Analyze Product Insights & Validate What to Build Next

In today’s competitive landscape of product development, analyzing and understanding the problem space along with validating assumptions are paramount.

💡 According to a study by McKinsey, 72% of product leaders agree that insights from customer feedback significantly influence their company's decision-making process. This underscores the importance of analyzing product insights and validating actionables in the product discovery journey.

Imagine a scenario where a leading e-commerce software platform (for now, let's call it "E-Com X") wished to improve its mobile user experience and conducted an extensive user research. They deployed surveys, gathered feedback from customer support interactions, monitored in-app engagement in product analytics platforms, and even conducted usability sessions. They gathered tons of both qualitative and quantitative data regarding user preferences, pain points, and feature requests.

But, it happens so that upon initial data capture, E-Com X found itself overwhelmed with a vast amount of feedback from its 1M+ user base. Their users expressed a range of opinions that were quite the opposite of each other, ranging from feature requests for improved checkout processes to complaints about slow loading times on certain devices.

Without any structured approach to analyzing and validating this data, E-Com X was at a fix on what to build next. As a result, it risked drowning in unprocessed information, leading to significant opportunity losses.

Importance of in-depth product insights analysis before feature-planning

Feedback validation is not just about having data; it's about extracting meaningful product insights that drive strategic decisions and product improvements.

Without this validation, you risk building features that might not resonate with your users or address their real needs, ultimately leading to wasted resources and missed opportunities.

And it’s a challenge that’s pretty common among product managers handling user feedback at scale. You might be -

• overwhelmed by tons of unstructured feedback data coming in from multiple sources
• confused because of data siloes created as a result of having very different sources of feedback in the first place
• frustrated with the noise in the data as the information comes in from different kinds of GTM teams (sales, success, support, marketing), leading to biases and opinions
• unaware about which product insight has the highest impact on your most important customer segment.

As a result, deciding which customer issues to address first or “what to build next” can be difficult, especially when there are competing priorities.

Exercise for defining product insights and validating them

1. Categorize feedback into the 3 big buckets: To start off, it’s always a good practice to have a broad understanding of user feedback, as to whether these fall into —

1. Customer needs
2. Customer desires
3. Customer pain points

These broad categories help you prioritize your intial action. If these are fundamental customer needs that are not getting fulfilled, you might want to revisit your product functionalities, whereas, if these are fresh & relevant customer desires, you might want to dive into a research on how it impacts your business goal. Customer pain points, on the other hand, should always have a separate prioritization on its own.

2. Once the feedback is organized into broader themes or buckets, the next step for you is to further subdivide these themes into more specific categories. Most common categories include —

1. Product Usage Feedback: Insights into how users interact with your product, highlighting areas of frequent use or confusion.
2. Feature Requests: Specific features that users have requested, indicating potential areas for product enhancement.
3. Complaints & Bug Reports: Issues and bugs reported by users, which need to be addressed to improve the user experience.
4. Praise or Appreciation: Positive feedback that highlights what users love about your product, which can be leveraged to reinforce successful features.
5. Sales Objections: Reasons why potential customers might hesitate to purchase or subscribe to your product in the first place.
6. Customer Churn Reasons: Negative feedback from users who have stopped using your product, providing valuable insights into areas that need improvement to retain users.

The next step would be to analyze these sub-categories to identify recurring topics. This method helps in pinpointing your specific product areas that need your attention. AI Product Discovery tools like Zeda.io helps you with —

• Frequency Analysis: Helps count the number of times specific issues or requests are mentioned. Higher frequency indicates a more pressing concern or popular request.
• Sentiment Analysis: Helps evaluate the sentiment associated with feedback to understand the emotional tone (positive, negative, neutral). This helps in gauging the urgency and impact of the feedback.
• Keyword Extraction: Helps identify common keywords and phrases used in feedback to detect recurring themes and topics.

3. Formulate testable hypotheses based on these customer & product insights: A well-defined hypothesis should be clear, concise and focused, addressing a very specific aspect of the product that you want to validate.

1. Identify common product insights that you want to explore further. This could be a common user need, a new feature request or a recurring complaint.
2. Once you have your product insights in plain sight, begin with a clear assumption, such as "We believe that by adding/implementing feature X, users will achieve Y." E.g. “By implementing a one-click checkout feature, our customer brands will be able to reduce cart abandonment by 15%”
3. Next, define the metrics that will indicate your success or failure of this experiment, ensuring they are specific and measurable. E.g. “We expect that this feature will increase our user retention by 30% over the next quarter.”
4. Finally, establish a set time frame for testing the hypothesis, allowing for a thorough evaluation of the results within a designated period.

4. Design and conduct experiments with “chosen users” to validate hypotheses:

Launch the prototype, MVP, or feature to the chosen user group - it can be your champion users, for instance. Make sure that the deployment process is smooth and that users understand how to interact with the new feature or product, with walkthroughs, tours and assisted-demos. For instance — you can test your prototype's visual design and layout with potential users to understand their preferences.

1. Create Multiple Mockups: Develop several high-fidelity, visually distinct mockups showcasing different product interfaces and style options.
2. Recruit Engaged Participants: Find participants, most preferably - your champion users, who can openly share their honest responses to your designs. Make sure they represent a good chunk of your target users.
3. Use Adjective Cards: Prepare index cards with a mix of positive and negative adjectives, or use pre-made decks like Microsoft's 118 Product Reaction Cards.
4. Conduct Sessions: Show each visual direction to participants and have them select cards that best describe their feelings towards each style and layout.
5. Analyze Feedback: Collect responses and, as a team, determine which design direction best aligns with your product goals based on the selected adjectives and attributes.

After the testing period, analyze the results to determine whether the hypothesis is validated or refuted, and whether you need to proceed with the actual product development.

Best practices for generating & validating product insights

✅ Best practice #1: Bring in quality feedback to start insight analysis.

‍

Often, product managers face the dilemma of either gathering too little feedback, which may not provide a holistic view, or drowning in excessive but superficial responses that lack depth.

Always try to collect a good number of detailed feedback entries for actionable insights.

And most importantly, make sure that these feedback entries include substantial context with respect to an idea, feature request, complaint, and so on - rather than 1-2 worded reviews.

Data cleaning plays a crucial role here. This involves encouraging users to provide comprehensive feedback, avoiding vague or superficial responses. For instance, employing detailed feedback forms with concise yet open-ended questions enables customers to express their thoughts freely, including subjective viewpoints.

By collecting meaningful and context-rich feedback, you can extract valuable insights that drive informed product improvements and decisions.

✅ Best practice #2: Regularly create custom reports to create your own problem statements, and validate them.

Without simplified reports, product managers often find themselves grappling with the situation of sifting through tons of insights - in an ad hoc way. There is a lack of systematic way to organize and analyze insights.

Make it a habit to create multiple custom reports to craft your own queries and develop your own problem statements and hypotheses, regularly.

With AI product discovery tools like Zeda.io, create reports using chart views that encapsulate —

• Product Insights (Y-axis) → You can view the key product improvements areas or further break down into insights.
• Measurement metric (X-axis) → These are the metrics with which you want to measure your product insights - no. of feedback, recurring revenue, open deals, etc.

These information cards let you dive into the top requests/ complaints/etc. with a quick AI-powered summary of the chart.

Using these reports, your product team and other stakeholders can receive a weekly digest that highlights key insights on product areas needing improvement, ensuring everyone stays well-informed.

This should also facilitate more data-driven decision making.

✅ Best practice #3: Save time by leveraging AI to summarize key product insights.

‍

Manually sorting through countless feedback entries

• takes a lot of time
• can lead to mistakes

… resulting in overlooked details and delayed decisions. This slow process makes it hard to quickly address user needs and make informed product improvements in a timely manner.

AI-powered summaries distill numerous feedback pieces into coherent product insights, providing a clear understanding of user needs and issues.

The supporting feedback acts as evidence for these AI-summaries, giving product managers a solid foundation for their data-driven decisions.

This approach not only saves time but also ensures that decisions are based on comprehensive and well-organized data, leading to more effective product improvements.

Wrapping up Stage 2 of product discovery

By thoroughly and consistently analyzing customer & product insights, you are not just building products that resonate with your users but most importantly — mitigating the business risks associated with ad-hoc product development.

In the end, it’s about building the right thing - and not just about merely building and shipping stuff. And that starts with the first two stages of product discovery - understanding the problem space and validating through actionable insights.

Let’s move on to the next stage where you learn how to incorporate these product insights into your product roadmap.

See next chapter →