In today’s forward-thinking, customer-focused organizations, product teams are constantly evolving and changing. New teams and functions are created as a way to empower the specialized expertise that’s leveraged to establish a competitive advantage. As the issue of acquiring and retaining users is critical to a product’s success, the evolution of product growth teams that focus on moving the needle on critical success metrics of their products should come as no surprise.
By all accounts, it looks like the product growth team trend is catching on. Google Trends for product led growth searches show an explosion of interest over the last few years. Searches for “growth product manager” over the last 5 years point to a 425% increase in monthly interest, while a cursory search on LinkedIn shows that there are 2,600 people listed there with titles that contain both product and growth. And apparently, there’s a good reason for all this interest: recent research shows that 62% of organizations agree that every company should embrace growth PMs.
At Loops, our mission is to help product teams use their data to easily discover growth opportunities. When we were just getting started in our product discovery process, we interviewed more than 300 different growth teams, including ones from highly successful companies such as Facebook, Pinterest, Google, and Lyft. Not only was this process insightful and critical to the development of our product, it also helped us uncover quite a few key processes and approaches shared by those growth teams that found success where others failed.
We’re excited to share a few of the most valuable of those lessons in this article.
1. They create processes and infrastructure that support the high-velocity experimentation that’s crucial for growth.
At their essence product growth managers are sort of like scientists. They leverage data to draw conclusions that form the basis for hypotheses about how to generate growth. And like any scientist worth their lab coat knows, a hypothesis is basically just an educated guess; that is, until it’s been tested and proven to be correct.
To ensure a steady stream of growth-oriented features for their products, growth teams must constantly experiment and quickly draw their conclusions. At leading companies like Google and Facebook, hundreds and sometimes thousands of experiments are run on a monthly basis for every single product.
Successful growth teams create a stable cadence of high-quality experiments that prove beyond a reasonable doubt that a proposed hypothesis is true, and that a certain initiative will actually move the KPI. Naturally, many of these experiments will fail, that is simply a part of the process. Nevertheless, each experiment, even those who fail, provides valuable learnings, insights and data points that compound over time. By maintaining high-velocity experimentation you can increase the likelihood to have successful experiments and gather plenty of learnings that enable you to build a better product.
But, ensuring a steady, high-velocity stream of high-quality experiments is easier said than done. It requires that processes and infrastructure be put in place that facilitate these complex procedures. Everything from tracking the number of expected experiments per month, communication frameworks, to capabilities that support detection of anomalies and failures in the experiments, and all the way to the implementation of more advanced models (such as Bayesian model analysis) to use for experiments with insufficient data, must be taken care of before any high-scale experimentation can be put into practice.
One product growth leader we recently spoke with told us how they handle experiments at an extremely successful lifestyle app with millions of monthly active users. They measure themselves by the number of experiments they run while using clear, predefined templates for sharing the experiment plan before it’s launched, and others for documenting results and lessons learned after it’s ended. These are then shared through dedicated email groups and Slack channels with relevant stakeholders. This process helps empower a healthy, focused discussion where every aspect of the experiment can be honestly and openly thrashed out.
2. They understand the tradeoffs and relationships between their different growth metrics and prioritize them based on their current growth model.
Like many things in life, product growth in one metric can often come at the expense of another. For example, optimization measures taken to increase acquisition may unintentionally reduce retention. Think of it as a two-steps-forward-one-step-back type of situation.
While all growth teams are acutely aware that these tradeoffs exist, the successful ones plan for them ahead of time and build clear growth models that take them into account. These growth models map all the different areas in the product; acquisition, retention, engagement, monetization levers and more, explain their co-influence, reflect the overall growth plan and put them into a detailed mathematical equation. Based on these models, successful teams can drive smarter prioritization decisions that measure both the growth delivered by a new initiative, and its tradeoff and potential negative impact. An example might be releasing an exciting feature that attracts new users while looking at the guardrail metric on 2nd-week retention to monitor for negative effects such as acquiring poor-quality users.
3. They set up a structured, data-driven model to set users up for success.
Product growth managers are data hungry creatures. Every move they make and every step they take is according to the data that flows from their product. Successful growth teams use this data to identify key Aha! moments and habit forming interactions with their products and translate them into clear, quantifiable models.
They then leverage these models to develop customized, highly specific user journeys that expose new users to those elements at exactly the right time. They deploy these journeys in a way that significantly increases the chance that a user will quickly find the value they’re looking for in the product and become successful users.
For example, with enough data in their hands and with the right statistical models, growth teams can identify when the Aha! moment occurs for their different target audiences. Successful product growth teams identify the exact combination and timing of features that get their users hooked and make sure they develop an activation flow that leverages these insights.
The Cold start problem is one example of a real life product challenge where some of the solutions can be found in product data. In this scenario, a product’s users only see its value after they’ve been using the product for a while; however, a difficulty seeing the product’s potential value early on holds users back from using it enough to realize said value. By leveraging data that shows when users are first realizing a product’s value, growth PMs can surface these critical moments earlier in the user journey so that users are more likely to continue using the product.
4. They constantly measure the impact of their actions in the long-term, instead of shipping a new feature and then moving on.
While excellent for testing functionality, experiment environments are lab-like in essence and fall short when it comes to replicating real-world results. In the real-world there is no vacuum, so different factors and changing circumstances often influence the way users actually interact with a product (e.g Novelty effect).
That’s why the impact of every change in the product, from the launch of a new feature to even the slightest modification in the onboarding flow, should be measured over the long-term. It’s the only way that PMs can be confident that the effect they’re seeing is real and was developed and measured using a clear, unbiased, data-driven approach to product development.
Growth teams are well-aware that the only way to truly understand the impact of their work is to constantly measure and analyze its impact to ensure that the results they’re seeing are credible. What sets the successful teams from the rest though, is their ability to go back to the data even long after the experiment has ended, learn from the process, analyze the long-term impact and iterate accordingly. In the messy environment caused by high-velocity experiments and endless KPIs, doing so is hard to prioritize and more so, to execute.
Conclusion
Product growth teams are still the new kid on the product block. This innovative function is focused on optimizing the product to achieve growth in prioritized KPIs such as acquisition, retention, conversion and more. These elite teams are fueled by product data generated when users interact with their products. They use this data to better understand how users gain value and to identify those key features and processes that create the most of it for their users.
The most successful among them leverage the approaches and processes described in this article. They measure the impact of their actions over the long-term so that they’re sure that their effect is significant and sustainable. They build growth models that account for the performance tradeoff of their actions, as any change that improves one metric could negatively impact another. They lean on data to quantify and model the flows that quickly surface value for their users. They also build infrastructure to support high velocity experimentation as a way to generate crucial data and ensure a steady backlog of high impact experiments. At the end of the day, everyone uses the data, everyone looks at dashboards. But what separates the successful growth teams from the rest, is the way they work with their data – they are able to deeply infuse it into their growth process, decision making and strategy, and unlock real exponential growth.
