<?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 Zunaid Ahmed on Medium]]></title>
        <description><![CDATA[Stories by Zunaid Ahmed on Medium]]></description>
        <link>https://medium.com/@zunaidahmed?source=rss-40ca26c96f4------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*_h_-ZsK2DCsoTA38.</url>
            <title>Stories by Zunaid Ahmed on Medium</title>
            <link>https://medium.com/@zunaidahmed?source=rss-40ca26c96f4------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 04:50:42 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@zunaidahmed/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[Outliers]]></title>
            <link>https://medium.com/@zunaidahmed/outliers-82d339696673?source=rss-40ca26c96f4------2</link>
            <guid isPermaLink="false">https://medium.com/p/82d339696673</guid>
            <category><![CDATA[success]]></category>
            <category><![CDATA[statistics]]></category>
            <dc:creator><![CDATA[Zunaid Ahmed]]></dc:creator>
            <pubDate>Tue, 25 Jun 2019 04:52:24 GMT</pubDate>
            <atom:updated>2019-06-25T04:52:24.174Z</atom:updated>
            <content:encoded><![CDATA[<p>Recently I came across a term called ’Kurtosis’. When I looked up its meaning on Google, it said kurtosis is the sharpness of the peak of a frequency-distribution curve.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aFpuysVFkj8JDuwoKz3T6Q.png" /></figure><p>I have always been an average student in Statistics. And my failure to comprehend what the meaning meant made me dig up a few more posts about it. After reading a couple of posts, I realized what Kurtosis actually means: <strong><em>A measure of outliers present in the distribution.</em></strong></p><p>Being an outlier is an interesting phenomenon. It is quite paradoxical that no matter how everyone tries to be an outlier, a high kurtosis or a high no of outliers would make the behavior to be ordinary.</p><p><em>What if India had 11 Sachin Tendulkars?</em></p><p><em>What if all the paintings made by Leonardo Da Vinci were a masterpiece?</em></p><p><em>What if all bands were as famous as Beatles?</em></p><h4>Can we really work towards being an outlier or to achieve extraordinary success?</h4><p>I was quite intrigued by how Malcolm Gladwell talks about this in his famous book Outliers.</p><p>In the book, he focuses on individuals who achieve extraordinary success- in sports, music or any pursuit. He tries to bust the myth of “self-made man”. Instead of hard work and talent, he talks about how factors beyond a person’s control influence a person’s success.</p><p>He also talks about how once an individual acquires related skills and traits in a particular field, there are other factors which help one succeed and not one’s increased abilities. The other things may be all — social factors, networking or a lucky breakthrough. For example — height in basketball players matters to a certain point. A player with an inordinate height may not necessarily be the best amongst the pack.</p><p>Obviously, talent and hard work are key ingredients for success. And it is the extensive practice which makes talented individuals truly world-class.</p><blockquote><strong>“Practice isn’t the thing you do once you’re good. It’s the thing you do that makes you good.”</strong></blockquote><p>Interestingly, he also highlights how the following factors can affect one’s chances of getting success:</p><ol><li>The month in which an individual is born — Sometimes, the system makes the one born in January to compete with someone born in December, almost a year older.</li><li>How one is brought up — Kids brought up in a more privileged environment are likely to have more practical intelligence through prod and encouragement from parents than kids who have had poorer backgrounds.</li><li>The year in which one is born — Unfair advantages in life can also come from being in the right place at the right time. In order to capitalize on a particular industry/opportunity, one may have an unfair advantage of not only when one was born but also how old one was when the industry was taking off.</li><li>Geographical and cultural advantages — Why Asians are good at Math!</li></ol><blockquote><strong>“Who we are cannot be separated from where we’re from.”</strong></blockquote><p>I believe people always aspire to stand-out. Their success may or may not be a factor of unfair advantages. However, by recognizing the reasons for unfair advantages, we can try creating a level playing field for everyone. But as far as the universe continues to be inherently varied and distinct in its form and nature, there will not be the same starting point for everyone.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=82d339696673" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Hooked or not?]]></title>
            <link>https://medium.com/@zunaidahmed/hooked-or-not-9d8fe265d592?source=rss-40ca26c96f4------2</link>
            <guid isPermaLink="false">https://medium.com/p/9d8fe265d592</guid>
            <category><![CDATA[product-development]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[hooked]]></category>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[product-design]]></category>
            <dc:creator><![CDATA[Zunaid Ahmed]]></dc:creator>
            <pubDate>Sun, 27 Jan 2019 12:57:07 GMT</pubDate>
            <atom:updated>2019-02-17T12:12:49.356Z</atom:updated>
            <content:encoded><![CDATA[<p>I sometimes wonder — what drives me to scroll my Twitter feed endlessly, with a purpose or without, at times like an impulse to do something when I know I’m not up to anything? Sometimes I would even do that when I am introspecting about something. Well, that’s crazy — taking out time to do nothing.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fgiphy.com%2Fembed%2F63I6FXZTXks2A%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fgiphy.com%2Fgifs%2Ftransparent-63I6FXZTXks2A&amp;image=https%3A%2F%2Fmedia.giphy.com%2Fmedia%2F63I6FXZTXks2A%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="640" frameborder="0" scrolling="no"><a href="https://medium.com/media/ada6a1cd60d224860953d0d114368e8b/href">https://medium.com/media/ada6a1cd60d224860953d0d114368e8b/href</a></iframe><p>Is it by coincidence or by design?</p><p><strong>Nir Eyal explains this through the concept of the Hook Model in his famous book — Hooked.</strong></p><p><em>The Hook model is a cycle consisting of 4 stages, which when repeated often enough, will lead the user to form a habit of using that product.</em></p><p>The four stages are:</p><p>1.<strong> The trigger:</strong> an external event that drives us to use a product for the first time.</p><p>I don’t remember what prompted me to use Twitter for the first time. But I guess it was one of those times when people were drifting away from Orkut to Facebook and Twitter.</p><p>1. <strong>The action: </strong>what action needs to be performed to use the product.</p><p>For Twitter, it is simple — Follow your interests -&gt; See what people are talking about -&gt; Join the conversation. I started by following what intrigues me most — cricket, movies, celebs, etc.</p><p>3. <strong>The reward:</strong> the motivation to complete the action.</p><p>Twitter fulfills my need to know people’s opinions about what’s happening around.</p><p>4. <strong>The investment: </strong>something of value like time, money or information that we have invested in the product.</p><p>With Twitter, each time I follow/unfollow people or interests my investment into Twitter increases.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/448/1*5be7trP2fpGJzs834O1Apg.png" /></figure><p>The last step leads to the first step by creating an external trigger whenever a followed account posts something. And when these steps get repeated over and over again, people start developing internal triggers instead of external ones. Eventually, people will have an impulse to use the product without any external driving force. Over time, the user behavior gets shaped for long-term, and we say — we have formed a habit or we have been ‘Hooked’.</p><p>So why is it not easy to give up my Twitter activity? Or for anyone else his Facebook account?<br>The trouble with habits is, it’s tough to change them — to develop one or to ditch one.</p><p>How we at Knudge implemented the Hook Model:</p><p><strong>1. Solving a problem:</strong> People use products to solve a problem they have. We help people improve their English skills. Thus, an external trigger — advertisement, a referral or a recommendation makes someone who is looking to improve his English use the app once.</p><p><strong>2. Enabling and Motivating potential users:</strong> A trigger needs to be backed by a sufficiently motivated user as well as by his capability to do so. We can’t do much about a user’s motivation, and hence we need to work on making it easier for a user to use the product. At Knudge, we provide options of different social sign ups along with email sign up. Thus minimizing any hindrance at the sign-up. Also, till now we have focused on a single page onboarding which tactically subscribes users to a daily feed — modeled to activate or motivate a user regularly.</p><p><strong>3. Rewarding through games and short quizzes:</strong> We have tried to implement an element of variability through different games and quizzes.</p><p><strong>4. Feeling of investment through a course subscription:</strong> Once a user has put in efforts or have spent a few days into the course, it is not easy for him to give up.</p><p><em>An interesting insight:</em></p><p>Of users who use the app for the first 30 days every day, 95% of those users use the app even after 90 days.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ymYJx_tCWr3naK993UP86A.png" /></figure><p><strong>The behavior clearly shows that a user often tends to develop a habit if he/she repeats a particular action frequently.</strong></p><p>How we can apply a Hook model to a product, depends on a product’s use case. We need to identify the parts or features which can be habit-forming and then adjust the aspects of the Hook Model accordingly.</p><p>The application of Hook Model goes a long way in developing a habit-forming product, but we need to ensure that we understand- what our customers want, how the product solves a problem and how we can apply the Hook Model during a user’s lifecycle.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9d8fe265d592" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Looking Back at 2017]]></title>
            <link>https://medium.com/knudge-me/looking-back-at-2017-knudge-me-91b0a0f65638?source=rss-40ca26c96f4------2</link>
            <guid isPermaLink="false">https://medium.com/p/91b0a0f65638</guid>
            <category><![CDATA[google-play]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[android]]></category>
            <category><![CDATA[growth]]></category>
            <category><![CDATA[english]]></category>
            <dc:creator><![CDATA[Zunaid Ahmed]]></dc:creator>
            <pubDate>Sun, 11 Mar 2018 14:00:13 GMT</pubDate>
            <atom:updated>2018-03-12T06:50:50.077Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eH2IJ-l8zvE5ulxy5YSwDQ.png" /></figure><p>Knudge.me had a fantastic 2017. We not only scaled different milestones but also carved out a unique place amongst our users who see Knudge.me app as their daily companion to help them get better at English.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/522/1*4B5WyUEcSgO_TgNtSSepVw.png" /></figure><p>Once we had crossed 50K, we knew we had cracked the code — we had understood what people wanted, and how they wanted to consume it. Correspondingly, we improvised on the learning methodology and focused simultaneously on improving our gaming engine and on personalisation. By the time we had reached 100K installs, we were on the ball and could see the bigger picture a little more clearly.</p><p>We grew 10X in a short span of 6 months and with zero acquisition cost. You can read more about it here:</p><p><a href="https://blog.knudge.me/how-we-grew-10x-from-100k-to-1-million-in-6-months-a5c32ebe69bb">How we grew 10x: From 100K to 1 Million in 6 months</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VU3TPbYadud-tDNTauemIg.png" /><figcaption>10X growth in 6 months</figcaption></figure><p>By October, we had reached 1 million users, and we did it quite convincingly. With over 15K ratings &amp; reviews, we had a rating of 4.8 on Play Store. Our review cloud highlighted words like</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9QmuzIKK5vYNxy3vvRUkmA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KwfzipbEbia_dfNR761Gwg.png" /><figcaption>4.8/5 — TOP RATED APP ON PLAY STORE</figcaption></figure><p>We were extremely pleased when our users wrote to us saying that they were indeed improving their English.</p><p>And we are delighted that we are directly/indirectly becoming a crucial part of their success stories.</p><blockquote>Message to our users: Tell us more about how Knudge.me changed your life. Do write to us at contact@knudge.me. This helps us stay motivated and work harder.</blockquote><h3><strong>What did the Knudge folks do in 2017?</strong></h3><p>There is no doubt that Knudge.me community has grown bigger in 2017. We appreciate the members for their love and for helping us spread the word about the app.</p><p>Our users are spread across more than 200 countries around the globe. From exam aspirants to job seekers, from working professionals to English enthusiasts, the Knudge.me community has made it clear that people of all ages and people with different goals aspire to be fluent in English.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xjRcdUF6MlPtluv46wveVg.png" /></figure><p>These amazing people have created some important milestones while they were on a mission to become proficient in English using Knudge.me:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YADpOuWvm_mfpYa-oBi-UA.png" /></figure><p>Congratulations! You guys make us proud and motivate us to go the extra mile!</p><p>For Knudge.me, 2017 ended with a big bang as it was listed under Google Play’s Best Apps of 2017.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wQn4iGMLnw7KPUFGLah-6Q.png" /><figcaption>Editors’ Choice App</figcaption></figure><p>We are now an Editors’ Choice app in Play Store. This not only validates the quality of the product but also substantiates the efforts the team has put in.</p><p>Check out the link for the entire list of Best Apps of 2017 here:</p><p><a href="https://play.google.com/store/apps/editorial_collection/promotion_topic_3002d08_apps_TP_bestof2017?hl=en">Best Apps of 2017 - Android Apps on Google Play</a></p><p>We promise to keep our noses to the grindstone and strive hard to be a part of this sought-after list next year as well.</p><h3><strong>The road ahead:</strong></h3><p>We have successfully carved out a fresh perspective to mobile-learning by combining gamification and learning. We are aware that our users who cut across all geographies are affluent, aspirational and growing. They have been the growth agents for us. With more than 2M installs on the Google Play Store already, our next steps moving forward would be to focus on extending the modes of learning on the platform and on improving the personalisation through machine learning.</p><p><strong>We are also super excited to share that the bulk of the making of the IOS app is almost done and it should be out very soon!</strong></p><p><strong><em>Looking forward to another fantastic year!</em></strong></p><p><strong>If you liked the post, please click ♥ below and share it with your friends so that they can enjoy it too. Also, do write to us at contact@knudge.me with your feedback/suggestion about the product.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=91b0a0f65638" width="1" height="1" alt=""><hr><p><a href="https://medium.com/knudge-me/looking-back-at-2017-knudge-me-91b0a0f65638">Looking Back at 2017</a> was originally published in <a href="https://medium.com/knudge-me">Knudge.me</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How we grew 10x: From 100K to 1 Million in 6 months]]></title>
            <link>https://medium.com/knudge-me/how-we-grew-10x-from-100k-to-1-million-in-6-months-a5c32ebe69bb?source=rss-40ca26c96f4------2</link>
            <guid isPermaLink="false">https://medium.com/p/a5c32ebe69bb</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[growth-hacking]]></category>
            <category><![CDATA[growth]]></category>
            <category><![CDATA[english]]></category>
            <category><![CDATA[marketing]]></category>
            <dc:creator><![CDATA[Zunaid Ahmed]]></dc:creator>
            <pubDate>Wed, 10 Jan 2018 07:15:04 GMT</pubDate>
            <atom:updated>2026-06-07T12:24:29.548Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DES-YW56BnX5B53w9dzPIw.png" /><figcaption>10X in 6 months!</figcaption></figure><p>Having faced the challenges of mastering the art of speaking fluent English, and having acknowledged how crippled it may make people feel, a year back, three friends embarked on a journey to help people realise the dream of improving their English skills and came up with Knudge.me.</p><p>Today, Knudge.me is not only one of the fastest growing self-learning apps but also one of the most beloved. It has helped more than 1 million learners to become proficient in English.</p><p><strong>What all did we achieve?</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2pTtHXqMh4CwU2ipSNyJYg.png" /></figure><p><strong>One question people often ask us is how did we achieve this growth? Is there any secret ingredient to it?</strong></p><p>Reaching the milestone of 100K took us some time. But it helped us understand the nuances of the product and the growth channels as well as what our customers want. We spent the next few months obsessing over the product and the customer experience. Our focus was to improve the product experience as well as the product discoverability. Some of the actions/strategies which helped us achieve this growth were:</p><ol><li><strong>Measure</strong> —It may sound tedious but this was the key element in our growth as it helped us analyse the gaps as well as the trends in the app usage. We would track metrics rigorously and repeat till we achieved the desired result. We used <a href="https://linktrace.in/">Linktrace.in</a> for app install attribution tracking. Read more about it <a href="https://linktrace.in/blog/deferred-deep-links-implementation-guide/">here</a></li></ol><p>Some of the key metrics which we measured were:</p><p><strong>Acquisition —</strong> New Installs, channels of acquisition &amp; Play Store conversions</p><p>Tracking Play Store conversions helped us improve it by a factor of more than 50%.</p><p><strong>Retention — </strong>D1, M1, M2 &amp; M3</p><p>We focused on improving D1 retention which eventually helped us to improve M1, M2 &amp; M3 retention.</p><p><strong>Engagement —</strong> DAU, MAU, DAU/MAU, Session Duration, No. of sessions/user &amp; Total Sessions</p><p>We focused on optimising our push notification strategy by keeping an eye on the uninstall rate. This helped us to streamline the growth of our Daily Active Users.</p><p><strong>Ratings — </strong>Play Store ratings &amp; reviews</p><p>1-star rating would take away our peace of mind for days. So we did all that, we could — to compensate for a low rating. After a point, users echo a similar sentiment and this helped us to improve the product gradually.</p><p><strong>Other —</strong> Bite-sized lessons &amp; No of games played.</p><p>Identifying a north star metric always helps. And for us, the key metrics were — bite-sized lessons consumed and no of games played.</p><p><em>Conclusion — Don’t fall for vanity metrics. Understand what matters to you. Even an iota of text change can impact the user flow and hence, any direct/indirect change should be measured.</em></p><p><strong>2. Quality &amp; Customer Feedback</strong></p><p>“Big things often have small beginnings” — We focused on doing one thing right at a time. But each small step has to be finite and complete — focussed on end-to-end user experience.</p><p>We can easily boast about how stable our app has been since its launch. We did push things fast but one thing we have never compromised upon is the quality. Whether it’s a text font or something as trivial as a font colour, it had to be the best. This has clearly reflected in our play store ratings. There are few apps on the play store with a million installs and even fewer with more than a million installs having 4.8 ratings.</p><blockquote>“I think our team would say to focus on the entire user experience — each pixel, button, interaction is an opportunity to delight your players/users.”</blockquote><p>— Paul Murphy, co-founder and CEO of Dots</p><p>He goes on saying: “Once we realized we had something, we measured and tested endlessly until it felt just right.”</p><p>All of this has been a gradual process by talking to our users, understanding their concerns and incorporating their viable feedback. We don’t skip any emails/reviews given by our users. We get more than 20 emails everyday, more than 40 reviews on the play store and more than 100 feedback responses. We talk to each one of them directly or indirectly to understand and address their concerns. At the end of the day, its the customer’s love which can help you grow organically.</p><p><em>Conclusion — Do not push half-baked features; and don’t make promises which you can’t fulfil.</em></p><p><strong>3. Growth Hacks</strong></p><p>We were lucky enough to receive this advice early on — d<em>on’t spend a penny on marketing.</em></p><p>There are a lot of things which one can do to drive installs. The challenge is to drive installs without spending anything on marketing. One needs to prioritise resources based on channels which can yield sustainable installs.</p><p>Some of the steps which we took to drive the growth:</p><p><strong>Use the product to drive growth up</strong></p><p>Identify key elements which are brag-worthy enough.</p><p><strong>Play store optimisations</strong></p><p>Title, Banner, Snapshots, Icon &amp; Keywords</p><p><strong>Social Activity</strong></p><p>All social platforms — Facebook, Twitter, Google Plus etc, Groups, Forums etc.</p><p><strong>Other</strong></p><p>Blogging, SEO, App Reviews etc.</p><p><em>Conclusion — Every install counts.</em></p><h3>The secret ingredient</h3><p><em>As Mr. Ping has famously said — “There is no secret ingredient. To make something special, you just have to believe it’s special.”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*E8UQ5aLf1-BlYx-50NUYNw.jpeg" /></figure><p><strong>If you liked the post, please click ♥ below and share with friends so others can enjoy it too. Also, do write to us at contact@knudge.me with your feedback/suggestion about the product.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a5c32ebe69bb" width="1" height="1" alt=""><hr><p><a href="https://medium.com/knudge-me/how-we-grew-10x-from-100k-to-1-million-in-6-months-a5c32ebe69bb">How we grew 10x: From 100K to 1 Million in 6 months</a> was originally published in <a href="https://medium.com/knudge-me">Knudge.me</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Mobile App Launch CheckList — Android]]></title>
            <link>https://medium.com/@zunaidahmed/mobile-app-launch-checklist-android-3f71355b2e86?source=rss-40ca26c96f4------2</link>
            <guid isPermaLink="false">https://medium.com/p/3f71355b2e86</guid>
            <category><![CDATA[knudgeme]]></category>
            <category><![CDATA[android-app-development]]></category>
            <category><![CDATA[app-launch]]></category>
            <category><![CDATA[product-launch]]></category>
            <category><![CDATA[android]]></category>
            <dc:creator><![CDATA[Zunaid Ahmed]]></dc:creator>
            <pubDate>Thu, 09 Feb 2017 22:34:27 GMT</pubDate>
            <atom:updated>2017-02-26T06:11:11.835Z</atom:updated>
            <content:encoded><![CDATA[<p>Sharing some learnings and steps that we followed for the launch of <a href="https://play.google.com/store/apps/details?id=com.knudge.me">Knudge.me</a>(a micro learning platform to help people improve their English and Math skills by leveraging AI and gamified learning. The app is rated 4.8 on play store with more than 60K installs(with zero paid marketing) within a span of 6 months.)</p><p><strong>The beginnings: </strong>The first step has to be the right one. Build first, improve later — doesn’t work when it comes to building a good product. Make sure you follow the best practices. The beginnings may take time but eventually you would save yourself both time and effort.</p><ul><li><strong>Use Source/Version Control </strong>— Be it Github or Bitbucket, whatever you are comfortable with. We chose Bitbucket since it offered free private repositories.</li><li><strong>Latest features </strong>— Android is evolving and so are the best practices around it. Google I/O’s app is one of the best source if you want to foray with a latest feature for your app or looking for a best structure for organising your code. Following MVVM + DataBinding saved us from writing extra lines of code. Less code means lesser bugs. Not just that, MVVM brought a lot of reusability to the presentation layer.</li><li><strong>Design </strong>— Material Design it is! Check any Google app like Youtube or any popular app like Instagram, Evernote etc. <a href="https://zeplin.io/">Zeplin </a>is a must have tool to easily collaborate and fasten up your pace of development.</li></ul><p><strong>The journey :</strong> Make sure every piece on your app is pluggable through Backend service. This helps to avoid unnecessary app updates and better control for any feature which can be switched on/off at any point of time.</p><p><strong>The closure : </strong>Think you got a working product? Let’s check if you got time to incorporate the following too.</p><ul><li><strong>Analytics </strong>— Wire every action, screen, goal to the analytics service. We chose Google Analytics since its free. GA count by sampling of data points and to track individual instances, you might have to go with CleverTap or MixPanel.</li><li><strong>Soft Update/ Hard Update</strong> — People don’t get everything right in first time. So let’s have a backup. Use Soft Update / Hard update feature to push the user to the latest version.</li><li><strong>Error Reporting</strong> — Our backend service is on ROR. We chose sentry for error monitoring and reporting. For App crash reporting we are using Crashlytics.</li><li><strong>Attribution Tracking</strong> — We were tracking our installs referrals source through GA unless it one day stopped giving results. We are yet to find the reason behind the same. You can explore Apsalar or any other service of this kind for the same.</li><li><strong>Testing and Test Environments </strong>— Separate out your production and test accounts for any service. App logs should be disabled for production apk. Any change to gradle or any library version change should be backward/forward compatible. 2G/3G/Wireless testing should be done separately. Also any UI change should be tested on different size mobile screens.</li><li><strong>Push Notifications</strong> — Since now Parse is already a history, a wise option is to go for FCM. In our case, we started with Parse and have now migrated to FCM.</li><li><strong>Deep Linking</strong> — We avoided it initially. But it is very important for transactional apps. Also deep linking helps in SEO. We are in process of evaluating Branch.io for the same.</li><li><strong>Email Service — </strong>We started with SendGrid which provides 12000 free emails per month. Our monthly emails requirement is now around 50K. And we are making the best use of Mailchimp credits now.(Knudge.me is part of Facebook’s FbStart program which offers free credits worth $40K for multiple services.)</li><li><strong>In app feedback and sharing — </strong>How are your customers going to write feedback? Is there any incentive to share the app?</li><li><strong>App size —</strong> To minimize the size of your app binary, make sure that you run the <a href="https://developer.android.com/tools/help/proguard.html">Proguard</a> tool.</li><li><strong>App Permissions- </strong>People don’t like giving any permission explicitly. Be clear with why you need permission and when do you want to prompt user for permission.</li></ul><h3><strong>Ready to Launch?</strong></h3><ul><li>A beta launch — It is very important to go for a beta launch. Remove all the pending bugs and issues.</li><li>A beta user — Don’t explain your product and let your beta user play around with the app. Improve the user flows by observing how users are interacting with app. Take feedback very seriously. Prioritise the feedback/suggestions. Delay the final launch if you have got some important data points to make relevant changes. After beta, we almost had to revisit our complete product but in the end that effort was worth it.</li><li>Set up all the possible user acquisition channels — A clean landing page, Facebook, Twitter, Instagram profiles should be up and ready.</li></ul><p><strong>Play Store — </strong>One of the most important step to get organic installs. Strategise for higher conversions and increased visibility on the store.</p><ul><li>ASO : Title, Short description, Full description should be carefully and strategically drafted. AppAnnie and SensorTower are the two tools that we use for Play Store Optimisations.</li><li>Graphics : Icon, Feature Graphic, Screen shots are the first impressions that a user would get before they are going to install the app. Make the best use of it. A demo video can also be used. <a href="https://theapplaunchpad.com/">AppLaunchPad</a> can be used to create some good graphics in quick and easy way. In case your app is compatible for Tablets, good to have screenshots for tablets too.</li><li>Privacy Policy — You can read this piece to get better idea about privacy policy.</li></ul><p><a href="https://www.xda-developers.com/google-plays-new-privacy-policy-rule-goes-into-effect-on-march-15th/">Google Play&#39;s New Privacy Policy Rule Goes into Effect on March 15th</a></p><p><strong>Server</strong></p><ul><li>Monitoring and Performance — We use New relic for application performance management and monitoring.</li><li>Logs — We are in process of setting up of ELK stack for log analysis and analytics.</li><li>Database back ups — This goes without saying.</li><li>Security and scalability — One aspect that we tend to avoid initially but should be appropriately handled to avoid any setbacks later.</li></ul><p><strong><em>And, that’s it. We are good to go.</em></strong></p><p><strong>Post Launch:</strong></p><ul><li><strong>Early Reviews — </strong>Ask your friends to rate app 5 stars. Try getting at least 100 5 star ratings. This would help to get the initial boost.</li><li><strong>App Update — </strong>Do a percentage roll-out. Monitor for any error and accordingly release the app to 100% users. App Update description should be elaborate and avoid writing only ‘Bug Fixes’.</li><li><strong>Monitor your App mentions and Reviews and Emails</strong> — Make sure you answer to any customer complaint and to thank them if they rate well.</li><li><strong>Product Hunt Launch</strong> — We went for product hunt launch after a month. Though product hunt launch should be done as early as possible to get the initial bump.</li><li><strong>A/B Testing — </strong>You can try A/B tests to evaluate better conversions for different graphics etc. Play store provides this feature.</li></ul><p><em>Some of the issues that we faced after the app launch. Taking the following into account can avoid you some pains :</em></p><ul><li>One outage : Log files, Db Backups, or any temp files should be constantly removed from the disk. And this gave us our first and the only outage.</li><li>Server Scaling : We had to scale our volume storage for our Database Server. This brought us a downtime of 9 minutes. This can be planned in advance to avoid any issue when the database size increases.</li><li>Volley: We were facing Volley HTTP library(Android) erratic behaviour due to its default settings configuration. We use to experience some random retries and some parameters used to be missed in the HTTP request. We have reduced the number of such issues but still once in a while we experience a missing parameter issue. In retrospect, we believe Retrofit would have been a better library to use. Learnings : Always go for library which has active development.</li><li>Push Notification: We made a wrong choice of going with Parse when it was supposed to come down within a year. It took a lot of time to gradually migrate all our users to FCM.</li><li>Google Play Services version change: We didn’t check the backward compatibility after changing the play services version. The app stopped receiving notifications. And once it so happened that we didn’t check the forward compatibility with Android N and that gave us some crashes.</li></ul><p><strong>If you liked the post, please click ♥ below and share with friends so others can enjoy it too.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3f71355b2e86" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Sidekiq and Scaling]]></title>
            <link>https://medium.com/@zunaidahmed/sidekiq-and-scaling-dd6f1969b68e?source=rss-40ca26c96f4------2</link>
            <guid isPermaLink="false">https://medium.com/p/dd6f1969b68e</guid>
            <category><![CDATA[rails]]></category>
            <category><![CDATA[sidekiq]]></category>
            <category><![CDATA[ruby-on-rails]]></category>
            <category><![CDATA[concurrency]]></category>
            <dc:creator><![CDATA[Zunaid Ahmed]]></dc:creator>
            <pubDate>Mon, 07 Nov 2016 12:21:58 GMT</pubDate>
            <atom:updated>2016-11-12T17:31:52.517Z</atom:updated>
            <content:encoded><![CDATA[<p>Sidekiq is an efficient way to process background jobs in ruby. Sidekiq makes use of threads to process multiple jobs at same time in the same process.</p><p>By default, a sidekiq process creates 25 threads. Though the amount of concurrency can be tuned by changing the concurrency in sidekiq configuration file.</p><p><strong>Some of the pertinent questions related to sidekiq scaling :</strong></p><ol><li>What is that optimum number of concurrency for a sidekiq process?</li><li>Does Increasing concurrency leads to increasing performance of sidekiq?</li><li>How we can get maximum throughput from sidekiq?</li></ol><p>Here, in this blog, I would mainly discuss about the issues that I faced while handling at least a million jobs using sidekiq and how I was able to improve the performance. (This was during my short stint in a well reputed tech based company. )</p><p><strong>The incumbent Set up :</strong></p><p><em>Server</em> : A single core system with 8 Gb RAM</p><p><em>Requirement</em> : 10 lakhs jobs to be processed.</p><p><em>Duration</em>: It took around 17–18 hours to process 10 lakhs jobs</p><p><em>Sidekiq</em>: The concurrency of sidekiq process was 100</p><p><em>Assumption</em>: Increasing sidekiq concurrency would increase the throughput.</p><p>Processing a million jobs in 17–18 hours, clearly shows that there were serious problems with the existing architecture, code or configurations. The dev ops increased the concurrency till they started facing Database connection issues.</p><p><strong>So what should be the max concurrency of a sidekiq worker ?</strong></p><blockquote>“Don’t set the concurrency higher than 50. I’ve seen stability issues with concurrency of 100, for example. Note that ActiveRecord has a connection pool which needs to be properly configured in config/database.yml to work well with heavy concurrency. Set the pool setting to something close or equal to the number of threads” - <strong>Mike Perham, Sidekiq author</strong></blockquote><p><em>Then how should we increasing sidekiq’s performance?</em></p><p>The first thing I experimented was to see the impact of increasing the CPUs on performance. The server configuration was changed to dual core. It had no impact for the following reason :</p><blockquote>‘Each Sidekiq process running on MRI will only use one core, regardless of the number of threads. To get the benefit of multiple cores, you should run several Sidekiq processes.’ — <strong>Sidekiq help</strong></blockquote><p><strong>Conclusion : To run multiple sidekiq workers to make use of all the available cores.</strong></p><p><strong>New System Configuration:</strong></p><p><em>Server</em> : Dual Core with 8 Gb RAM</p><p><em>Requirement</em> : 10 lakhs jobs to be processed.</p><p><em>Sidekiq</em> : 3 sidekiq workers with configuration of 50 each.</p><p><em>Results</em> : It took around 3–4 hours to process a million jobs.</p><p><strong>The change in set up resulted in 4–5 times better performance. This was a significant improvement with scope to further improve it by tweaking the concurrency number, sidekiq workers and multiple cores.</strong></p><p>Attaching a response to this post from the Sidekiq Author himself, <strong>Mike Perham</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OSiYVuxmq7i1dc67SouM7g.png" /></figure><p><em>PS: It would be great if someone can share a way through which it can be determined the opt number for sidekiq concurrency , sidekiq workers and multiple cores.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dd6f1969b68e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Innovative apps for effective learning, building habits and improving long term memory]]></title>
            <link>https://medium.com/@zunaidahmed/innovative-apps-for-effective-learning-building-habits-and-improving-long-term-memory-bbf20013f223?source=rss-40ca26c96f4------2</link>
            <guid isPermaLink="false">https://medium.com/p/bbf20013f223</guid>
            <category><![CDATA[education]]></category>
            <category><![CDATA[android]]></category>
            <category><![CDATA[mobile-learning]]></category>
            <category><![CDATA[learning]]></category>
            <category><![CDATA[knudge]]></category>
            <dc:creator><![CDATA[Zunaid Ahmed]]></dc:creator>
            <pubDate>Sun, 30 Oct 2016 14:03:10 GMT</pubDate>
            <atom:updated>2016-10-30T14:03:10.661Z</atom:updated>
            <content:encoded><![CDATA[<p>As our life becomes increasingly digital everyday, it transforms behaviours, habits and preferences of learners convincingly. Today’s student is distracted, impatient and gets overwhelmed easily. And one must look for answers outside of traditional ways and channels to find an effective solution.</p><p><strong>So how do we improve our ability for learning to learn in such a complex environment?</strong></p><p>Here are some ways which can be essential in achieving this goal, by understanding how a human brain is wired to learn . Let’s discuss each of these:</p><p><strong>Micro Learning :</strong> The concept of bite-size learning is simple. Instead of delivering training over a continuous course, you break it into bite-size chunks so learners can easily fit it into a vacant time slot in their busy daily schedules.</p><p><strong>Mobile learning :</strong> The on-the-go lessons should continue even when one is waiting for a bus or sitting in a cafe. Gaining knowledge for long term is a continuous process.</p><p><strong>Spaced Learning :</strong> Learning is greater when studying is spread over time, as opposed to studying the same amount of time in a single session. Spacing effect minimises our forgetting curve.</p><p><strong>!(Don’t Repeat Yourself) : </strong>Repeating what you have learnt creates retention of information for a longer duration. Reinforcing the repetitions by diversifying the learning resources solidify what you have learnt by retaining the interest through varied study material.</p><p><strong>Visuals and Graphics : </strong>A large body of research claims that images create visual memory in our brains and help us to retain and retrieve information effectively.</p><p><strong><em>5 most innovative apps,which can provide effective learning solutions by building habits and by improving long term memory are:</em></strong></p><p><a href="https://www.elevateapp.com/"><strong>Elevate :</strong></a><strong> </strong>Its a brain training app designed to improve attention, speaking skills, processing speed, memory, maths skills, and more. The brain training training games helps user to improve one’s cognitive skills over time effectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vwPhOthStyB5GzcUeKaPPA.png" /><figcaption><strong>Elevate</strong></figcaption></figure><p><a href="http://www.thefabulous.co/"><strong>Fabulous</strong></a><strong> : </strong>It motivates user to reach their Health &amp; Productivity goals by following a step-by-step program. It provides science based coaching by building habits to increase energy, feel vibrant health, lose weight and sleep better.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bgrbe9jr5SPpAMjaafj4QA.png" /><figcaption>Fabulous</figcaption></figure><p><a href="https://www.yourprimer.com/"><strong>Primer by Google</strong></a><strong> : </strong>Bite-sized lessons in marketing skills. Interactive activities will help you learn concepts quickly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZR7m9BlleNHgHWK0z4BxMw.png" /><figcaption>Primer by Google</figcaption></figure><p><a href="http://www.knudge.me/"><strong>Knudge.me</strong></a><strong> : </strong><a href="http://knudge.me/">Knudge.me</a> aims to make learning easy and helps build long term memory. The platform works on a complex algorithms developed using advanced techniques like adaptive learning, spaced learning which are aimed to provide effective learning. The notification based micro learning platform teaches through infographics and gamification. Download the app from : <a href="https://play.google.com/store/apps/details?id=com.knudge.me&amp;utm_source=medium&amp;utm_medium=blog&amp;utm_campaign=interesting_apps">Knudge.me</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OSXjDYlG5BYrkvEAkPaOuw.png" /><figcaption>Knudge.me — Learn smartly!</figcaption></figure><p><a href="https://curiosity.com/"><strong>Curiosity</strong></a><strong> : </strong>Get smarter every day with daily “Smart Facts” that help you to improve your knowledge while having fun!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hrGEy-XcS4SXKnWQfSsANQ.png" /><figcaption>Curiosity</figcaption></figure><p><strong>If you liked the post, please click ♥ below and share with friends so others can enjoy it too.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bbf20013f223" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>