration management tools like <a href="https://en.wikipedia.org/wiki/Puppet_(software)">puppet</a> and <a href="https://en.wikipedia.org/wiki/Chef_(software)">chef</a> were introduced. Deployments became more automated and the infrastructure was becoming more codified. Developers became more involved with the operation of the system. They were online during deployments and were also added to the pager rotation. The DevOps culture was being adopted.</p><p>This was a huge improvement, however the infrastructure was still fairly mutable. Servers were never completely rebuilt. Additionally, critical non-server infrastructure such as load balancers were still being managed in the traditional fashion. On the positive side, continuous integration was improved and changes were rolled out in a semi-automated fashion. Upgrading applications via a package (<a href="https://en.wikipedia.org/wiki/RPM_Package_Manager">RPM</a>) managed by puppet became a trivial task and <a href="https://en.wikipedia.org/wiki/Continuous_delivery">continuous delivery</a> became a reality.</p><p>Once the system was at AWS, we began to experiment with immutable infrastructure as we could now build the entire stack from the operating system to the application. RPM’s were replaced by <a href="https://en.wikipedia.org/wiki/Amazon_Machine_Image">AMI</a>’s as the new common build artifact for deployment and eventually <a href="https://en.wikipedia.org/wiki/Docker_(software)">containers</a> were introduced as we marveled at how truly lightweight they were. All infrastructure, including the troublesome load balancers, was codified and deployed using the wonderful open source tools from <a href="https://en.wikipedia.org/wiki/HashiCorp">HashiCorp</a> such as <a href="https://www.terraform.io/">terraform</a>.</p><p>Our clusters became self-healing and even self-scaling. Deployments occurred on a daily basis with little fanfare with the developer of the actual code often deploying it.</p><p>Gradually, our servers had morphed from pets into cattle.</p><h3>The 5 S’s</h3><p>Our goals for DevOps at Whistle can be summarized in the following order</p><ol><li><strong>Stability</strong> Without stability, our products can not be trusted by our customers to reliably tell them about their beloved pets.</li><li><strong>Scalability </strong>Directly tied to stability, we need the ability to grow quickly and easily.</li><li><strong>Security </strong>Our customer’s data and system reliability can not be compromised.</li><li><strong>Simplicity</strong> Automate as much as possible with the goal of increasing the “<a href="https://en.wikipedia.org/wiki/Bus_factor">bus factor</a>”.</li><li><strong>Saving</strong> Finally, we need to continually scrutinize our use of resources without jeopardizing any of the previous goals. Idle CPU, memory and disk is a waste of money.</li></ol><p>Our transition from pets to cattle has positively impacted each of these goals. However, the goals are never completely attained. The good news is that the DevOps culture is flexible enough that it allows us to make continual progress.</p><h3>Lessons Learned</h3><ul><li><strong>Be patient.</strong> Any migration of a complex system is challenging and paradigm shifts do not occur overnight. The benefits of DevOps are not always simple to comprehend and mistakes made along the way, may make lasting bad impressions. Also, set the expectation that DevOps is a journey and is never “done” Milestones can be set and achieved, but the software world evolves at warp speeds and there are always new technologies to explore and adopt.</li><li><strong>Culture is Critical</strong>. DevOps is truly a culture that must be adopted by all stakeholders in an organization. Sometimes the most difficult challenges will be non-technical in nature as change is difficult for most people. Without cooperation, you are doomed to fail. For example, without true <a href="http://12factor.net/">12 factor apps</a>, immutable infrastructure will be difficult, if not impossible to achieve.</li><li><strong>Don’t get too hung up on tooling.</strong> In the devops world, there is no shortage of competing tools. Take some time to evaluate and choose the best tool for the job, but prepare to remain flexible when something better comes along. Avoid using different tools for the same basic tasks.</li><li><strong>Take advantage of </strong><a href="https://en.wikipedia.org/wiki/Software_as_a_service"><strong>SaaS</strong></a><strong> whenever possible.</strong> Focus on your core product/technology, because chances are good that everything else can be done better and more efficiently by someone else. Surely, you’re not managing your email servers anymore, do you really need to manage your <a href="https://en.wikipedia.org/wiki/Big_data">Big Data</a> database servers?</li></ul><h3>Pets Have Priority</h3><p>Pets come first at Whistle. Fortunately, the DevOps culture has been adopted across the company and our awesome applications have been designed and developed with that in mind. Without solid infrastructure, we lose our focus on making our applications even better. Internally, our applications are named after actual employee pets at Whistle. Some examples are <em>arlo</em>, <em>manolo</em> and <em>hank</em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b8fRzDprbPI6jG9u8dpLPw.png" /><figcaption>Arlo</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/736/1*Mc1Jo0lS6bjizQVhwb7nyg.png" /><figcaption>Manolo</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JE-uQhPGmpXbzmoBgme-Tw.jpeg" /><figcaption>Hank</figcaption></figure><p>As far as server naming, <em>ip-172–16–50–232</em> suits us fine and you can make her out in the herd below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/683/1*JUrL6gJYvs5ajPAzIDF_og.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=70f8c4b5f4ba" width="1" height="1" alt=""><hr><p><a href="https://medium.com/whistle-product-engineering-blog/the-pets-vs-cattle-analogy-at-whistle-70f8c4b5f4ba">The Pets vs. Cattle Analogy at Whistle</a> was originally published in <a href="https://medium.com/whistle-product-engineering-blog">Whistle Product &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What Product Means at Whistle]]></title>
            <link>https://medium.com/whistle-product-engineering-blog/what-product-means-at-whistle-e0a2db8c0e31?source=rss----e6aa629c6ed2---4</link>
            <guid isPermaLink="false">https://medium.com/p/e0a2db8c0e31</guid>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Daniel Harman]]></dc:creator>
            <pubDate>Thu, 03 Mar 2016 17:35:45 GMT</pubDate>
            <atom:updated>2016-03-03T17:39:44.488Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HGfTBGRg-umaF4hN09Y8aA.png" /></figure><p>Here at Whistle, everyone lives and breathes our products, far beyond than just the Whistle Product team. It is important to us that everyone who joins the Whistle family understands how they contribute to the Whistle “product”, regardless of role or responsibility. One way is by clearly articulating what “product” actually means at Whistle via a Product Manifesto; an attempt to define what matters to us as Whistlers and why. Would be interested to hear your thoughts!</p><blockquote>This is <strong>Product</strong>. We <strong>care</strong> about our customers, both<strong> people and their pets</strong>. <strong>Never,</strong> are they simply users, but rather they are the Whistle family. Product <strong>encompasses everything</strong> that we do here at Whistle. <strong>Every experience</strong> that touches our customers, <strong>is the Whistle product</strong>. We strive to uncover better ways of <strong>building products</strong> <strong>that truly matter</strong> to our family. When we don’t know the answer, <strong>we own it</strong>. We <strong>experiment</strong>, <strong>analyze</strong>, <strong>foster feedback</strong> and <strong>iterate</strong> to find answers that drive development of great products. Products that are truly valuable, that help guide pet owners <strong>through their pet parent journey.</strong> Products that are focused on raising <strong>happy, healthy and safe pets, in a fun and playful way.</strong> Importantly, product is <strong>never simply just pixels and bytes</strong>, but rather <strong>whom we all are here at Whistle.</strong></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e0a2db8c0e31" width="1" height="1" alt=""><hr><p><a href="https://medium.com/whistle-product-engineering-blog/what-product-means-at-whistle-e0a2db8c0e31">What Product Means at Whistle</a> was originally published in <a href="https://medium.com/whistle-product-engineering-blog">Whistle Product &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Fancy Fields]]></title>
            <link>https://medium.com/whistle-product-engineering-blog/fancy-fields-d7cfc2f68e6d?source=rss----e6aa629c6ed2---4</link>
            <guid isPermaLink="false">https://medium.com/p/d7cfc2f68e6d</guid>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[front-end-development]]></category>
            <category><![CDATA[react]]></category>
            <dc:creator><![CDATA[Emmy Armintrout]]></dc:creator>
            <pubDate>Thu, 03 Mar 2016 00:05:59 GMT</pubDate>
            <atom:updated>2016-03-04T05:01:43.597Z</atom:updated>
            <content:encoded><![CDATA[<p>Do you use React? Are you building a form? Are you fancy?</p><p>If so, then you are really good at Googling, because this article is for you.</p><p>Fancy Field has a smooth float label and multiple message customization options. It relies on CSS animation whenever possible and has few prop requirements. It will impress your parents.</p><p>Fancy Field is the product of months of Development and Testing. It has been upgraded to handle multiple parameter types. For example, when you’re filling out a form about your dog, you probably expect a real-time recommendation for how much daily exercise your dog should get. Right!? Just pass Fancy Field a focusText prop and boom. FANCY.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/295/1*80j-erYAIIXJpZ98pPTv_A.gif" /></figure><p>If you’re a person of questionable character and have a Not Dog, you shouldn’t select a dog breed. Disable that right away!*</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/358/1*xC0eK7_Oq7EsyUnbgYS-ww.png" /></figure><p>Being Fancy means being frank. Let the people know why they’re wrong. Position that error message where you WANT. It’s your party.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/520/1*TbtUQ6SruAxFxofNhcsWVw.gif" /></figure><p>Fancy Field does not stand alone. There’s also Fancy Button, which can actually become Fancy Checkbox, Fancy Toggle, or Fancy Switch with the addition of one prop. Fancy Button inspires the forms of your dreams.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/281/1*25kR3XV8DWnyHkGFUiQfXA.png" /><figcaption>A poll about business.</figcaption></figure><p>All the stuff you need, plus More Fancy Options, is available at the <a href="https://github.com/emmya/fancy-fields"><strong>Fancy Fields GitHub</strong></a><strong>. </strong>Easily tweak styles in fancy-variables.scss. We all know you want to pick your own $fancy-color and $error-color.</p><figure><a href="https://www.instagram.com/cassiethadoge/"><img alt="" src="https://cdn-images-1.medium.com/max/640/1*BBR2d9ysi0uaplhWXcWCcQ.jpeg" /></a></figure><p>Fields would have never been this Fancy without the help of designer <a href="https://medium.com/u/5d3d1285a52c">Arman Nobari</a>, my coding guru Brian, and the general wonderfulness of <a href="https://medium.com/u/3eb9753f23a4">whistlelabs</a>.</p><p><em>*Not actually that fancy, I just wanted to draw a frowny face.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d7cfc2f68e6d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/whistle-product-engineering-blog/fancy-fields-d7cfc2f68e6d">Fancy Fields</a> was originally published in <a href="https://medium.com/whistle-product-engineering-blog">Whistle Product &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Whistle Research Strategy: Hang Out With Your Users]]></title>
            <link>https://medium.com/whistle-product-engineering-blog/how-whistle-uses-product-mixers-in-our-research-strategy-141158a806d9?source=rss----e6aa629c6ed2---4</link>
            <guid isPermaLink="false">https://medium.com/p/141158a806d9</guid>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[user-research]]></category>
            <category><![CDATA[design]]></category>
            <dc:creator><![CDATA[Malia Eugenio]]></dc:creator>
            <pubDate>Wed, 17 Feb 2016 00:55:45 GMT</pubDate>
            <atom:updated>2019-01-17T16:42:58.656Z</atom:updated>
            <content:encoded><![CDATA[<p>by <a href="https://twitter.com/maleug">Malia Eugenio</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GhofpYIoPk8GqtQ3_5uXgw.jpeg" /></figure><p><strong>At Whistle, the quality of connection between humans and their pets is our modus operandi.</strong></p><p>To us, user-centered design is more than a buzzword, it is one of our values. As a company, Whistle takes users and their pet relationships very seriously. After all, we are focused on building products that help enhance this relationship, so that pets and pet owners alike can live healthier lives — together. That’s why we keep user needs, attitudes, and behaviors in the vanguard of everything we build.</p><p>As a product team, we don’t see the people who buy our products as simply users, yet a central part of the Whistle family. So, we decided to invite a group of them to come hang out with us.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*A6fwp7765fWTahlF0-a5hw.jpeg" /></figure><h3>Sprucing Up User Research</h3><p>User research is not simply about asking users what they want. In fact, that’s exactly what it is <em>not</em> about. To us, It’s about building a symbiotic relationship between the people making Whistle, and the passionate users who use our product to keep their pets safe and healthy each day. We find so much value in engaging our users regularly and doing so informs our design process sufficiently, while helping us set clear objectives for our product roadmap.</p><p>We can sit here and talk about “archetypal users” all day, revamping personas based on quantitative data and demographic information. We can send out survey after survey to gauge interest in potential features, but that can only get us so far. We need something more contextual. We don’t just want to hear customer feedback, we need to hear customer <em>ideas</em>. We need our design hypotheses to be challenged and debunked.</p><p><strong>We need something a little more real.</strong></p><p>By having product mixers at the forefront of our research strategy, we invite users to come have dinner and share <em>their</em> stories about <em>their</em> pets and experiences using Whistle. More importantly, we use the mixers as a means of amplifying the voices of the people on the other side of the product.</p><p>On one hand, it gives users an opportunity to tell us about their experience in a direct and candid way. On the other hand, it helps us test put our assumptions to the test early on. This allows us to refine our orginal hypotheses about users and their behavior with our product, and figure out how to prioritize features and concepts respectively. This keeps us on our toes. If we seek to be proven wrong by users, we will be led to the <em>right</em> answer, <em>faster</em>.</p><p>We’ve also found that the product mixer is a great way to involve Whistlers outside of the product and design team. Engineering, marketing, finance, and the executive team are all invited to participate as facilitators. These team facilitators learn how to ask effective, open-ended questions during the user testing and interview portion of the mixer. Allowing other roles outside of the product team to gain exposure to users makes for a more unifying product development process in which we are all keeping the end user in mind. It seems that each time, everyone learns something relevant to what they do here at Whistle, simply by spending time with a Whistle user.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PtucwnhIds5FfDsr8tHHlQ.jpeg" /></figure><h3>The Simple Act of Observation</h3><p><em>Apart from getting to know our customers face-to-face, there are 2 main components to the product mixer strategy…</em></p><ol><li>Hear directly from users about their current experience using Whistle</li><li>Expose early concepts to see if we are on the right track</li></ol><p>In a previous mixer, we used a card-sorting exercise to sort out a new navigation schema. We then presented them with two paper prototypes, and ended up implementing both of the designs after gaining further validation. Exposing content in such a raw way allowed us to take user interviews to a new level. From observing their reactions to an early version of a feature redesign, we gained the insights that we needed in order to hone in on the concept further and thus, worked the newly validated solutions into the Whistle experience.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nbad7Dws3S-Sh4eRx2LyfQ.jpeg" /></figure><h3>A Beacon of User Insights</h3><p>Recently, we took a more collaborative approach and made the structure of the mixer less interview-focused. As a tactic of observatory design research, we asked them to whip out their phones, open the Whistle app, and tell us everything they love and hate about certain features. This “live audit” methodology helped us benchmark how users currently interact with certain features, and allowed us to identify patterns within the context of them actually using the app.</p><p>Adding a little flavor to this approach, we tossed in a participatory design exercise and asked users what a more optimal pet activity timeline would look like to them. Of course, attempting to sketch a UI can be scary if you’ve never done it before. However, using this exercise as a primer to encourage ideation with the users allowed us to at least understand what feelings, attitudes, and emotions are behind their interactions with the product. Digging deeper into this behavioral layer exposes so much delicate information that allows us to increase the pathos of our product design process, which thusly increases the emotional appeal of the app itself.</p><h3>Research Analysis: Mixing it All Up</h3><p>A user experience is only as good as the research analysis behind it — both quantitatively and qualitatively speaking. By using the process of deduction to understand the research results of our product mixers, analysis helps us reach an understanding of what users need as we break down complex ideas into smaller, fragmented concepts.</p><p>While the whole is indeed more than the sum of its parts, the summation of user insights as a driving force behind design decisions can drastically improve the whole of an experience. We believe that by engaging with users directly through interviews and product mixers, we can distill our learnings as a way to increase our awareness of who our users are, and continue to move forward in building products that help people take better care of their pets.</p><h4>Stay tuned for more here on how we do user research and analysis at Whistle!</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=141158a806d9" width="1" height="1" alt=""><hr><p><a href="https://medium.com/whistle-product-engineering-blog/how-whistle-uses-product-mixers-in-our-research-strategy-141158a806d9">Whistle Research Strategy: Hang Out With Your Users</a> was originally published in <a href="https://medium.com/whistle-product-engineering-blog">Whistle Product &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[OUR FIRST POST: Introducing the Whistle Product and Engineering blog]]></title>
            <link>https://medium.com/whistle-product-engineering-blog/our-first-post-introducing-the-whistle-product-and-engineering-blog-a203c1271110?source=rss----e6aa629c6ed2---4</link>
            <guid isPermaLink="false">https://medium.com/p/a203c1271110</guid>
            <category><![CDATA[design]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[product-design]]></category>
            <dc:creator><![CDATA[whistlelabs]]></dc:creator>
            <pubDate>Thu, 11 Feb 2016 20:39:04 GMT</pubDate>
            <atom:updated>2016-02-17T01:03:33.997Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>OUR FIRST POST: Introducing the Whistle Product and Engineering blog</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*E3ZZkpcXGVqydv2nyR5CcQ.jpeg" /></figure><p>Welcome to the Whistle product and engineering blog! This is the best place to hear from the engineers, designers, and makers behind the smartest pet products. We will be sharing insight to some of our challenges, our philosophies, and lessons learned. These posts are written by the team members who craft, tinker, refine, and ultimately create the amazing Whistle products pet parents rely on to help them care for their furry family members. We hope this blog will facilitate bigger conversations about design, engineering and what it’s like to work here at Whistle.</p><p>Give us a follow and drop us a line — we’re really excited to share what we’re working on and hear from you!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a203c1271110" width="1" height="1" alt=""><hr><p><a href="https://medium.com/whistle-product-engineering-blog/our-first-post-introducing-the-whistle-product-and-engineering-blog-a203c1271110">OUR FIRST POST: Introducing the Whistle Product and Engineering blog</a> was originally published in <a href="https://medium.com/whistle-product-engineering-blog">Whistle Product &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>