<?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[Love, Bonito Growth Journal - Medium]]></title>
        <description><![CDATA[This is our way of cracking growth — we’d love to share and exchange some ideas with you - Medium]]></description>
        <link>https://medium.com/lovebonito-growth-journal?source=rss----4abe546138af---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Love, Bonito Growth Journal - Medium</title>
            <link>https://medium.com/lovebonito-growth-journal?source=rss----4abe546138af---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 13:16:54 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/lovebonito-growth-journal" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Using Automation to Improve Customer Lifetime Value]]></title>
            <link>https://medium.com/lovebonito-growth-journal/using-automation-to-improve-customer-lifetime-value-c971dc114d34?source=rss----4abe546138af---4</link>
            <guid isPermaLink="false">https://medium.com/p/c971dc114d34</guid>
            <category><![CDATA[growth]]></category>
            <category><![CDATA[ecommerce]]></category>
            <category><![CDATA[crm]]></category>
            <dc:creator><![CDATA[Julie Martin]]></dc:creator>
            <pubDate>Tue, 03 Nov 2020 11:38:51 GMT</pubDate>
            <atom:updated>2020-11-03T12:15:06.076Z</atom:updated>
            <content:encoded><![CDATA[<p>Over the last 12 months at Love, Bonito, we automated customer engagement at various points in their journey with the goal of reducing friction across the user experience.</p><p>We rolled out these automations across our website, emails and paid ads.</p><p>Here’s an overview of our methodology:</p><p><strong>1. Identify and map the key points in the customer’s journey</strong></p><p>Before mapping out the flows, we spent some time identifying the key points in the customer journey using the AARRR framework:</p><ul><li>Acquisition: Lead generation</li><li>Activation: Lead conversion into first-time shopper</li><li>Retention: First-time shopper into a second-time shopper (“aha” moment)</li><li>Revenue: How to transform second-time shopper into repeat revenue-generating customers</li><li>Referrals: How to integrate a customer referral program</li></ul><p>We identified that Activation and Retention are the 2 most important points of the customer’s journey for an omnichannel commerce brand like Love, Bonito.</p><p><strong><em>Tip #1 — Create a visualisation of the customer lifecycle</em></strong></p><p><em>In order to help visualise the existing customer flows, we created a general customer timeline and lined up the different flows below. This helped us to see at which point in the customer journey we could create additional engagement.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6nuKbYZISFE8q0VhZ2Fmqw.png" /><figcaption><strong>Example of customer lifecycle</strong></figcaption></figure><p><strong>2. Build and automate flows with 1–2 scenarios</strong></p><p>We built the most important flows and split the traffic into 2 groups and 1 control to test different variables, such as:</p><ul><li>Timings of the email: e.g. 30 days vs. 60 days after first purchase</li><li>Content: e.g. brand-oriented vs. more personalised content based on customer data</li></ul><p>It is important to add a group of users that will not be affected by the experiment, aka the “control group”, so that we can compare the differences in conversion, ceteris paribus (with all other variables being the same).</p><p>Next, we determine the size of your test group. In order for an experiment to be successful, results need to show a statistical significance less or equal to 5%.</p><p>If the journey will affect a large number of users and we expect the impact to be important, then the test group could be a smaller percentage. If we expect to reach only a few hundred customers every week, then the test group will need to be larger (usually 50%).</p><p><strong><em>Tip #2 — Avoid creating many variations for the same experiment, which will result in too small test groups, where results will be statistically insignificant!</em></strong></p><p><em>The first time we did an automated flow, we sent out the 2nd email to 5 different customer groups each with an allocated time interval. The intention was to find out the most effective time interval to use in the long run. However, after a few weeks we realised that the number of customers in each group was too small to have statistically significant results. In general, we need a few thousand customers in each group to start having significant results. We reduced the number of variations to 2 in the next experiment.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/514/1*QIjVCOSBi4WMtK_sNqpTeQ.png" /><figcaption>Experiment with 2 variations</figcaption></figure><p><strong>3. Run the experiment</strong></p><p><em>How long?</em></p><p>Most of our flows run from 7 to 60 days after a significant number of users performed a target action. A subscription model might consider a customer as churning after 2 months without ordering or using the service, whereas a traditional retail model might use a window of 6 –12 months.</p><p><em>What data are we looking for?</em></p><p>Before starting the flow, we need to know what are the key metrics that will determine if the experiment is successful. It is recommended to have 1 primary metric, and 1 to 2 secondary metrics. For example, if our journey is about the second purchase, the primary metric could be the % of customers that make the second purchase after X days and the secondary one would be the average time to second purchase.</p><p><strong>4. Analyse results. Repeat.</strong></p><p>Once the experiment is over, we compare the key metrics from the control group and the test group. Depending on the results, we will decide to keep the flow running or change the content of the experiment to improve results in future iterations. At times, we use online significance calculators such as <a href="https://neilpatel.com/ab-testing-calculator/">this one</a> to determine if the results are significant.</p><p>Whether the results were successful or not, we always share the learnings internally in our Growth newsletter. See our <a href="https://medium.com/lovebonito-growth-journal/how-do-we-spend-3-hours-a-week-to-run-a-b-tests-at-zero-cost-e3c4b25c9960">first article</a> for more details on this newsletter!</p><p><strong><em>Tip #3 — Try again with different engagement channels</em></strong></p><p><em>When analysing the results of a specific experiment, we consider using different channels for next iterations. For example, if the channel was email only, maybe adding an onsite pop-up with the same personalised message would help to convert users that did not open the emails. Onsite channels worked better for us when the active user base was significant enough to get impressions.</em></p><p><strong>In conclusion — What’s next for us?</strong></p><p>After 1 year of running experiments, we now have an increased proportion of onsite traffic coming from automated emails (+300% between the beginning and the end of H1 2020!). Thanks to the launch of our recent loyalty program LBCommunity+, we are now able to connect the offline customer transaction data with online data for even more accurate targeting. Next steps for us will be to use predictive data models to identify which customer segments are more likely to engage with our content, as well as flag segments with high churn risk.</p><p>Kudos to the team who helped ideate and create the flows:</p><p><a href="https://medium.com/u/29e336693121">Yan Pui</a> | <a href="https://medium.com/u/1fd37392b3e8">Oliviakoh</a> | <a href="https://medium.com/u/91b349bd4e54">Lovin See</a> |and our amazing Brand team!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c971dc114d34" width="1" height="1" alt=""><hr><p><a href="https://medium.com/lovebonito-growth-journal/using-automation-to-improve-customer-lifetime-value-c971dc114d34">Using Automation to Improve Customer Lifetime Value</a> was originally published in <a href="https://medium.com/lovebonito-growth-journal">Love, Bonito Growth Journal</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[3 Quick Hacks to Double Your Facebook Advertising ROIs]]></title>
            <link>https://medium.com/lovebonito-growth-journal/3-quick-hacks-to-double-your-facebook-advertising-rois-3ffd08daca05?source=rss----4abe546138af---4</link>
            <guid isPermaLink="false">https://medium.com/p/3ffd08daca05</guid>
            <category><![CDATA[facebook]]></category>
            <category><![CDATA[campaign-optimisation]]></category>
            <category><![CDATA[marketing]]></category>
            <category><![CDATA[social-media-marketing]]></category>
            <category><![CDATA[social-media]]></category>
            <dc:creator><![CDATA[Product @ Love, Bonito]]></dc:creator>
            <pubDate>Thu, 25 Jun 2020 02:09:28 GMT</pubDate>
            <atom:updated>2020-06-25T02:12:39.662Z</atom:updated>
            <content:encoded><![CDATA[<h4>by <a href="https://medium.com/u/1fd37392b3e8">Oliviakoh</a> and <a href="https://medium.com/u/91b349bd4e54">Lovin See</a></h4><p>As a small in-house team, optimising campaigns can be overwhelming.Where should we start? How should we structure our optimisation strategies?</p><h4><strong>Introduction to Optimisation</strong></h4><p>The key to tackling this is to always look back at your numbers and figure out your pain points. Here are three key questions we asked ourselves.</p><ul><li>Has our audience grown over time?</li><li>Are our creatives delivering the results we want?</li><li>How do we tackle budget constraints</li></ul><p>We are here to share 3 quick yet effective strategies that have helped us boost our campaign performance.</p><h4><strong>Tip #1 — Set broader objective for smaller target audience to improve ROAS</strong></h4><p>We realised our retargeting audience reach was not growing and as a result CPMs were increasing overtime.</p><p><strong>Our hypothesis: </strong>Would laxing our Facebook campaign objective help to drive higher reach, while improving efficiency?<em> If your audience is already of high quality and ready to convert, all you need Facebook to do is to serve them the ad to keep your brand top of mind.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Idh8SdiPgsu8Tut7kYPgYA.png" /></figure><p><strong>Test setup</strong></p><p>Campaign objective: Reach optimisation vs. Value (ROI) optimisation<strong><br></strong>Measurement: Conversion Rate (CR) — Clicks to Purchase<br>Target audience: Cart abandoners<br>Creative: Test best performing campaigns<br>Timeframe: 2 weeks</p><p><strong>Results</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*SMttqSNCuPZVHC8Q" /><figcaption>Our results were statistically significant across two different markets, we see an uplift in both CR and ROAS with a laxed campaign objective</figcaption></figure><p><strong>Takeaway</strong></p><ul><li>Don’t limit your campaigns with your regular purchase funnel objectives. Instead learn how to marry your audience and objectives to maximize performance.</li></ul><h4><strong>Tip #2 — Always test your creative elements to achieve higher CTR</strong></h4><p>Keeping creatives fresh helps to maintain our ad engagement metrics like Click-Through-Rate (CTR) and View-Through-Rate (VTR). There are multiple variables you can test with such as:</p><blockquote>- Creative Copy</blockquote><blockquote>- USP call out (E.g. Free Delivery, Bestsellers)</blockquote><blockquote>- Hero Image</blockquote><blockquote>- Number of images in a creative</blockquote><blockquote>- Animation &amp; Colour Effects</blockquote><p>When it comes to evaluating creative performance, we want to focus on CTR <em>(click-through-rate)</em> as our key metric. <em>With a higher CTR, we will be able to drive higher traffic which, in turn, would potentially lead to an increase in conversions.</em></p><p>While our campaigns have consistently been successful at driving conversions, we were not certain which elements would help to drive incremental CTR and engagement. Hence, the variable we tested here — our creative copy.</p><p><strong>Our hypothesis:</strong> A more personalised and informal messaging on creatives will drive higher engagement.</p><p><strong>Test setup<br></strong>Measurement: CTR<br>Target audience: Interest &amp; Lookalike<br>Creative: Chinese New Year Launch (Control Creative vs Creative A vs Creative B)<br>Timeframe: 2 weeks</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/906/1*jmV3VrzdZLDhX7xsueznaQ.png" /></figure><p><strong>Results</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/906/1*UD5GLE9jPmmOmhGvagAeDA.png" /><figcaption>Our results here were statistically significant (95% confidence interval) and proved our hypothesis across the three different markets.</figcaption></figure><p><strong>Takeaway</strong></p><ul><li><strong>Don’t stop after one test! </strong>Keep testing other elements to investigate what would drive a more engaging creative.</li></ul><h4><strong>Tip #3 — Group new markets when you have limited budget for better performance</strong></h4><p>When it comes to market expansion, your initial marketing budgets might be limited as these are unexplored territories.</p><p><strong>So how should campaigns be structured in order to maximise spends and not let small budgets limit your campaign?</strong></p><p>Here are 2 possible ways of setting up your campaigns</p><blockquote><strong>Campaign Structure 1:</strong></blockquote><blockquote>- All markets</blockquote><blockquote><strong>Campaign Structure 2:</strong></blockquote><blockquote>- Campaign 1: High CPM markets</blockquote><blockquote>- Campaign 2: Low CPM markets</blockquote><p>In campaign structure 1, we have all markets under the same campaign but split by ad sets. <em>Here we are allowing Facebook to optimise towards the market that the platform thinks would most likely be interested in the products or services that we are offering.</em></p><p>In campaign structure 2, we split our new focus markets into 2 groups — one with higher CPM and the other with lower CPM. <em>Here, we are separating campaigns by CPM so that we ensure High CPM markets get to spend as well.</em></p><p>We suggest for you to run a split test with multiple campaign variations to find out which works best for your business.</p><h3><strong>In Conclusion…</strong></h3><ul><li>Start small — <strong>don’t let the unknown scare you.</strong> Have a clear and simple approach. For example, choose one metric you want to improve on and work backwards towards a solution.</li><li>Never be satisfied with the status quo — <strong>What doesn’t get measured, doesn’t get managed. </strong>Optimisations should never stop. Always be on the lookout and keep asking the team questions on how you can make your campaigns better. Even though test #1’s results were significant — it’s worth retesting this multiple times to prove our hypothesis works.</li><li>Set a <strong>prioritisation framework</strong> — here’s <a href="https://medium.com/lovebonito-growth-journal/how-do-we-spend-3-hours-a-week-to-run-a-b-tests-at-zero-cost-e3c4b25c9960">an article</a> on how we do this</li><li><strong>Stay relevant </strong>— every account structure, business model is different. Make sure to tweak your optimisations to your business needs.</li></ul><p>Optimisations should never be done based on pure speculation and hearsay. Have structured approach in mind but keep things fun and creative by experimenting with different strategies to help amplify your results.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3ffd08daca05" width="1" height="1" alt=""><hr><p><a href="https://medium.com/lovebonito-growth-journal/3-quick-hacks-to-double-your-facebook-advertising-rois-3ffd08daca05">3 Quick Hacks to Double Your Facebook Advertising ROIs</a> was originally published in <a href="https://medium.com/lovebonito-growth-journal">Love, Bonito Growth Journal</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A/B test at zero cost: A busy marketer’s guide]]></title>
            <link>https://medium.com/lovebonito-growth-journal/how-do-we-spend-3-hours-a-week-to-run-a-b-tests-at-zero-cost-e3c4b25c9960?source=rss----4abe546138af---4</link>
            <guid isPermaLink="false">https://medium.com/p/e3c4b25c9960</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[growth-hacking]]></category>
            <category><![CDATA[growth-marketing]]></category>
            <category><![CDATA[ecommerce]]></category>
            <category><![CDATA[ab-testing]]></category>
            <dc:creator><![CDATA[Isabella Zhou]]></dc:creator>
            <pubDate>Fri, 24 Apr 2020 03:36:17 GMT</pubDate>
            <atom:updated>2020-04-28T08:54:32.137Z</atom:updated>
            <content:encoded><![CDATA[<p>While running A/B tests can sound exciting, the execution and planning of it can be chaotic, especially with a small team. Given the 10,000 other immediate priorities we might have on our plate, we often ponder on whether these tests might create any value.</p><h3><strong>Before we start — Some hard truths upfront on A/B tests:</strong></h3><p><strong>Guess what? Running A/B tests can be useless at times </strong>😪</p><ul><li>Time-consuming: 2 weeks are typically needed for one of the variants to be statistically significant during testing</li><li>Lack of actionable next steps: Test results can be counter-intuitive and inconsistent; you have either test again at another time or leave it for now after a month of testing</li></ul><p>With this, we recommend <a href="https://cxl.com/?s=test">a bunch of articles</a> from CXL for further reading. <strong>So why are we still running these tests?</strong></p><ul><li>Validation: It is human nature to make assumptions based on instincts, but the reality might be quite contradictory. Testing is the best way to put things into perspective and leverage data to validate assumptions. Removing guesswork out of the picture, why not?</li><li>As the saying goes, “small tweaks can move the needle”: We recently conducted a test: by placing an urgency label “selling fast” on popular products; we saw a 20% higher add-to-cart rate compared to the original version.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TnS_93LVQFlzuuUXaPj3Zw.png" /><figcaption>Images are for reference only, might not reflect the actual test</figcaption></figure><ul><li>Team work: it helps bring minds together to explore something out of the norm. If you work in a fast-paced startup environment, there is no doubt that your team has tons of daily work to execute. To alleviate them of their daily grind, create something different and exciting for them to participate in. You might be surprised by the creativity and novel ideas!</li></ul><h3><strong>Okay, so how do we do it all within 3 hours a week?</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yYw9akKEhdWfhmXW3qMBNQ.png" /></figure><h4><strong>1. Have a mini squad to run the show</strong></h4><p>Set up a mini squad involving: marketing, product managers (PM), data analysts and product designers for brainstorming and prioritization (in fact, we just have 5 people and 1 from each team as a rep)</p><ul><li><strong>PM/Onsite Manager AKA lead of the A/B test:</strong> Leads the experiment’s brainstorm sessions and documentation. She/he initiates and suggests test ideas and decides what to test first based on the priority framework created, striking a balance between user needs and business impact</li><li><strong>Data analyst:</strong> Helps with initial data exploration if necessary and provide guidelines and statistical suggestions to make sure test results are scientifically significant</li><li><strong>Marketing: </strong>Considers challenges from acquisition channels to website conversion path, thinking about questions like how do we improve conversion rate so that ad spend efficiency will be improved as the additional benefit? She/he has a vested interest to collaborate and ensure that onsite funnels have the most optimal setup</li><li><strong>Product Designer: </strong>Shares user feedback based on ongoing surveys, contributes ideas to UX improvements; she/he creates design changes we need to run the experiment</li></ul><h4><strong>2. Be committed</strong></h4><p>Set regular cadence for discussion — we have a bi-weekly meeting to facilitate that, because without consistency and commitment, no work gets done!</p><p>In reality, we have been cancelling this meeting frequently due to other urgent priorities but we recently overcame this challenge by committing to attending this meeting even <em>if</em> there were no test ideas. We used the time to analyse the performance of ongoing tests and have open discussions on what worked <em>or </em>didn’t<em> </em>work. When different minds come together, it typically leads to a lot of insightful discussions from diverse perspectives.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*YFp1rcvYdEYZ6Kz9" /><figcaption>This chart shows that it also takes us almost <strong>6 months</strong> to get to the right pace, so don’t worry if tests cannot be conducted regularly. It’s better to have 1 test than none at all 😎</figcaption></figure><p>Create a platform to share the results in your organization — this forces us to do it regularly as we have an audience to share it with. We have been circulating a monthly Growth Newsletter company-wide to share details about the tests we’ve run (an example below)</p><figure><img alt="Love, Bonito Growth Newsletter Example" src="https://cdn-images-1.medium.com/max/1024/1*9EFnAYQhmJ8W50QTdlsu8Q.png" /><figcaption>This is one of our recent Growth Newsletter, super simple and keeping things to the minimal 😉</figcaption></figure><p><em>Tip: Create this newsletter via Mailchimp (best of all…it’s free!), so you can also track open rate and click-through rate to steadily evaluate how you’re cultivating an interest in tests across the organization. The eventual results were encouraging and we’ve since received numerous test ideas from other teams! (Most importantly as growth hackers, optimize everything you put out — we even optimize our newsletter subject lines!)</em></p><h4><strong>3. Starting my first test</strong></h4><p><strong>Step 1: Build a prioritization framework for everyone to enter ideas and rank their ideas</strong></p><p>Create a spreadsheet to track all ideas with structured inputs from contributors. We were inspired by the book “Hacking Growth: How Today’s Fastest-Growing Companies Drive Breakout Success” by Sean Ellis (Available on Amazon: <a href="https://www.amazon.com/Hacking-Growth-Fastest-Growing-Companies-Breakout/dp/045149721X">HERE</a>)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*p6SbrCr-Q_9WPmA_" /></figure><ul><li><strong>Number it:</strong> #JustHousekeepingThings Make sure everything is organized for documentation memos!</li><li><strong>Objective: </strong>What do you want to achieve, e.g. improve add-to-cart rate?</li></ul><blockquote>Try to break down a big KPI into mutually exclusive smaller goals. At any one point of time, you are improving one part of the funnel instead of all. For example, we cannot improve conversion rate directly but we can improve one part of the funnel leading up to conversion rate so that the overall conversion rate is improved</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*D7IvJU_J8DtwFFup" /><figcaption>Each rectangular represents one part of the funnel leading up to Conversion Rate (CR)</figcaption></figure><ul><li><strong>Hypothesis:</strong> by doing X action, I predict add-to-cart rate will be improved by 20%</li></ul><blockquote>We usually conduct a quick analysis of the metrics to date and estimate number of users affected by this hypothesis; we also plot out the historical trends of those metrics to extract user behaviours to predict the improvement with the test</blockquote><blockquote>It’s dangerous to start a test without understanding the impact, e.g. you may propose a test for 1 user instead of 1000 users</blockquote><ul><li><strong>Target Audience: </strong>Is this an experiment for all visitors, new visitors or a segment of users?</li><li><strong>Key results to track: PICK 1 PRIMARY GOAL </strong>— misconception: try to test as much as possible from one test so we get more learnings. This isn’t wrong and you can still set up secondary goals for an experiment. However, if the primary goal isn’t clear, when you review the experiment design you’ll be lost and miss the original objective</li></ul><p><strong>Step 2: Prioritize it</strong></p><p>Each point has a rating from 1 to 5 and the tests scoring the highest average will be prioritized.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*DLZ7NVdMz9GJa9dW" /></figure><p><strong>Step 3: Choose an A/B testing tool and create a test</strong></p><p>There are a number of tools in the market but we’ve been on Google Optimize for several reasons:</p><ul><li>It’s FREE and directly linked to Google Analytics to track results, create goals and segmentations based on your historical data</li></ul><blockquote>Limitations:</blockquote><blockquote>5 tests are running concurrently but we’re unlikely to exceed that</blockquote><blockquote>1 primary goal in every test, which is great because it helps us to be laser-focus and overly greedy; however, you can have multiple secondary goals</blockquote><ul><li>It can be easily implemented via GTM or a line of simple code</li><li>You can easily modify anything on a specific page (you name it) via simple HTML/CSS; if you want to go hardcore, there is the JS option too</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VPz-DKNK9hOAfs_o" /></figure><ul><li>It takes care of all statistical calculations and show you the significance of the results</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*IS7rvMDEucvgnjfA" /></figure><p><strong>Step 4: Run it</strong></p><p><strong>Step 5: document it</strong></p><p>Each time a test is selected, a 1–2 page memo should be created according to the following template. The memo will help document the expected results and conclusions from the experiment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/728/0*d3nZI2b_IGCfNzAi" /><figcaption><em>(We referenced an online resource to adapt their template)</em></figcaption></figure><h3><strong>In a nutshell…</strong></h3><ul><li>Set a regular cadence to run tests, even if it’s 1 test at a time</li><li>Stay curious and keep it going — test results might be counter intuitive or meaningless but when you look back 6 months down the road, you will be surprised by how much you’ve accomplished!🤘</li><li>Foster teamwork to gather more test ideas, as we like to say we are #bettertogether and you’ll be amazed by the creativity and diversity of ideas from your fellow team-mates</li><li>Create your <em>own</em> framework to organize tests — what we’ve shared has been based on what has worked for us. you may find other ways that are better suited to your team, company or needs!</li></ul><p><strong><em>Kudos to the A/B testing squad crews who have made this happen:</em></strong></p><p><a href="https://medium.com/u/8a5dccd6f97"><em>Julie Martin</em></a><em> |</em><a href="https://medium.com/u/13e33978127"><em>Song Jiayi</em></a><em>| </em><a href="https://medium.com/u/29e336693121"><em>yan pui</em></a><em> | </em><a href="https://medium.com/u/1fd37392b3e8"><em>Oliviakoh</em></a><em> | </em><a href="https://medium.com/u/91b349bd4e54"><em>Lovin See</em></a><em> | </em><a href="https://medium.com/u/6530c9e3c026"><em>Bluebeard</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e3c4b25c9960" width="1" height="1" alt=""><hr><p><a href="https://medium.com/lovebonito-growth-journal/how-do-we-spend-3-hours-a-week-to-run-a-b-tests-at-zero-cost-e3c4b25c9960">A/B test at zero cost: A busy marketer’s guide</a> was originally published in <a href="https://medium.com/lovebonito-growth-journal">Love, Bonito Growth Journal</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>