In mobile app development, team collaboration and understanding data correctly are bigger hurdles than one might think. These issues have a direct knock-on effect on user retention, a key metric in an industry where let’s face it, 74% of users drop the app after just a day.
This article digs into the nitty-gritty of keeping users engaged in the long haul. Only about 4% stick around four weeks after installation, and we need to talk about why.
I’ll be sharing insights and stories from my experience but also from those who’ve been there, and done that – shedding light on what works and what doesn’t. You will find a mix of professional experiences and tried-and-tested strategies, highlighting everything from the role of in-app purchases to how platform choices and personalization can make a real difference in keeping users around.
We will be unpacking real-life examples and expert takes, and I hope this article will become your go-to guide for better collaboration, smarter use of data, and ultimately, improving those app retention rates. So, let’s dive in and see what we can learn and apply to your projects.
First, let’s look at some data
In-app purchases significantly boost retention by 33% compared to non-IAP apps. Platform and industry differences are notable, with iOS apps outperforming Android by 52% in retention rates and the gaming sector leading at a 36% retention rate, outshining social networking apps.
Personalization proves to be a game-changer, offering a 50% better retention rate when onboarding is tailored.
BUT. To get to do personalization, CRO, experimentation, whatever we are calling it these days, there is a very big need for proper collaboration to happen, to ensure that product teams behind mobile apps, SaaS, and any type of products see the value in using data.
Not too long ago, I was brought on board to help a retailer that had a challenge of understanding user interactions with their app and its features and whether their innovation efforts were translating into customer loyalty and satisfaction.
The key to unlocking this impasse lies in how data, particularly qualitative insights, are utilized and communicated across teams. Being in product for a long time, I’ve learned that the key isn’t just to bulldoze through these complexities. Instead, it’s about striking a delicate balance, knowing when to leverage data and when to step back, and respecting the diverse goals and needs within the team.
Together with my team at Media Monks decided to tackle this retailer’s challenge head-on with NLP and Semantic Analysis to decode the sentiment behind a large number of app reviews on Google Play and Apple Store. This wasn’t about collecting data for the sake of it. We were on a mission to grasp the ‘why’ behind the ‘what’ in user feedback. Using the tool I’ve built together with Krasimir Bambalov, a Customer Voice Analysis tool, we cut through the noise to pinpoint the real emotions and reasons driving user reviews and how the existing app UX affects customer loyalty and conversions.
First, we assessed the Sentiment Scores and magnitudes and created a bespoke sentiment score classification model to maximize the opportunity of actually identifying pain points from negative reviews.
While sentiment analysis offers valuable insights into the emotional tone of the text, it also has limitations. This is why we moved further to Semantic Analysis, to be able to identify the topics from each review, associate them with the score, and create a prioritization model based on these insights to help the product team understand what are the biggest opportunities and pain points.
This is a mockup of how it should look like using the Zero-Shot Classification interface from HuggingFace for this review.
In my role as a strategist, I have to connect the dots between data insights and the product team’s real-world needs. The product team members often don’t have time to spend understanding, and using qualitative data because they are constantly faced with fast decision-making, dealing with feature releases, bugs, making qualitative data just a nice to have.
So I guided them to see its value in refining our product. We didn’t just share findings; we facilitated a game-changing discussion, turning feedback into plans for action and aligning it with our product goals for a user-centric development path.
During a key meeting with the UX, UI, development, and data teams, we broke down the user feedback into clear, executable steps. We aimed to ensure that every hypothesis and test would lead us closer to features that met user needs effectively.
This wasn’t about addressing problems—it was about changing our strategy. Integrating user feedback directly into our product strategy allowed us to enhance our app in a way that actively engaged our users.
My story is very similar to the story that Bhav Patel, Founder of CRAP Talks and Director of Experimentation and Analytics at Lean Convert shared with me:
“At Hopin, I was responsible for driving our metrics framework. We realized pretty quickly that we needed to integrate qualitative measures and feedback into our ways of working, but not the previous “spray and pray” method we’d used before. We identified key areas of the site to add customer effort score (CES) surveys in to based on our “aha moment” metrics. That became our real-time measure of customer effort/experience. Working closely with our VP of Product Support, I then leveraged the AI functionality in our CS tool to categorize the CS tickets that were being raised by customers. These were summarised and placed directly into our MBR presentations alongside the new quantitative metrics. Finally, we leveraged NPS to paint an overall picture of how customers felt about Hopin.”
Els Aerts, Co-founder at AGConsult and user research and experimentation extraordinaire shares a case study she worked on for itsme®, one of the most popular apps in Belgium, which highlights how lean user testing contributed significantly to the app’s success. Els shares her experience supporting them: “The team was very intentional about qualitative feedback. They really understood the value. We discussed sprint planning together, the possible impact of the research results, etc. It was one of the few times that user testing was not seen as a burden or something that would slow product development down. But as something that would greatly increase its value. Which it did. The user research uncovered some absolutely critical issues that would have made the app bomb if it had gone live like that. Thanks to the iterative user testing and the team being fully open to qualitative user feedback, it was a great success.“
Moving from Els’s success story, let’s explore a different aspect of user engagement Tracy Laranjo, Fractional Head of CRO & Experimentation, shares a powerful approach to leveraging qualitative data: “I’ve found that the most impactful way to get qualitative data alignment is by vividly portraying real issues through the end user’s perspective to your stakeholders. Instead of just describing problems, show the reality of end users through their own eyes and words. Utilize tools like screen recordings, heatmaps, and direct quotes from CX tickets and reviews to tell a compelling story. This approach is cost-effective and can reveal significant unknown-unknowns, greatly influencing the direction of product development and enhancing user experience.”
The Core Issue of Misalignment in Teams
The problem often boils down to one thing: misalignment. When teams are isolated, the product’s vision becomes blurred. It’s a story of disjointed efforts where the richness of data is lost in translation, hidden behind complex dashboards, or trapped in quantitative metrics that fail to capture the full picture.
This is particularly evident in the mobile app sector, where the average day-one retention rate is around 25%, and by day seven, it drops to 10.7%. By the three-month mark, only 29% of users remained engaged, highlighting the urgency for a more cohesive development approach.
The biggest challenge is harmonizing data strategy with UX/UI design and development efforts.
If done right, this ensures that every product evolution is deeply attuned to user feedback and business goals within the existing organizational constraints.
Talking to Eva Gonzalez, Co-founder of Datola and Data Analyst Tech Lead at Empathy.co about this, she shared: “In my opinion, successful products arise from a combination of several factors: creativity, innovation, technology, market research, data (quantitative and qualitative), user knowledge… Different teams are needed to cover each area and it is essential the alignment between them. But it is not an easy task, each team could have different objectives and end up with opposite strategies. In my experience, communication and leadership from high-level managers are key to keeping the strategies aligned across all teams in the company and that will be the first step to achieve the best version of the product.”
- 39% of employees report a lack of collaboration within their organizations. This points to a significant gap in teamwork, which is seen by three-quarters of employers as an important factor in company success.
- 97% of workers and employers believe that the lack of team alignment negatively influences the success of a task or project. This statistic highlights the critical impact of proper alignment and collaboration within teams on the overall success of projects.
Fosca Fimiani, Analytics and CRO Leader at Merkle Italy touches on the issue of compartmentalization in large organizations, where specialization can inadvertently lead to barriers against comprehensive understanding:
“In agencies, particularly larger ones, a common issue is being restricted within your specific domain, often unaware of the broader landscape. As an analyst working with data and tracking, there’s often a desire to implement more UI/UX elements or leverage MMM for better budget allocation. However, you can get stuck relying on clients to implement these or waiting for them to search for suitable providers. Yet, by looking beyond your immediate scope, you might find opportunities to collaborate with other teams and possibly score a significant upsell. Effective communication and understanding your business’s capabilities are essential.
The creation of silos within larger companies has resulted in significant specialization in specific topics but often led to a limited understanding of the broader context. When engrossed in our individual responsibilities, it’s easy to become narrowly focused and overlook the interconnectedness of various aspects that could also influence our work. While dealing with really poor performance of a checkout process, our focus may just be on problems related to payment methods or the amount of code affecting the loading time, whereas the real issue might lie in the UI. This tunnel vision could result in a loss of time and money due to a lack of communication and cross-team work.“
This sentiment of stepping out of confined roles is echoed by Eddie Aguilar, a seasoned experimentation expert and Founder at Blazing Growth, who believes that fostering curiosity within teams is the key to unlocking true product potential.
“In my experience, the foundation of productive product development lies in robust communication and precise measurement plans. I’ve seen teams charged with building features without understanding their real-world performance—because that curiosity was never sparked within them. Too often, product managers are executing orders from management without space for dialogue or question, a practice that’s all too common in corporate America. But when we open the door to understanding user engagement—like dissecting why an app’s onboarding has unnecessary complexity—it fosters a deeper connection with their work and the user’s experience.”
Eddie continues, “Measurement plans are vital, yet they vary widely. A product manager’s focus should pivot from vague goals like ‘more leads through signup’ to tangible metrics such as ‘screen load speed’ or ‘API request speeds’—factors that truly affect a user’s interaction with the app daily. It’s a misalignment I’ve often challenged because I’m driven by my curiosity, even if it means rocking the boat. The goal is to create systems that enhance performance, not resentment.”
Bridging External and Internal Insights for Roadmap Success
In a discussion about collaboration and integrating insights from internal and external resources with Kuba Bomba, Chief Product Officer at Piwik PRO, he told me more about their methodical approach to incorporating feedback in product development. He explains: “In the development of the Piwik PRO Analytics Suite, we prioritize feedback from a variety of stakeholders, both within and outside the organization. These insights significantly aid our product managers in crafting future roadmaps. Whenever we embark on developing a major feature, we initiate a product discovery phase. This includes conducting customer interviews selected based on feedback or other data sources like CRM and product analytics. Our collaboration extends to partners, often power users, who provide early feedback and test features before their official launch.”
The Piwik PRO strategy underscores the value of community engagement in enhancing product adoption. Kuba notes, “The introduction of our free Piwik PRO Core plan dramatically boosted product adoption. User feedback, primarily gathered through our community forum, frequently translates into new feature requests.”
Kuba’s approach is rounded off with an internal mechanism that empowers employees to funnel user feedback directly to product teams. “We have a structured process enabling employees to submit user feedback, ranging from feature requests to UI/UX improvements. Each piece of feedback is shared on Slack for immediate discussion, ensuring that the context remains fresh and actionable.”
What is the Team Fragmentation Puzzle in Mobile App Development?
The Team Fragmentation Puzzle refers to the complex challenge of aligning various specialized teams and integrating their distinct perspectives, skills, and data insights to create a cohesive and effective product development process.
This puzzle arises from the siloed nature of modern work environments, where teams such as UX/UI designers, developers, data analysts, and product managers often operate in isolation.
The challenge lies in harmonizing these disparate elements to ensure a unified approach to product strategy, design, and implementation, all while keeping pace with market needs and user feedback. Solving this puzzle is key to enhancing efficiency, innovation, and the overall success of product development initiatives.
Applying the TOSCA Framework for Clarity
To dissect this issue, let’s apply the TOSCA framework, identifying the Trouble, Owner, Success criteria, Constraints, and Actors involved.
- Trouble: Clients face fragmented product development processes, with siloed teams leading to a misalignment between business objectives and user satisfaction.
- Owner: The responsibility typically falls on product managers, who must navigate between development teams, user experience designers, and data strategists.
- Success: Success is achieved when cross-functional inputs are seamlessly integrated, resulting in a product that resonates with users and meets business goals.
- Constraints: Existing organizational structures, company culture, and a lack of established processes for data-centric collaboration.
- Actors: Product managers, UX/UI designers, developers, data analysts, and strategists.
*The TOSCA framework is a structured approach for crafting clear and concise problem statements by identifying the Trouble, Owner, Success criteria, Constraints, and Actors involved.
Talking to my friend Rommil Santiago, Founder of ExperimentNation about his challenges with the team fragmentation puzzle he shares:
“During my time at places that did product well (because many don’t), most of the major functions (e.g. product, marketing, growth, and design) held weekly/bi-weekly face-to-face conversations with actual end users to understand their lives, how they used tools in their field (not just ours), their aspirations, etc. We did this religiously. It was often a very open conversation where we asked questions, answered questions from users, we shared ideas, – in general, we did our best to show real empathy for our users.
While we didn’t uncover new pieces of information all the time, overall, we uncovered enough insights regularly to make it worthwhile. Based on these sessions we got so many test and feature ideas that I strongly encourage product managers to make this a part of their regular routines.”
Kelly Wortham, founder of ForwardDigital.org and the Test & Learn Community (TLC), brings a unique perspective to data-driven decision-making in product development. She says: “This approach is more than just number crunching; it’s about cultivating an environment where curiosity is a driving force. We’re looking at data not as a barrier to productivity, but as a foundational element for innovative solutions.”
Kelly highlights the role of data in fostering a collaborative and innovative culture. “Integrating data into everyday discussions opens the door to insights that aren’t just about product enhancements but also about how our teams operate and excel together. The true value in our product teams comes not only from the data we collect but from how we articulate its relevance. Moving beyond the ‘what’ of data to explore the ‘why’ and ‘how’ is key. How can these insights elevate user experience? Why are these metrics important for our product’s evolution? These questions should be a regular part of our dialogues, transforming data from mere figures to a critical tool for innovation and advancement.”
Strategies for Effective Mobile App Development Collaboration
To effectively tackle the fragmentation puzzle in product development, you should adopt a multi-faceted approach that emphasizes the importance of both quantitative and qualitative data.
Start by balancing the hard numbers with the rich, contextual insights that qualitative data offers – remember, while metrics tell you the ‘what,’ it’s the stories behind these numbers that reveal the ‘why’. Equally important is the presentation of this data. Simplify it to ensure that team members of all technical levels can grasp and engage with the information, fostering a collective understanding.
Tim Ceuppens or as he calls himself “The Data Raconteur”, CMO at Pit&Pit shares: “A story I always tell is this one: a few years ago I was tasked with optimizing the lead gen. We had a critical point where people dropped out. The more information tabs were used frequently, this we saw in the data.
But then: nothing happened. This we also saw in the data. Then we used Hotjar to find why.
Once we had that we could just present conversion metrics. But that doesn’t do much by themselves. So we asked the board to pull out their phones. Go to a page. Use the information tabs and then ask them to click the sign-up button. To which they went: what button? And we went: exactly. We should probably add that. Now that may seem silly, but the fact that they experienced it meant that that and a few other changes became no-brainers to implement. We could never have had everything approved that quickly by just showing numbers and guesstimates.”
Lastly, the key to unifying fragmented teams lies in nurturing continuous dialogue. Facilitate regular cross-functional meetings, not just check-ins but open forums for sharing insights, raising questions, and collaboratively exploring solutions. This approach not only demystifies data but also turns it into a powerful tool for team alignment and effective product development.
What are other experts recommending
Jim Gordon, an analytics industry leader with a wealth of experience as a product owner and SaaS entrepreneur, shares his candid insights:
“In the charged dialogue about dismantling silos within product management, there’s a nuance that often gets overlooked. Not all silos are detrimental; some are repositories of excellence, and breaking them down can be counterproductive. From my journey, I’ve come to appreciate that data-driven decision-making should enhance, not hinder, the product owner’s vision. Objectivity is essential, but it’s not the only currency in play. It’s a contentious standpoint, but the truth is that product teams don’t always need an analyst if they’re proficient in self-service and meeting their objectives. The real challenge isn’t tearing down silos for the sake of it, but rather being strategically useful. There’s an art to knowing when your expertise is truly needed and when stepping back is the wiser move.”
Reflecting on his time as a product owner, Jim remarks, “The relentless pursuit to break silos must be met with discernment. The goal is to be worth someone’s time. It’s about understanding when to be a soundboard and when to take initiative. For those in visionary roles, it’s about being the catalyst that aligns with their dynamism, helping them capitalize on opportunities.”
He concludes with a pragmatic take, “Not every silo warrants demolition. It’s about being that ‘very useful engine,’ to borrow from Thomas The Tank Engine. For the visionaries and the executors alike, the key is to offer solutions that truly fit their needs and drive the vision forward.”
Jim Gordon’s perspective sheds light on the much-needed balance of contribution and adaptation within team structures. It’s about more than just tearing down silos; it’s about integrating meaningfully, ensuring each effort aligns with the product’s broader goals.
Segueing from Jim’s point, Shiva Manjunath, experimentation industry leader and host of the experimentation Podcast ‘From A to B’, underscores a similar but distinct aspect of team dynamics. He points out that high-caliber team members alone don’t guarantee success: “It doesn’t matter how good your team members are – if your data is bad and no one is talking to each other, you are creating limitations in the growth of your program.
Sports example because that’s just what I do. You can have Michael Jordan, Lebron James, and Steph Curry on your basketball team. If they aren’t working in a system that puts all of their skills together, they won’t succeed. Most people struggle to define, and communicate, concrete product goals. If there is a clear vision of what the primary goal is for users, alignment from there is much easier. So if you are struggling to get results, take a step back and identify if you are focused on the proper goals, and if you can even track success for those goals.”
Adding up to the importance of communication and cross-functional collaboration, Andra Baragan Founder at Ontrack Digital, a seasoned experimentation strategist in the field, shared her insights: “One of the biggest lost opportunities I see in organizations is the lack of communication between departments. This is especially true when multiple external agencies are involved but fail to work towards the same strategy. The qualitative research done in conversion optimization is immensely beneficial beyond just marketing, as it taps into the customer’s voice, which is key for any business.
We’ve seen impressive results when clients apply CRO and qualitative research insights across all business aspects. When multidisciplinary teams engage, they often produce stronger solutions and uncover more opportunities, thanks to their varied understanding of the business. In terms of disseminating these insights, we focus on making our audits and A/B test reports easily digestible for anyone in the company. The goal is to clarify how we arrived at these insights, their impact on business objectives, and the next steps. Keeping information simple, relevant, and actionable, and ensuring it reaches the right people, leads to much better results and stronger collaboration in the organization.”
Gerda Vogt-Thomas, Co-founder at Koalatative and experimentation expert, brings insights from Sean Ellis’s book ‘Hacking Growth’: “Sean Ellis emphasizes the importance of reporting data discovery in the simplest and most accessible way possible. If no one outside of the data analysts can understand your findings, they are unlikely to result in meaningful action. Massive spreadsheets, database queries, and highly technical displays, valuable for professionals, may be daunting to other team members.”
Gerda continues with her perspective on the application of these principles: “Most of us tend to assume that everyone knows what we know, but this is far from the truth. As marketers, we focus on getting to know our users, yet spend little time applying this mindset within our organizations to understand the people we need to communicate our findings to. Qualitative data interpretations can be especially complicated due to their subjective nature. If you’re analyzing open-ended, long-form text answers about why someone didn’t buy your product, you’re faced with the spectrum of human emotion.
Furthermore, differing interpretations of a customer survey by team members can lead to varied results. There’s also evidence that many companies don’t act on the qualitative data they gather, leading to analysis paralysis. To engage others in the work you’re doing, it’s essential to explain how it benefits them and to offer simple, actionable next steps. Storytelling is a valuable skill, but as an analyst, you need to understand and explain the next steps based on the story uncovered and how this contributes to the greater good of the organization.”
Main takeaways
- Enhance team collaboration: Emphasize the importance of proper collaboration across product teams to leverage data effectively for personalization and user engagement.
- Prioritize qualitative data analysis: Utilize NLP Semantic Analysis, and NPS analysis to decode user sentiments, understanding the ‘why’ behind the ‘what’ in user feedback.
- Integrate user feedback into product strategy: Convert user feedback into actionable plans, aligning it with product goals for a user-centric development path.
- Address misalignment in teams: Recognize and tackle the core issue of team misalignment to prevent disjointed efforts and loss of valuable data insights.
- Simplify data presentation: Communicate data findings in a simple, accessible manner to ensure understanding and action across teams.
- Foster continuous dialogue: Encourage regular cross-functional meetings for sharing insights, raising questions, and collaboratively exploring solutions.
- Apply lean user testing: Implement iterative user testing and be open to qualitative user feedback, as demonstrated in the itsme® case study.
- Portray real issues vividly: Use tools like screen recordings, heatmaps, and direct quotes from CS tickets and reviews to bring end-user perspectives to stakeholders.
- Leverage a variety of feedback sources: Incorporate insights from internal and external stakeholders, as practiced by Piwik PRO, for comprehensive product development.
- Balance silo strengths with team integration: Recognize the value of specialized silos while strategically integrating teams to align with broader product goals.
In essence, the fusion of diverse team strengths with comprehensive data understanding is the bedrock of successful product development. It’s about moving beyond the ‘ship it’ mentality to a culture of ‘validate and understand’. As we look ahead, the key lies in combining the strengths of diverse teams with a holistic understanding of data, driving not just user engagement but long-term loyalty and success.