<?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 Jega Pradeeba on Medium]]></title>
        <description><![CDATA[Stories by Jega Pradeeba on Medium]]></description>
        <link>https://medium.com/@pjega92?source=rss-84225be8d71------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*Kdd5HWqYsfsgUB-o</url>
            <title>Stories by Jega Pradeeba on Medium</title>
            <link>https://medium.com/@pjega92?source=rss-84225be8d71------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 18:44:21 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@pjega92/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[A Netflix Special: Product Strategy Case Study (Tech Focused)]]></title>
            <link>https://medium.com/@pjega92/a-netflix-special-product-strategy-case-study-tech-focused-ec291c7b5597?source=rss-84225be8d71------2</link>
            <guid isPermaLink="false">https://medium.com/p/ec291c7b5597</guid>
            <category><![CDATA[technical-analysis]]></category>
            <category><![CDATA[product-strategy]]></category>
            <category><![CDATA[netflix]]></category>
            <category><![CDATA[case-study]]></category>
            <category><![CDATA[product-management]]></category>
            <dc:creator><![CDATA[Jega Pradeeba]]></dc:creator>
            <pubDate>Mon, 22 Jul 2024 10:13:39 GMT</pubDate>
            <atom:updated>2024-07-22T10:20:09.599Z</atom:updated>
            <content:encoded><![CDATA[<h3>A Netflix Special — Product Case Study (Tech Focused)</h3><h3>Contents:</h3><ul><li><a href="#9bf8">Snapshot</a></li><li><a href="#93de">Background</a></li><li><a href="#319a">Defence: The Simian Army</a></li><li><a href="#04f6">Midfield: AWS and Infrastructure</a></li><li><a href="#4e60">Offense: Personalisation / Search and R+D</a></li></ul><h3>Snapshot</h3><p>Netflix has transitioned from a media-logistics business to a media-tech giant, with the sole purpose of giving their users the ability to watch what they want, when they want. In this case study, we analyse how Netflix were able to leverage technologies to build such a dominant product on three levels: defence (how it meets the bare minimum of user expectations), midfield (how its infrastructure allows it to assist in meeting user expectations and facilitate innovation), offense (how Netflix’s innovations and leveraging of data allow it to further its competitive moat):</p><p>The Defence: The Simian Army represents Netflix’s defensive strategies, focusing on resilience and system robustness. By deliberately introducing failures into their system, Netflix enhances its ability to withstand real-world issues, ensuring high availability and reliability for users.</p><p>The Midfield: AWS and cloud infrastructure are akin to the midfield, serving as the backbone that supports both defensive and offensive plays. This infrastructure allows Netflix to scale efficiently, handle massive amounts of data, and maintain operational flexibility, enabling rapid deployment and innovation.</p><p>The Offense: Personalization, search, and R&amp;D are akin to some of Netflix’s offensive strategies, as they are crucial for strengthening its competitive edge. Through advanced algorithms and continuous experimentation, Netflix tailors content to individual preferences, enhancing user engagement and satisfaction. This forward-looking approach drives growth and sets Netflix apart from competitors.</p><h3>Background</h3><p>Netflix is the leading streaming entertainment service, with around <a href="https://www.similarweb.com/blog/insights/media-entertainment-news/streaming-q1-2023/#:~:text=post%2Dpandemic%20world.-,Market%20Share%20Over%20Time,when%20compared%20to%20leading%20platforms.">44%</a> market share, offering a wide array of TV shows, movies, anime, documentaries, and more across a variety of genres and languages. Founded in 1997 by Reed Hastings and Marc Randolph in Scotts Valley, California, it initially started as a DVD rental service by mail. However, Netflix pivoted to streaming in 2007, marking the beginning of its transformation into a global streaming powerhouse.</p><p>Today, Netflix operates in over 190 countries, providing millions of subscribers with content that can be watched virtually anywhere, on any internet-connected screen. Known for its significant investment in original content, Netflix has produced critically acclaimed series and films, further solidifying its position as a leader in the entertainment industry.</p><p>Netflix employs around 12,000 staff, with two-thirds being <a href="https://aws.amazon.com/solutions/case-studies/netflix-case-study/">technical</a>. Despite transitioning from a media-logistics service, to a media-tech company, at its heart it still serves the same purpose: give their users the ability to watch what they want, when they want. Technology just enhanced that service offering, whilst finding new ways to create more value for each user.</p><p>“At Netflix, we want to entertain the world. Whatever your taste, and no matter where you live, we give you access to best-in-class TV series, documentaries, feature films and games. Our members control what they want to watch, when they want it, in one simple subscription.” — <a href="https://about.netflix.com/en">Netflix</a></p><p>Throughout the OTSOG case series so far, we tend to analyse companies from both a technical-product and product-marketing lens. For this Netflix Special however, we’ve decided to double-down on the technical angle, simply because there are a number of learning points on this angle alone that are worth an in-depth exploration.</p><p>To simplify matters further, we’ve split the technical strategies into 3 parts, as mentioned in the <a href="https://docs.google.com/document/d/15QIzqPg2iEOsw1UQPxHAS1ZsEF3-bIEAB_l4jsFOffs/edit#heading=h.kaoo06icts92">snapshot</a>: the defence, midfield and offense. Hopefully, this does not come across as an oversimplification, but we believe that it helps provide a clearer framework to understand the multifaceted approach Netflix employs to innovate, secure, and expand its platform effectively.</p><p>The strategies discussed here are by no means an exhaustive list, but they illustrate key elements of how Netflix leverages technology to maintain its industry leadership and innovate within the streaming landscape.</p><p>I hope you enjoy the read.</p><h3>The Defence: The Simian Army</h3><p>Arguably, a good metric of the capabilities of a company’s tech team, is their ability to build and ship open-source, developer-centric products, that are widely used, but not known to the average consumer. Not only do such contributions showcase technical ability and thought leadership, but they also highlight a commitment to advancing the technology ecosystem: Facebook built React (one of the most widely-used JavaScript libraries), LinkedIn built Kafka (a real-time data-streaming platform), Google built Kubernetes (a system for automating and deploying containerised applications), and Netflix built Chaos Monkey and subsequently the Simian Army.</p><p>Chaos Monkey is a tool that intentionally introduces failures into the infrastructure to test its resilience and reliability. It works by disabling computers in Netflix’s cloud production network to test how remaining systems respond to the outage, thereby ensuring that Netflix’s services can tolerate the failure without any customer impact — “we have to be stronger than our<a href="https://netflixtechblog.com/the-netflix-simian-army-16e57fbab116"> weakest link</a>” — Director of Cloud &amp; Systems Infrastructure at Netflix, Yury Izrailevsky.</p><p>The Netflix Cloud team describe its philosophy as this:</p><p>“Imagine getting a flat tire. Even if you have a spare tire in your trunk, do you know if it is inflated? Do you have the tools to change it? And, most importantly, do you remember how to do it right? One way to make sure you can deal with a flat tire on the freeway, in the rain, in the middle of the night is to poke a hole in your tire once a week in your driveway on a Sunday afternoon and go through the drill of replacing it. This is expensive and time-consuming in the real world, but can be (almost) free and automated in the <a href="https://netflixtechblog.com/the-netflix-simian-army-16e57fbab116">cloud</a>.”</p><p>“By running Chaos Monkey in the middle of a business day, in a carefully monitored environment with engineers standing by to address any problems, we can still learn the lessons about the weaknesses of our system, and build automatic recovery mechanisms to deal with them. So next time an instance fails at 3 am on a Sunday, we won’t even <a href="https://netflixtechblog.com/the-netflix-simian-army-16e57fbab116">notice</a>.”</p><p>From the success of the Chaos Monkey, the Simian Army was born. This was a suite of different types of chaos monkeys, that would introduce different kinds of chaos and hence test different vulnerabilities, assessing Netflix’s ability to survive them — similar to how the human body has different types of immune system cells, but all still serving the same purpose, to ensure the health of the host. The army<a href="https://netflixtechblog.com/the-netflix-simian-army-16e57fbab116"> includes</a>:</p><ul><li>Latency Monkey — introduces artificial delays, making them intentionally slow, to see if the serves still work well when they are requests to the server are delayed.</li><li>Conformity Monkey — finds instances that don’t adhere to Netflix’s best practices and shuts them down.</li><li>Doctor Monkey — checks if our services are healthy by looking at their regular health checks and other signs, like how busy the CPU is. If it finds services that aren’t doing well, it takes them out of use.</li><li>Janitor Monkey — searches for unused resources and disposes of them to ensure the cloud environment is running free of clutter.</li><li>Security Monkey — finds security violations or vulnerabilities, and terminates those instances.</li><li>Chaos Gorilla and many more</li></ul><p>Netflix’s strategy of deliberately introducing problems into their systems to uncover weaknesses early has revolutionised the way companies ensure the robustness of their distributed systems. This approach, known as chaos engineering, has established a new standard for testing system resilience, inspiring a shift across the tech industry towards more reliable and fault-tolerant software development practices.</p><p>By identifying and addressing vulnerabilities before they escalate, the Simian Army fortifies Netflix’s defence on service delivery, ensuring Netflix can continue to do the thing that customers are paying it for — let them watch the content they want, when they want it, without failure.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/736/0*ujEA0q7TUeEQCP-f" /></figure><p><em>Figure 1 — A pictographic of Netflix’s Simian Army, </em><a href="https://in.pinterest.com/pin/643170390539215087/"><em>source</em></a><em>.</em></p><h3>The Midfield: AWS and Infrastructure</h3><p>“We serve hundreds of thousands of requests per second..we serve over 1 billion of hours of content per month..and all of this is being run out of AWS infrastructure” — <a href="https://aws.amazon.com/solutions/case-studies/netflix-case-study/">Eva Tse</a>, Director of Big Data Platform, Netflix</p><p>Amazon Web Services (AWS) is a comprehensive cloud computing platform, providing a wide range of services like computing power, database storage, and content delivery. It’s known for its scalability, reliability, and flexibility, making it a popular choice for businesses of all sizes. What makes the use of AWS for Netflix even more interesting, is that Amazon Video is a direct competitor to Netflix, so why use it?</p><p>Uprooting your set infrastructure is often driven by necessity. For Netflix, this necessity came from two <a href="https://www.youtube.com/watch?v=XrWII4ewrXA">catalysts</a>:</p><ul><li>First, an initial wake-up call — a two-day outage of their streaming services</li><li>Second, the uptick in streaming demand — which provided the momentum to see them through the transition.</li></ul><p>Before 2008, the strategies and practices for handling system failures in software development were mainly focused on traditional IT infrastructure, i.e., dedicated hardware and in-house data centres. Developers and engineers employed methods like fault tolerance and redundancy, but the widespread adoption of cloud computing and distributed systems was still very nascent.</p><p>Netflix started to rethink this following a 2-day <a href="https://www.youtube.com/watch?v=XrWII4ewrXA">outage</a> to their streaming service, due to a hardware failure (specifically their data storage infrastructure) in 2008. The outage underscored the limitations of traditional, physical infrastructure in meeting the demands for continuous service availability. Netflix realised that to achieve the level of resilience and operational agility required, they needed an infrastructure that could dynamically scale and recover from failures seamlessly.</p><p>Cloud computing, with its promise of high availability, scalability, and cost-efficiency, offered a solution. AWS, known for its robust cloud services, presented an opportunity for Netflix to leverage these cloud benefits. By adopting AWS, Netflix could focus on improving and expanding its service offerings without the constraints of managing physical hardware, thus ensuring that its streaming service could reliably meet the rapidly growing demand from users worldwide.</p><p>But another existential threat was also needed to push them through the transition. In 2009, for the first time, streaming demand <a href="https://venturebeat.com/social/netflix-streams-already-rushing-past-dvds-in-2009/">overtook</a> their DVD rental demand, meaning they needed a vast increase in their data centre capacity. This is important because, like many companies at the time, their IT infrastructure was scaled to how many employees they have (i.e. internal company demand-driven), but, especially for customer-facing companies like Netflix, it needed to scale to how many customers they have.</p><p>This shift from DVD to streaming is the crux of the need for AWS. Because for any one interaction between a user and the movie they watch, Netflix had a 100-fold increase in the amount of traffic to their data centre per movie/TV show (things like storing where you stopped, what shows you’ve seen etc); and combined with a 10-fold increase in views per week, Netflix now had a <a href="https://www.youtube.com/watch?v=XrWII4ewrXA">1000-fold increase</a> in traffic to their data center.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*eNyTZArJTUXDNQf_" /></figure><p><em>Figure 2 — The image presents the logistics of a DVD rental business, showing limited customer interactions with a data centre for personalised browsing and a physical process for mailing DVDs to and from a shipping site.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*HYfZfYYJ_a8ImdrN" /><figcaption><a href="https://aws.amazon.com/solutions/case-studies/netflix-case-study/">Source</a></figcaption></figure><p><em>Figure 3 — The image outlines the exponential growth in a streaming business, highlighting a 10x increase in weekly views, 100x more data centre traffic per view, and a 1000x rise in traffic with each customer shift to streaming.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sipgxnF7glBPxH1u" /><figcaption><a href="https://aws.amazon.com/solutions/case-studies/netflix-case-study/">Source</a></figcaption></figure><p><em>Figure 4 — The diagram shows the workflow of a streaming service with a data centre processing personalised content and a CDN distributing encoded video to users. It highlights infrastructure components for delivering a seamless binge-watching experience.</em></p><p>To figure out how to build data centres fast and keep up with the demand, Netflix had two options:</p><ul><li>(1) recruit a data centre operations team, try to guess how much capacity they would need and build it before it was needed (massive upfront costs).</li><li>(2) use the elastic compute services of AWS (EC2) which scaled with the demand needed, despite getting into bed with one of their biggest competitors (Prime Video).</li></ul><p>After a meeting with Amazon and an understanding that AWS and Prime Video were completely separate businesses, Netflix was satisfied enough to build with AWS and hasn’t looked back since.</p><p>A few case examples of the core benefits of AWS for Netflix:</p><p><strong>Cost efficiency:</strong></p><p>With AWS’s pay-as-you-go model, Netflix pays only for the compute time that is actually used. This model allows for billing by the hour or even by the second, depending on the instance type, providing flexibility in resource utilisation. As a result, tasks can be processed faster by using more machines for a shorter time without increasing the total compute hours and, consequently, the cost.</p><p>To illustrate: Encoding movies and shows (i.e. optimising raw video files for online streaming) requires substantial compute power, especially as the volume of content grows. AWS’s pricing structure (where 10 instances for 10 hours costs the same as 100 instances for 1 hour) enables Netflix to scale up their compute capacity as needed to swiftly manage large <a href="https://aws.amazon.com/solutions/case-studies/netflix-case-study/">backlogs</a> of shows. This capability allows for rapid content availability to customers without incurring additional costs.</p><p><strong>Speed of iterations:</strong></p><p>Before the adoption of cloud computing, Netflix software engineers had to request additional physical servers from the IT department to conduct tests and experiments, a process that could take days or even <a href="https://aws.amazon.com/solutions/case-studies/netflix-case-study/">weeks</a>. This separation of servers was necessary to maintain isolation and prevent any experimental changes from impacting the live product.</p><p>However, with AWS, engineers can now launch virtual instances for their experiments almost instantaneously — <a href="https://aws.amazon.com/solutions/case-studies/netflix-case-study/">within seconds or minutes</a>. This capability greatly enhances their flexibility, allowing them to run more experiments at a faster pace, thereby significantly speeding up Netflix’s innovation cycle (something we cover in more detail under <a href="https://docs.google.com/document/d/15QIzqPg2iEOsw1UQPxHAS1ZsEF3-bIEAB_l4jsFOffs/edit#heading=h.ta7ean2l5wfd"><em>The Offense</em></a>).</p><p><strong>New ways of working in the cloud:</strong></p><p>During the production of “<a href="https://aws.amazon.com/blogs/media/the-crown-in-the-cloud/">The Crown</a>” Season 4, the series encountered the daunting task of generating over 600 visual effects (VFX) shots amid the global pandemic constraints.</p><p>The transition to a cloud-based workflow on AWS empowered a relatively small in-house VFX team of 10 artists to work remotely and efficiently, leading to the delivery of the 10-episode series in a mere eight months.</p><p>“Our users expect that when they turn on the Netflix service, that it will just work. And that is partly due to the fact that we’re on AWS Cloud”- <a href="https://aws.amazon.com/solutions/case-studies/netflix-case-study/">Eva Tse</a>, Director of Big Data Platform, Netflix</p><p>As you can see, AWS allowed Netflix to orchestrate their defence and attack strategies:</p><ul><li>Defensively, AWS enabled them to not only maintain their core service offering — allowing users to watch content on-demand — but also to enhance its robustness and scalability.</li><li>Offensively, AWS facilitated much faster experimentation and iteration rates, allowing Netflix to develop new features that increased stickiness and attracted more subscribers to the platform.</li></ul><h3>The Offense: Personalisation, Search and R+D</h3><h3>Personalisation</h3><p>“Personalization is one of the <a href="https://research.netflix.com/business-area/personalization-and-search">pillars</a> of Netflix because it allows each member to have a different view of our content that adapts to their interests and can help expand their interests over time..</p><p>It enables us to not have just one Netflix product but hundreds of millions of products: one for each member <a href="https://research.netflix.com/business-area/personalization-and-search">profile</a>.”</p><p>Netflix uses a number of different machine learning and recommendation algorithms to drive the personalisation and search experiences. These include:</p><ol><li><strong>Collaborative Filtering</strong> for predicting preferences through similarities across its user base, leading to tailored content recommendations.</li><li><strong>Deep Learning </strong>for intricate pattern recognition in user behaviour, refining the precision of personalized recommendations and features like custom artwork.</li><li><strong>Ranking Algorithms</strong> for curating and sequencing content, maximizing personalization by highlighting the most pertinent titles.</li><li><strong>A/B Testing</strong> as a fundamental method for continuously improving and refining personalization and search algorithms by testing new ideas and understanding their impact on user engagement.</li></ol><p>The outcomes of these processes are additional features of the platform such as “Because you watched..”, “Matched for you…” lists of movies / shows. Whilst this is beyond the initial core product offering of Netflix (which again is giving users the ability to watch what they want when they want it), it’s showing users what they might want to watch based on an understanding of their preferences. In other words, it’s solving another huge problem for users, figuring out what to watch.</p><p>By Netflix recommending shows with the highest likelihood of a match, they achieve three things:</p><p><strong>[1] Reduced latency in content discovery: </strong>When users are on the platform they either know what they want to watch or they don’t. By automatically recommending shows directly on the home page, based on what they’ve watched previously (shown as a % match), Netflix can minimise the time between users entering the platform and watching something.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*W6-fQOFrK2bKnRC6" /></figure><p><em>Figure 5 — The Netflix homepage comes with recommended shows that have a certain % Match based on your view history.</em></p><p><strong>[2] Enhancing stickiness to the platform: </strong>When users are presented with films that align with their interests, they are more inclined to stay on the platform as it saves them the effort of searching for similar content elsewhere.</p><p><strong>[3] Optimizing Content Catalogue</strong>: While having a diverse catalogue is important, the ability to recommend content effectively can reduce the need for an “infinitely large catalogue.” By understanding and catering to user preferences, Netflix ensures that even a finite selection of content can meet a wide array of user desires, making the service feel personalized and comprehensive.</p><p>“Personalized recommendations on the Netflix Homepage are based on a user’s viewing habits and the behaviour of similar users. These recommendations, organized for efficient browsing, enable users to discover the next great video to watch and enjoy without additional input or an explicit expression of their intents or <a href="https://arxiv.org/pdf/2105.14134.pdf">goals</a>.”</p><h3>Search</h3><p>Netflix’s search feature is powerful. It uses a combination of natural language processing, text analytics, machine learning and collaborative filtering again to build new connections between catalogue<a href="https://research.netflix.com/business-area/personalization-and-search"> titles</a>.</p><p>Whilst at a fundamental level, the search feature aligns with a more defensive strategy (because it directly serves the core tenet of Netflix — getting users to the shows they want to see), we wanted to mention it here because Netflix deliver value beyond simply this core necessity of search, and it links well with personalisation.</p><p>This idea is captured well here:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*PJvTy-3U7Am-VTjL" /></figure><p><a href="https://arxiv.org/pdf/2105.14134.pdf"><em>Figure 6 </em></a><em>— the evolution of search results at Netflix. (a) showing exact match results, evolving to (d) showing exact match as well as “based on..” results to showcase catalogue + inspire for further viewing.</em></p><p>In this image, (a) shows the results of the show that exists in the Netflix catalogue exactly related to what the user is searching for. This neither increases nor diminishes the value created by Netflix, it is neutral, as it meets user’s expectations (met expectations == neutral value creation).</p><p>The following three images reflect the evolution of their search function, where now you can not only find the thing that you are looking for, but you can find similar shows which are ordered by attributes, e.g. others have also liked, based on a similar type of show, or similar category. This helps users do two things: understand that Netflix has a wide catalogue, and get inspired/explore other shows that they may want to watch, beyond what they searched for (beyond expectations == additional value creation).</p><p>To make the importance of this more obvious we can compare it to one of Netflix’s competitors, Disney Plus.</p><p>If we’re searching for a specific title, e.g. “The Book of Boba Fett” (a Star Wars series), users will typically type in the most convenient keywords e.g. “Boba Fett” in this case. But the top search result is a completely different show, and you have to scroll down to find it. The first row gives you the trailer of <em>The Book of Boba Fett</em> but the real link to the series is located on the second row (8 titles away).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ZXpj9g2ZPw8y53bX" /></figure><p><em>Figure 7 — Screenshot of Disney Plus when searching for “boba fett”, the actual series is 8 titles away.</em></p><p>Even if the user goes a step further and types in “The book of boba”, the top result is a completely different show again, and the user can only find it now 6 titles away.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*H7YHkJMhZ-voZanr" /></figure><p><em>Figure 8 — Screenshot of Disney Plus when searching for “the book of boba”, the actual series is still 6 titles away.</em></p><p>Compare this to the search of Netflix. If we search for <em>Naruto</em>, the first result is <em>Naruto</em>, with the following two results directly related shows, being sequels of <em>Naruto</em> (<em>Naruto Shippuden</em> and <em>Boruto</em>). This is an important difference in the two platforms. When a user is searching for a show, they need to know whether it is in the catalogue or not.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ZgSmb4ZVkOc1RMos" /></figure><p><em>Figure 9 — Screenshot of Netflix, when a user searches Naruto, it is the first title, with related titles next to it.</em></p><p>If a user searches for something on Disney Plus, they cannot immediately know whether it is there or not, they have to keep looking through, compared to Netflix, where if it’s not the top result, they can be pretty confident it is not there. Whilst this difference may seem trivial, these product frustrations accumulate in the user’s mind.</p><p>Let’s take another example. There are no explicit <em>Superman</em> films in either the Disney Plus catalogue nor Netflix, yet the search results are very different and the frustrations are self-explanatory:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*4PFeUJu97yIPt87N" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*YFBhfyTzy7TGpmoL" /></figure><p><em>Figures 10 and 11 — A comparison between searching for “Superman” on the Disney vs Netflix platform. Whilst neither have exact titles, Disney displays shows with titles similar to the keyword searched, whilst Netflix displays shows similar to what the user is actually looking for with being explicit (i.e. superhero movies).</em></p><p>Whilst it may seem tedious to dive into such a niche example, these frustrations matter to the user, and because they matter to the user, they should matter to the business.</p><p>In business 101, we learn about the crucial gap between the value users perceive and the price they pay. While much emphasis is placed on finding ways to enhance this perceived value (marketing etc), the importance of reducing product frustrations is often overlooked.</p><p>These seemingly minor annoyances do accumulate in the users’ minds, diminishing the overall perceived value for users. And if that value is reduced to the level of the price point, or even below, then users will not want to use your platform — something that likely contributes to what they got wrong in the Disney Plus platform:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*N38VEzRXRzMrAY2S" /></figure><p><em>Figure 12 — A simple diagram showing how product frustrations accumulate to diminish the overall perceived value of the product by users. If it diminishes the value enough, users a likely to stop using the platform and switch to another.</em></p><h3>Research and Development</h3><p>It’s also worth paying respect to Netflix’s emphasis on R+D and their cross-institutional collaborations. For example, their paper on the use of variational autoencoders (a type of neural network) to improve collaborative filtering (for personalisation), was done between <a href="https://arxiv.org/pdf/1802.05814.pdf">researchers</a> from Netflix, MIT and Google AI.</p><p>This willingness to partner outside of just the Netflix R+D team, doesn’t just underscore the strategic advantage of leveraging collective expertise across industries and academia, but something deeper. It represents an attitude and culture at Netflix, where instead of being obsessed with trying to <em>increase their slice of the pie </em>(competitor-focused), they are obsessed with <em>increasing the size of the whole pie </em>(user-focused), and in doing so continuously redefine the landscape of digital streaming and entertainment.</p><p>Case #3<br>© All rights reserved 2024 OTSOG Media</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ec291c7b5597" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Monzo: A Product Strategy Case Study]]></title>
            <link>https://medium.com/@pjega92/monzo-a-product-strategy-case-study-94cbffbe4ad3?source=rss-84225be8d71------2</link>
            <guid isPermaLink="false">https://medium.com/p/94cbffbe4ad3</guid>
            <category><![CDATA[product-strategy]]></category>
            <category><![CDATA[fintech]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[case-study]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Jega Pradeeba]]></dc:creator>
            <pubDate>Mon, 22 Jul 2024 05:36:24 GMT</pubDate>
            <atom:updated>2024-07-22T05:53:34.024Z</atom:updated>
            <content:encoded><![CDATA[<h3>Monzo — A Product Strategy Case Study</h3><h3>Contents</h3><ul><li><a href="#ed79">Key takeaways</a></li><li><a href="#695f">Founding of Monzo</a></li><li><a href="#d37b">Technical Product Strategy</a></li><li><a href="#cc4c">Product Marketing Strategy</a></li><li><a href="#17b8">Culture / Leadership</a></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uvO8uSzQLckkFuEYqkuurQ.png" /><figcaption><a href="https://monzo.com/blog/2019/08/30/is-monzo-a-debit-card">Image Source</a></figcaption></figure><h3>Key Takeaways</h3><ul><li><strong>Built for scale </strong>— Monzo’s infrastructure uses a modular, fragmented services infrastructure (microservices approach), utilizing advanced tools like AWS and Kubernetes to ensure agility, resilience, and scalability.</li><li><strong>Designed for virality</strong> — Its intuitive mobile app, eye-catching ‘hot coral’ debit cards, and referral programs have not only attracted users but also encouraged them to share their positive experiences with friends and family, creating a viral effect that has propelled Monzo into the mainstream of modern banking.</li><li><strong>Rooted in transparency and community</strong> — Monzo stands out for its commitment to open communication and engaging with its user base. This approach has helped foster a strong sense of trust and loyalty among its customers, and distancing itself from traditional banking.</li></ul><h3>Introduction</h3><p>The essence of any product is the feeling users get from it. For Monzo and other neobanks, the core tenet of that feeling is trust. That trust needed to take 2 forms though: users needed to trust that Monzo would provide them with the functionalities they would expect from a normal bank (namely if I leave my money here, can I trust you to look after it?), and at the same time, can users trust the startup bank to also represent the complete opposite of what traditional banks represent (frustration, time-wasting, lack of transparency).</p><p>In this case study, we look at how Monzo was able to address these two points, specifically looking at its technical product strategy and product marketing strategy, and how these helped Monzo shift the banking landscape.</p><p>I hope you enjoy the read.</p><h3>The founding of Monzo</h3><p>Monzo, a neobank (digital-only banking platform) based in the UK, has redefined the landscape of banking with its customer-centric approach and innovative technology. Founded in 2015 by Tom Blomfield, Jason Bates, Gary Dolman, Paul Rippon, Jonas Huckestein, it quickly distinguished itself by offering a seamless, intuitive banking experience.</p><p>The founding team had a strong mission driving them of simply<em> making banking better</em>, and they had the experience to help get them there with three of the five being early members of Sterling Bank (another neobank), and Tom Blomfield (CEO of Monzo) having founded another fintech company, GoCardless, incubated at Y Combinator in 2011.</p><p>“Tired of the way traditional banks do business, we believed we could build a real alternative. One that’s totally transparent, treats people fairly, and caters to the way we all actually live our lives.” — Tom <a href="https://monzo.com/blog/2018/02/19/third-birthday">Blomfield</a>, CEO of Monzo.</p><p>Starting initially as a prepaid card combined with a digital app to track expenses, Monzo evolved into an official bank in 2017, and today boasts <a href="https://monzo.com/annual-report/2023/">7.5 million</a> users, making it the UK’s <a href="https://monzo.com/annual-report/2023/">7th largest</a> bank by customer count.</p><h3>Product Strategy</h3><p>Monzo’s success was never down to solely how good its marketing was nor just its technical architecture, but a combination of both sides of the business and the synergy between them.</p><h3>Technical Product Strategy</h3><p>From the beginning, Monzo was building to scale. The company’s strategy involved the deployment of robust cloud-based solutions, agile software architectures, and the ability to generate insights from their data.</p><p>This approach was crucial for two main reasons:</p><ul><li><strong>Banking is a ubiquitous service:</strong> Success in banking requires the capacity to handle large numbers of customers, and vast numbers of transactions per customer, efficiently. In addition, as we will see later, Monzo used referral systems to market the product. For any one user who refers Monzo to another two users, each one of them can refer another 2 users, and soon you get exponential growth. Whilst that’s great from the marketing KPIs, it can be a real headache if your product infrastructure is not set up to handle that kind of exponential growth in demand from users.</li><li><strong>Customer Attraction to Neo-Banks:</strong> The primary appeal of neo-banks like Monzo lies in their ability to streamline and simplify banking processes, thereby eliminating common frustrations. If Monzo had encountered operational issues due to a high number of users, it would have contradicted this value proposition, deterring customers and diminishing their trust and satisfaction. Once trust is lost in products, particularly financial products, it’s very difficult to get back. So in other words, Monzo had no room for error.</li></ul><p>Bottomline, this foresight of building to scale was necessary to prevent operational bottlenecks that could erode customer trust, a key factor for any financial service.</p><h3>Amazon Web Services</h3><p><em>“By using AWS, we can run a bank with more than 4 million customers with just eight people on our infrastructure and reliability team.” — </em><a href="https://aws.amazon.com/solutions/case-studies/monzo/">Matt Heath</a>, Distributed Systems Engineer, Monzo</p><p>Amazon Web Services (AWS) is a comprehensive cloud computing platform, providing a wide range of services like computing power, database storage, and content delivery. It’s known for its scalability, reliability, and flexibility, making it a popular choice for businesses of all sizes. By leveraging AWS’s cloud infrastructure, Monzo could manage its banking services with a lean team, overseeing operations for millions of customers. There were 2 main angles for facilitating this:</p><p>First, regulatory compliance, as AWS’s reputation for meeting the Financial Conduct Authority’s <a href="https://aws.amazon.com/solutions/case-studies/monzo/">(FCA) guidelines</a>, such as through AWS CloudTrail, made it a trustworthy choice for hosting FCA-regulated workloads.</p><p>Second, Monzo adopted a technical strategy centered on a microservices and containerized architecture, leveraging the scalability and agility afforded by this design (i.e. a flexible system made of small, independent components that expand and contract with demand). By harnessing the power of <a href="https://aws.amazon.com/solutions/case-studies/monzo/">AWS’s cloud environment</a> along with Kubernetes for streamlined container management, Monzo was able to efficiently operate a multitude of <a href="https://monzo.com/blog/2020/10/19/autoscaling-monzo">microservices</a>, enhancing both performance and reliability.</p><p>The use of this architecture is like constructing with many smaller Lego bricks as opposed to one big “monolithic” brick. Each microservice is a separate ‘Lego brick’, small and manageable on its own. When combined, they form a larger, more complex structure. This modular approach enabled Monzo to efficiently build and adjust its digital banking services, piece by piece, to create a cohesive and flexible system, and adapt as it learned about user behavior and needs.</p><p>This infrastructure facilitated Monzo’s ability to provide “real-time” banking services, a stark contrast to the traditional banks’ slower processes, and supported Monzo’s exponential growth in services and customer base. The microservices architecture also enabled rapid deployment of new features and quick adaptation to changing customer needs, underlining the importance of flexibility in product development.</p><p>[A more detailed journey of Monzo’s backend architecture which can be found <a href="https://monzo.com/blog/2016/09/19/building-a-modern-bank-backend">here</a>].</p><h3>Google BigQuery</h3><p>“As a tech team we value simplicity, and with Google BigQuery we’ve created the simplest and most scalable setup we could imagine.For four or five years, banking conferences have discussed the need to bring all data together in one place for analysis.</p><p>With Google BigQuery, we’re actually doing it.” — <a href="https://cloud.google.com/customers/monzo">Dimitri Masin</a>, Head of Data at Monzo</p><p>One potential drawback of a microservices architecture, is that it can complicate analytics compared to a monolithic architecture — i.e. because data is fragmented across independently operating services, it can be more difficult to gather data and draw insights.</p><p>Google BigQuery is a fully managed, serverless data warehouse and analytics platform that allows organizations to analyze large datasets in real time. Monzo’s utilization of Google BigQuery enabled them to consolidate their data, ensuring that all teams had access to the same up-to-date information. This unified source of truth not only simplified data access for new team members but also facilitated more efficient and scalable analytics, even as Monzo’s user base continued to grow.</p><p>Furthermore, using BigQuery also reduced the gap between engineering, data, and business, as it enables non-technical staff to self-serve <a href="https://cloud.google.com/customers/monzo">85%</a> of their business intelligence queries without consulting the data team:</p><p>“We do all of our analysis on the fly because Google BigQuery can execute such gigantic joins of tables at speed. That’s an incredible advantage. We define and analyze segments as we think of them, instead of creating an ETL (extract, transform, load — a more traditional/time-consuming process to gather insights from data) process and realizing the next day that we want something else”, Dimitri Masin (Head of Data).</p><p>For example, Monzo’s data team set up dashboards via BigQuery for customer support, showing trending issues users had. They were able to identify the most frequently recurring problems by looking at customer behaviour on the app just before the put in the support request. This granular insight on user behaviour and where things were going wrong, allows them to reduce the number of support requests they received by <a href="https://cloud.google.com/customers/monzo">50%</a> over 10 months. Again, the impact this alone would have on the trust and confidence users have with Monzo is significant.</p><p>What’s interesting, is not just taking this strategy from a data and business-centric view, but also purely financially. “In other companies of our size you need a data engineering team of at least two to four people constantly on-hand to maintain and run day-to-day analytics infrastructure,” says Dimitri. “Google <a href="https://cloud.google.com/customers/monzo">BigQuery</a> doesn’t need a dedicated team to maintain it. In the two and a half years since we set up the solution, it’s been so robust and so scalable that it’s required no maintenance work whatsoever.”</p><p>The bottom line is that Monzo leveraged Google’s BigQuery to overcome the drawbacks of a distributed microservices backend architecture, giving them a single source of truth that even non-technical staff could leverage.</p><p>Opinion:</p><ul><li>One thing you may have noticed is that by leveraging services like AWS and BigQuery, the Monzo team was able to maximize their technical capabilities whilst minimizing the size of the team, which is often a startup’s biggest expense.</li><li>Despite Monzo doing this back in 2016/2017, as such services expand and improve, it’s likely this lean set-up will be how most startups (and even more established companies) operate.</li></ul><h3>Other Points</h3><p>Another notable mention is their commitment to a mobile-first approach. In contrast to many traditional banks, which transitioned from non-digital to digital services optimized for web/desktop computers before eventually focusing on mobile, Monzo prioritized mobile usability from the outset. This approach reflects a forward-thinking perspective on banking services, acknowledging the growing importance of mobile devices in modern banking. It’s worth noting that even today, several major banks struggle to offer a seamless mobile experience.</p><p>Opinion:</p><ul><li>I think Monzo understood that when it comes to banking, users just want to get things done. And I think that’s the sweet spot for mobile services in general — mobile apps are primarily for task-orientated activities (outside of social media that is). This is in contrast to more immersive activities like media consumption, in-depth research, or leisurely shopping, which often suit desktop environments better — e.g. it is why Amazon did not need to be mobile first. If traditional banks had adopted a mobile-first strategy earlier, they likely would have given neobanks like Monzo a better run for their money.</li><li>I believe this also highlights the importance of building a flexible infrastructure from the ground up. Moving from non-digital to web banking to mobile-first, traditional banks were always on the back foot. But how we as consumers interact digitally is always changing, beyond mobile we’re already seeing the application of VR and AI pins. It’s easy to say banks should have been mobile-first from the beginning, but it is difficult to anticipate trends — so I don’t think we should. Instead, I think it is about building a flexible infrastructure that can adapt to new use cases.</li></ul><h3>Product Marketing Strategy</h3><h3>User Experience</h3><p>While UX/UI design is typically considered an integral component of technical product strategy, its role in Monzo’s case goes beyond functionality. It plays a critical role in shaping the brand’s identity and molding customer perception. Monzo’s user-friendly design isn’t just about ease of use; it serves as a psychological bridge connecting the simplicity of banking tasks with the overall user experience. This aspect of design has been instrumental in establishing a positive and enduring impression of the brand within its user base.</p><p>A great example is the onboarding process:</p><p>“Our vision is to let new customers create a Monzo account in minutes — anytime, anywhere. We want our customers to start the process as they enter a queue at their local coffee shop and have an account created by the time their coffee is served.” — Head of Marketing and Community at <a href="https://www.jumio.com/app/uploads/2021/11/jumio-monzo-case-study.pdf">Monzo</a></p><p>It’s worth remembering that the whole point of Monzo, and other neobanks, was to offer a new alternative to the frustrating processes of traditional banks, and so by that very premise, every activity done on Monzo needed to be easy, painless and frustration-free. During the setup process, Monzo faced high abandonment rates, particularly during online identity verification.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/1*_lLoHK1wax9pYuP1m-7-jg.png" /><figcaption><a href="https://goodux.appcues.com/blog/monzos-fast-friendly-finance">Image source</a></figcaption></figure><p>To address this, they partnered with Jumio, a software identity verification service. One significant insight during testing revealed that many new users abandoned the process when asked to take a selfie. Monzo recognized that users might not be in a suitable environment for taking a verification selfie, such as being at work or in a public place. To mitigate this, they implemented a strategy of sending reminder emails over the following days to encourage users to complete the process. This multi-touch approach, akin to the way Duolingo sends reminders to stay on track, significantly improved new account conversions.</p><p>Furthermore, Monzo’s decision to leverage Jumio’s backend infrastructure for identity verification, rather than building it in-house, resulted in a significantly faster verification process. Approximately 95% of accounts were verified within just 270 seconds. This accomplishment was crucial in achieving Monzo’s goal of delivering a swift and effortless customer experience, especially when compared to the potentially lengthier verification processes offered by traditional banks.</p><h3>Marketing Campaigns</h3><p><strong>The Golden Ticket</strong></p><p>One of Monzo’s more famous marketing strategies was the golden ticket — a referral program to boost customer engagement and loyalty.</p><p>After launching their beta version in 2017, Monzo gave out “<a href="https://www.extole.com/blog/how-monzo-crowdfunded-a-bank-using-relationship-marketing/">golden tickets</a>” to existing customers. These customers could then give these tickets to their friends, who could skip the waiting list and get immediate access to the Monzo <a href="https://monzo.com/blog/2018/01/10/golden-tickets-are-back">app</a>. The idea was influenced by the gamification and waitlist-style referral system from Robinhood, a popular online trading platform. This method proved highly effective in building customer loyalty, as studies show referred customers are <a href="https://www.extole.com/blog/referral-marketing-customers-significantly-more-valuable-loyal-ama-study/">18%</a> more loyal than those acquired through other means (plus referred customers are <a href="https://www.extole.com/blog/referral-marketing-customers-significantly-more-valuable-loyal-ama-study/">25%</a> more profitable per year than un-referred customers).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/416/0*Svblc8LtZPGHjRAt" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*FPFXZvWYkZscOG_n" /><figcaption><a href="https://hashir.blog/2021/02/invite-only-apps/">Image source</a></figcaption></figure><p><em>A screenshot of the waiting list to get the Monzo bank card when they were launching their beta in 2017/2018, and to the right a screenshot of the ‘golden ticket’ you could send to your friends to skip ahead in the queue.</em></p><p>The campaign was hugely successful with <a href="https://tomblomfield.com/post/691384431502557184/monzo-growth">40%</a> of its signups in 2017 coming from the golden tickets alone, helping them push from 70k users in 2016 to 600k by the end of <a href="https://tomblomfield.com/post/691384431502557184/monzo-growth">2017</a>. I think it is easy to stop here and believe that this cleaver marketing strategy was just that, clever marketing, but there are a few more nuances that are worth mentioning.</p><p>First, timing. The Monzo team did not push this campaign in 2016, when they were still perfecting the product. The campaign increased their number of users nearly <a href="https://tomblomfield.com/post/691384431502557184/monzo-growth">ninefold</a>, if they had done this any earlier when the product was not perfected, they would have dug their own grave. So you have to give the leadership team the props for actually having the patience to push only when they were confident and ready — something a lot of other startups fail to do.</p><p>Secondly, trust. As previously touched on, trust is the core emotion for financial products, particularly for neobanks, who have to deal with the cognitive bias that e.g. only traditional banks are safe. But this is particularly important for referrals. It’s one thing for you to take a chance, and leave your money with a new startup and things go wrong, it is a completely different thing if you told your friend or family member to leave their money in a startup, and things go wrong.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*FirgEhrfk4lMBVqV" /><figcaption><a href="https://monzo.com/">Image source</a></figcaption></figure><p>So the success of the golden ticket campaign was not just a testament to the clever marketing strategy itself, but it was another nod to just how good the product was, because people were telling their loved ones to join, which meant they had strong confidence in the safety and efficiency of the Monzo product.</p><p>Opinion:</p><ul><li>I genuinely believe that this idea of gradually building consumer confidence is such a core tenet of their success: build product → test on a few thousand customers → perfect product → viral marketing campaigns.</li><li>It’s also worth noting that Monzo never came out of the gates being a fully-fledged bank. They began as a prepaid card, users could see the benefits of digitalising their transaction history in a user-friendly way, without forking over all of their money, and allowed Monzo themselves to collect data, analyse, learn, plan and build upwards. They eased consumers into using their product as a bank, gradually building up users’ confidence in them.</li></ul><h3>Design</h3><p>Beyond just making the interface easier to understand and navigate, Monzo is well known for its creative design thinking. One of the best examples of this is Monzo’s famous coral-pink cards.</p><p>From the beginning, the colour was chosen as a marketing strategy: “We mostly wanted it so when we went to a restaurant or something, people would ask about the card,” head of design Hugo <a href="https://www.wired.co.uk/article/monzo-card-design">Cornejo</a>. “Then when we realised people really liked it, we kept it.” As a user of Monzo myself, I have to admit, I only came to know about it after seeing my friend’s bright pink card, and asking about it.</p><p>“In order for any product that’s customer-facing, design must happen. You might do it with care, you might do it as an afterthought. In our case, we know of the importance of it because of the hot coral card — people really like it, but it doesn’t happen by accident. It was a lucky strike, but there’s some reasoning and design behind it.”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*K2DBnHRQxc0WBzwX" /><figcaption><a href="https://www.cnbc.com/2024/06/03/digital-bank-monzo-posts-first-full-year-of-profit-after-more-than-doubling-revenue.html">Image source</a></figcaption></figure><p>Throughout the app, you can tell the Monzo team has put a lot of thought behind their design — making it seem clean, modern and approachable. Given that their customer base is mostly millennials and Gen Z, who are used to and largely appreciate design, it definitely does not go unnoticed, and further cements the idea that Monzo is so diametrically different from traditional banks.</p><h3>Transparency as part of its brand</h3><p>Despite its success, it wasn’t always smooth sailing for Monzo, and yet even when things did go wrong, they maintained their core value of transparency.</p><p>For example, during a sudden and unexpected outage (due to some changes to their backend service), Monzo did not hesitate to warn their customers not to use that app and in fact to “carry another card”.</p><p>“Because we jumped straight on it and put an immediate message out saying ‘please carry another card out, we’re having an outage and we’re looking into it and will update you immediately’, the PR that came back from it was really strong… From the depths of despair hope blossomed — people said this is a really good firm to be with.” - Gary <a href="https://www.matterofform.com/news/articles/monzo-gary-dolman">Dolman</a>, the CFO of Monzo</p><p>Given the deteriorating trust consumers had in traditional banks since the 2008 financial crisis, which was epitomized by a lack of transparency, it was never going to be enough for neobanks to win by just having a more convenient user experience. Monzo knew this from the start, and their efforts in being an ‘emotionally’ superior product by simply being honest and transparent won consumers over.</p><p>Opinion:</p><ul><li>It may also be worth noting that the types of consumers a startup neobank attracts as its initial customer base, is always going to be those who are willing to accept some degree of risk/understanding that there will be teething problems.</li><li>This means that the presence of these issues was never going to make users turn away (although obviously, this depended on frequency and severity of issues), but the point is that the initial customer base was looking for something — a bank they could trust, that would be honest with them, and that represented the opposite of what traditional banks stood for.</li><li>This is a core part of their product marketing that deserves nothing but praise and serves as a great learning point for other startups.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*xISsJv-iVn5WSbcY" /><figcaption><a href="https://monzo.com/blog/2019/09/08/why-monzo-wasnt-working-on-july-29th">Image source</a></figcaption></figure><h3>Culture and Leadership</h3><p>The transparency and honesty that Monzo showed its customers, was not just a marketing strategy, but it was part of their culture. Culture is always top-down, and I think it’s fair to say that the senior leadership had a strong emphasis on engraining their values into the roots of the company.</p><p>For example, Dolman, Monzo’s CFO, attributes their success to the team’s fearless approach to challenges — characterized by the <em>five finger push, </em>where not all attempts may succeed, but it’s the one that breaks through that leads to significant achievements:</p><p><em>“</em><a href="https://www.matterofform.com/news/articles/monzo-gary-dolman"><em>Five finger push</em></a><em> — four of those fingers won’t break through, but the fifth will and it will break through to something big. It’s not being scared to fail in certain aspects.”</em></p><p>This philosophy encouraged not fearing failure in certain aspects, recognising it as an integral part of the journey to success. In its quest to create the “best current account in the world,”</p><p>Monzo showcased a blend of ambitious goals and technological advancement, while maintaining a unique focus and ignoring its competitors. Dolman emphasizes the importance of concentrating on their own strengths and innovations, rather than being preoccupied with what others in the field are <a href="https://www.matterofform.com/news/articles/monzo-gary-dolman">doing</a> — (the old adage of <em>competition is for losers [Peter Thiel]</em>). This strategy allows Monzo to stay ahead, with a firm belief in their ability to outperform and outpace the competition.</p><p><em>“You worry about your own game, not what the opposition are doing and let them catch up with you — not the other way round. You are constantly saying ‘these are our ideas; this is what we want to do’, and you have the courage to publish it.</em></p><p>What you are banking on is the fact that you are going to be able to do things better and faster than the opposition.” — Gary <a href="https://www.matterofform.com/news/articles/monzo-gary-dolman">Dolman</a>, the CFO of Monzo.</p><p>These principles are not only pivotal to Monzo’s success but also serve as valuable lessons for other sectors, demonstrating the power of positive leadership and a strong, cohesive company culture.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=94cbffbe4ad3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Midjourney — A Product Strategy Case Study]]></title>
            <link>https://medium.com/@pjega92/midjourney-a-product-strategy-case-study-fd3a097e6c7b?source=rss-84225be8d71------2</link>
            <guid isPermaLink="false">https://medium.com/p/fd3a097e6c7b</guid>
            <category><![CDATA[case-study]]></category>
            <category><![CDATA[generative-ai-tools]]></category>
            <category><![CDATA[product-strategy]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[product-management]]></category>
            <dc:creator><![CDATA[Jega Pradeeba]]></dc:creator>
            <pubDate>Sat, 20 Jul 2024 02:26:53 GMT</pubDate>
            <atom:updated>2024-07-26T02:32:49.796Z</atom:updated>
            <content:encoded><![CDATA[<h3>Midjourney — A Product Strategy Case Study</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TFi-JFZl3pOGcGrvwqQk3g.png" /><figcaption><em>Source: </em><a href="https://legacy.midjourney.com/showcase/recent/"><em>Midjourney showcase</em></a></figcaption></figure><p><strong>Intro</strong></p><p>Midjourney is an AI-driven platform that specialises in generating images from natural language inputs. It was built by an independent research lab with the same name, consisting of only 11 full-time staff to date, yet generating over $200m ARR and 16 million users — having only started less than 2 years ago.</p><p>In this case study, we will examine the key factors contributing to its success, focusing on technical product and product marketing strategies, what lies ahead for the company, and some more high-level points for consideration.</p><p>I hope you enjoy the read.</p><p><strong>The Beginnings of Midjourney: Foundation and Development<br>Early Days: The Formation by David Holz</strong><br>Midjourney was established in 2022 by David Holz. Before that, Holz’s background includes studies in physics and math, followed by a pursuit of a PhD in applied math, a period during which he also worked at NASA and the Max Planck Institute. In 2011, he relocated to San Francisco to start Leap Motion.</p><p>Leap Motion, known for its precise hand-tracking technology similar to a 3D mouse, gained significant traction with over 300,000 developers using the technology, attracted funding from major VCs — an experience that would later influence Midjourney — and eventually was acquired by Ultrahaptics for $30M in 2019. [<a href="https://aituts.com/david-holz/">1</a>]</p><p><strong>Transition to Midjourney:</strong><br>After Leap Motion, Holz found himself contemplating the future and what people would need in an uncertain world. He identified three core pillars he believed to be essential: reflection, imagination, and coordination. This contemplation and his personal philosophy significantly influenced the conceptualization of Midjourney. [<a href="https://aituts.com/david-holz/">1</a>]</p><p><strong>Origin of the name:</strong><br>Holz attributes the concept of Midjourney to Daoist influence, specifically Zhuangzi. On the name itself though, Holz says he feels like we are actually mid-journey — that “we come from a rich and beautiful past, but ahead is this wild, unimaginable, unfathomable future”. [<a href="https://www.reddit.com/r/MidJourneyUpscaler/comments/17i9a8y/what_is_the_meaning_of_midjourney_founder_david/">2</a>]</p><p><strong>The Vision is bigger than you think:</strong><br>Holz envisioned Midjourney not just as a technology product but as a platform for creativity and expression. He focused on creating an AI-powered platform that would transform text prompts into visual imagery, enabling users to explore new realms of creativity and imagination. This vision was rooted in his belief in the power of AI to expand human imagination and capability.</p><p>“We don’t think it’s really about art or making deepfakes, but — how do we expand the imaginative powers of the human species?” [<a href="https://www.theverge.com/2022/8/2/23287173/ai-image-generation-art-midjourney-multiverse-interview-david-holz">3</a>]</p><p>Right now, there is a lot of fear-mongering around AI, but Holz sees things differently, he likens AI to an engine — engines are without feeling, without motivation, and without direction — it is only the human’s application of the engine that those things are derived.</p><p>An engine is a tool that is to be wielded by humans, to get us from A to B much more efficiently, and in doing so, take us on a journey that opens our minds to new ways of thinking and looking at the world. Midjourney was designed to help influence new frontiers of imagination, to be a creative partner, not a replacement.</p><p>In an interview with The Verge, Holz goes on to compare the discovery of these AI models to the discovery of water. In the sense that they both represent fundamental elements that can be harnessed for transformative purposes. Just as water can be both a peril and a boon to humanity — capable of causing harm but also essential for life and progress — AI too holds dual potentials. The point is not to be afraid of its potential, but to understand how we can build tools to harness its potential to better our lives:</p><p>- “How do we teach people to swim? How do we make boats? How do we dam it up? How do we go from people who are scared of drowning to kids in the future who are surfing the wave? We’re making surfboards rather than making water. And I think there’s something profound about that.”</p><p>Opinion:<br>I don’t think the importance of the vision of a company gets as much emphasis as it should. Perhaps we forget that humans are driven by emotion, and when you develop a moonshot, something that is a little bit abstract, but is exciting and adds value to the world — supercharging human imagination in this case — that is what galvanizes the team around you and what attracts the best talent to you.</p><p><strong>The Ethos</strong><br>The ethos of Midjourney, as encapsulated in the quote, “It’s just about having a home for the next 10 years to work on cool projects that matter — hopefully not just to me but to the world — and to have fun,” reveals a company culture deeply committed to passion-driven innovation and the pursuit of projects with global impact.[<a href="https://www.theverge.com/2022/8/2/23287173/ai-image-generation-art-midjourney-multiverse-interview-david-holz">3</a>]</p><p>This eleven-person team operates without external capital, a unique position that liberates them from the typical financial pressures and constraints often faced by startups. This independence is key to understanding their approach. The absence of external financial motivations and external forces trying to pull the strings, allows for a purer focus on crafting a product that genuinely connects with users, fostering a deeper level of engagement and satisfaction.</p><p>Bottomline is Holz has managed to protect the core of the business and his team from being pulled in the wrong direction, optimising for a product that users love — nothing more, nothing less.</p><p>Opinion:<br>I think this is an important lesson for early-stage founders. Whenever we read about startups in the media, it&#39;s usually about how much money the startup raised and therefore how much it is valued at. So in our minds we associate the amount of money you can raise to the amount of value a startup creates.</p><p>I.e. we assume: value a startup creates == valuation of the company == f(amount of money raised from VCs)<br>I think founders, and perhaps even VCs, really need to disconnect from this mental model, and realise that the true value of a company, is a function of the value created for each individual user of your product. And when you optimise for this, as Midjourney clearly do, the financial rewards will come — they were always a lagging indicator anyway.</p><p><strong>Product Strategy:</strong><br>It&#39;s worth looking at their product strategy from two angles: the technical product strategy, i.e. more focused on how they were building and optimising their model, and the product marketing strategy, more focused on how they positioned and their product and engaged with users.</p><p><strong>An intro to the product:<br></strong>Midjourney, like other generative AI platforms, operates on a foundation of advanced machine learning techniques, primarily using diffusion models. The core principle behind these models is to start with a sample image, incrementally add noise, and then train the model to reverse this process, effectively learning to generate new images that are similar to the original. This process allows for the creation of unique, yet familiar, visual content.</p><p>The training of these models is a data-intensive process. Platforms like Midjourney typically gather vast amounts of data from the internet, which includes scraping images and text. For instance, Midjourney has utilized open-source training models and extensive datasets, such as the 2 billion image-text pairs from the English subset of CLIP’s open dataset, created by the German non-profit LAION. This approach of aggregating and processing massive datasets enables these AI models to learn and replicate a wide range of styles and content, making them versatile tools for various creative applications. By continuously analyzing user interactions and preferences, platforms like Midjourney further refine their models, ensuring that the generated content resonates with user expectations and emerging trends.</p><p><strong>Technical Product Strategy</strong></p><p><strong>Offense — Default style of MJ:</strong></p><p>Midjourney’s default style is intentionally designed to be more artistic and interpretative than the specific input provided by the user, serving a distinct purpose in its product strategy. According to Holz, the rationale behind this approach is that users often don’t know exactly what they want when generating images. If precise replication were the goal, one could simply use Google Images. However, platforms like Midjourney aim to elevate human imagination, necessitating a more creative and proactive role in the image generation process.</p><p>The essence of Midjourney’s approach is to avoid the mundane. For instance, a simple prompt like “dog” could yield a straightforward photo, but that lacks creativity and purpose in the context of AI-driven artistry. Instead, Midjourney aspires to produce works that are not just responses to prompts but artistic interpretations. This is evident in the model’s inclination towards whimsical, abstract, and somewhat peculiar outputs, often blending elements in unexpected yet aesthetically pleasing ways. Holz notes that the model has preferences, such as favouring certain colours and styles, which contributes to its unique artistic identity.</p><p>The style is often described as “a bit whimsical and abstract and weird,” noting that it blends elements in unexpected and aesthetically pleasing ways, often favouring certain colours and themes. This distinctive style is a deliberate choice, ensuring that the output is more than just an answer to a query — it’s a creative journey, offering users a blend of beauty, surprise, and artistic flair. [<a href="https://80.lv/articles/midjourney-founder-talked-about-its-art-style-copyright-what-ai-is/">4</a>]</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/482/1*6jAFKYSQ6JhCL7U_2RkmMA.png" /></figure><p>Source: <a href="https://dev-to-uploads.s3.amazonaws.com/uploads/articles/fwx57bnug0hnu4j64ns6.png">Miss Journey</a> — a default face the model has a tendency to draw</p><p><strong>Defense — On restrictions<br></strong>What’s important about Midjourney’s product strategy, is not just what it allows users to do, but also what it does not allow users to do.</p><p>Another important part of Midjourney’s product strategy is not only the creative possibilities of what users can do, but also the limitations it places on what users cannot do. Considering the boundless scope of user imagination, generative AI platforms have sometimes been used to create content that is graphic and violent.</p><p>One way Holz and his team found a way to deal with this, was to inject accountability of the images created by putting that user’s name on the images created — “When you put someone’s name on all the pictures they make, they’re much more regimented in how they use it. That helps a lot.” Essentially, by promoting transparency over who made it, it almost created a self-policing mechanism.</p><p>On top of this, they added in more robust guard rails, from moderators to the team actively intervening and banning specific words such as “ultragore and everything within a mile of that”</p><p>Opinion:<br>I think the point here is that yes there’s this moonshot vision that’s clear to the team and the users, but Midjourney also set up guardrails to ensure that no one veers off from the track, and ends up morphing the platform into something the team did not want it to be.</p><p><strong>On artists:<br></strong>The issue of copyright within platforms like Midjourney and ChatGPT are still in open discussion right now, so I won’t delve into this too much.</p><p>But it is worth noting that the fact that Holz and his team have been aware of addressing the concerns of the artistic community from the start, has likely also played a role in their success. This is not just to do with anticipating liability issues, but for a product that is built on a community, the community needs to maintain trust in the product, and by actively engaging with their artists it is no wonder why around 4 million of its users are working professional artists.</p><p>“We do have a lot of artists in the community, and I’d say they’re universally positive about the tool, and they think it’s gonna make them much more productive and improve their lives a lot. And we are constantly talking to them and asking, “Are you okay? Do you feel good about this?” We also do these office hours where I’ll sit on voice for four hours with like 1,000 people and just answer questions.” [<a href="https://www.theverge.com/2022/8/2/23287173/ai-image-generation-art-midjourney-multiverse-interview-david-holz">3</a>]</p><p><strong>Product Marketing Strategy</strong></p><p><strong>Partnership with Discord<br></strong>Their partnership with Discord has more to it than at first glance.<br>Firstly, by not being on a simple-to-access website, Discord in a way, acted as the sandbags, holding back the flood of users that inevitably come with virality — a problem that OpenAI faced. In fact, it allowed the team to continue to engage with the community they did have, and perfect their product, before opening the floodgates — which they are now doing.</p><p>Secondly, even in its very early stages, Midjourney still had to support hundreds of thousands, and soon millions, of users trying to access their model. By piggybacking off of Discord’s infrastructure, to handle the traffic, Midjourney was able to keep its head well above the water, and is also a contributing factor to why the team could stay as small as it has.</p><p>Third is engagement. It’s not simply just that Discord already had a large and active community, but also how the engagement specifically ended up influencing their output. The idea is essentially a “round-robin” story — where one person starts, and then another person adds to it, and then another, and another, and by the end, you create something that no one individual could have dreamed of.</p><p>Holz explained it pretty well in his interview with the Verge:<br>“We found very quickly that most people don’t know what they want. You say: “Here’s a machine you can imagine anything with it — what do you want?” And they go: “dog.” And you go “really?” and they go “pink dog.” So you give them a picture of a dog, and they go “okay” and then go do something else.</p><p>Whereas if you put them in a group, they’ll go “dog” and someone else will go “space dog” and someone else will go “Aztec space dog,” and then all of a sudden, people understand the possibilities, and you’re creating this augmented imagination — an environment where people can learn and play with this new capacity.”</p><p>Furthermore, because it is so community-driven, you automatically end up with art that is fun, diverse and completely original. And because you’re showcasing the generated images back into the community, there is a much higher chance of things going viral — and when they do, it reinforces interest in the Midjourney platform — case and point the Pope in a puffer jacket image that broke the internet, and led to even more mass interest in Midjourney.</p><p>Opinion:<br>I think this point here opens up a whole new can of worms. Because yes it AI can augment human output, but humans adjusting the output from another human’s AI output, creates a flywheel that is way beyond what any one person, or even isolated interaction with AI can achieve. And honestly, its this that is the future of AI — human + AI + more human inputs creating a flywheel of innovation.</p><p><strong>Sandbox-and-watch strategy</strong><br>Perhaps one of their main product strategy could be described as sandbox-and-watch. By this I mean, they created the playground of what could be done, put in some general guidelines of what should not be done, and then just watch the community take the product in its different directions.</p><p>Whilst this has led to Midjourney being used to create some incredibly fun art (an anticipated outcome), as well as using it for bad (also anticipated), some users have been using it for art therapy — where people create images of their loved ones who recently passed away — definitely not anticipated by the team.</p><p>And it’s not just a few users, around 20% of all users on Midjourney use it for art therapy. In fact the man behind the viral <em>Pope in a puffer jacket</em> initially started using the platform to create images of his brother who passed away. [<a href="https://www.buzzfeednews.com/article/chrisstokelwalker/pope-puffy-jacket-ai-midjourney-image-creator-interview">5</a>]</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/746/1*wOBGEhkpdH8EPKKGl8kBEw.png" /></figure><p>Opinion:<br>I think there is something to this strategy — because by simply providing the sandbox, you are stepping away from any confirmation bias you/team has (because when you build something, you just assume it will be used in the way you’re anticipating), and instead, you open the door to diverse and unforeseen user innovations. This not only challenges your initial assumptions but also enriches the product’s evolution, driven by actual user creativity and need.</p><p><strong>Business model<br></strong>A few brief points worth noting about the business: From a top-line perspective, Midjourney are currently doing around $200m ARR, and for a team of only 11, that’s pretty impressive. It has around 16million users, with 30% of their users being professionals, which likely includes industries like graphic design, marketing, and perhaps even entertainment. [<a href="https://www.theverge.com/2022/8/2/23287173/ai-image-generation-art-midjourney-multiverse-interview-david-holz">3</a>] How exactly the outcomes of the legal proceedings around generative AI will impact Midjourney, and this segment of its customers, is still yet to be seen.</p><p>On the cost side, the expense of training image models is significant, around $50,000 per training session, especially considering that multiple iterations are often necessary to achieve the desired accuracy and quality of the models. This iterative process, which might require “three tries or 10 tries or 20 tries”, implies a considerable investment in research and development. “It is expensive. It’s more than what most universities could spend, but it’s not so expensive that you need a billion dollars or a supercomputer.” They are also running on $20,000 servers, which they rent. The point here is that the compute power is enormous for simply just generating the images, in the order of thousands of trillions of operations (petaops) per second — “there has never been a service before where a regular person is using this much compute”. Nonetheless, Holz anticipates costs will drop as competition increases and investors plough in more money.</p><p><strong>Opportunities</strong><br>This month, Jan 2024, Midjourney have released their platform onto their own web service in an effort to increase accessibility and stay competitive. Though it’s worth noting, they are still sticking to the strategy of holding back the flood until they are ready as the website will initially be available only to people who’ve racked up more than 10,000 images via Midjourney on Discord — allowing them to test and refine the platform’s experience.</p><p>The next step beyond image generation is obviously video generation. Many platforms are already making notable strides in video generation technology. As this field evolves, the platform that excels in producing high-quality video content is poised to gain a substantial competitive edge.</p><p>The ability to generate videos effectively and innovatively could become a crucial determinant in leading the market. This suggests that the future of AI in visual media might very well hinge on mastering video generation. It’s likely that the increased revenues from greater accessibility via the web platform, will help fund the training of the models for video enhancement.</p><p><strong>Threats: Competition:<br></strong>There are a number of similar platforms to Midjourney, from OpenAI’s DALLE to Stable Diffusion. The graph above shows that pre-V5 release, the gap between these three platforms has been very tight, and it’s likely that this will continue to be an arms race between the top players.</p><p>What’s important here, is that whilst objectively they all seem to do the same thing, generate new, creative images, they still do it in very different ways. The graph below is from a quantitative study analysing the performance of the 3 different models against real images. FID (Fréchet Inception Distance) is simply a way to evaluate the quality of images generated by models — where the lower the FID score, the more realistic the image.</p><p>As you can see, Stable Diffusion is much better can generating hyper-realistic images, but the importance of this completely depends on who’s using it and for what. Midjourney has a core artistic community, likely because the Midjourney is not so hyper-realistic, with its default style, it provides a better use case for continuing to “expand the imaginative powers of the human species”. [<a href="https://arxiv.org/pdf/2210.00586.pdf">6</a>]</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*II8695R2FrrJ94svdUl51g.png" /></figure><p>Case #1 — initially written in Jan 2024<br>© All rights reserved 2024 OTSOG Media</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fd3a097e6c7b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>