The 10 KPIs every product leader needs to know in 2025

Proven metrics for driving product—and business—success

Introduction: The modern challenges and opportunities of product leadership

What does it mean to measure the success of your product and company?

Sure, there are a few mainstays that we can always count on—increasing a product’s active users, adoption, and retention will always be fundamental to ongoing business success.

But things are slightly different these days. For example, Gartner predicts that 75% of enterprises will operationalize AI by 2025. AI is rapidly changing modern enterprise operations, and many of these businesses are working in real time to understand just how to measure success. All the while, work and leadership practices are changing—automation, vibe coding, and user intelligence are just the tip of the iceberg.

Being a product leader in 2025 requires a careful balancing act between technical fluency and strategic vision, bringing clarity to the noise out there. With new technologies and new ways of doing things, product leaders are finding that now is an incredibly exciting time to lead a product to success.

In this guide, we’re taking our experience working with 14,000+ customers over the last year to discuss the 10 KPIs that every product leader needs to know. While we’re sticking with primarily what we think are crucial KPIs for every product team, we’re also digging into how AI changes the equation.

Understanding what’s essential for measuring product success

Let’s zoom out and understand the different types of KPIs at a high level. Here are three categories to keep in mind:

• Business outcomes: What are your most crucial business outcomes, and how does your product impact key business and financial outcomes in the short and long term? These might be closely tied to reducing churn, increasing long-term user retention, or bringing in new users or subscribers.
• Product usage: These KPIs reflect how users behave inside your product. Which features do they use the most? Where are they getting stuck? These metrics are critical for understanding how users do or do not find value in your software.
• Product performance: These KPIs measure how well your product performs its intended function. What is your product response time? How much downtime did you experience last month? It’s often best to set an internal benchmark for quality metrics and measure yourself against it monthly or quarterly.

The 10 essential KPIs for 2025

When selecting our top KPIs, we considered indicators that would help product leaders track crucial aspects of their apps, such as costs, feature adoption, and user sentiment. These metrics address the dual demands of innovation and stability, helping you understand how your users engage with your software so that you can reduce churn while getting key insights on what they might want in the future.

1. Predictive customer lifetime value (CLTV)

It’s common knowledge that it costs significantly more to gain new customers than to retain them. This is why it’s critical to understand the lifetime value of a long-term customer.

Traditional CLTV models, which rely on historical averages, struggle to account for dynamic variables like AI-driven behavior shifts or supply chain disruptions. More competitive operations are turning to predictive models driven by AI to understand how present-day decisions can impact future revenue.

Predictive CLTV integrates machine learning to analyze real-time data streams, including user interaction patterns, macroeconomic trends, and geopolitical events. They can use transaction data, user behavior metrics, and feature or product preferences.

For example, a B2B software provider might use predictive CLTV to allocate resources to high-potential industries, such as healthcare or renewable energy, where regulatory changes create growth opportunities.

Why is CLTV important?

Predictive CLTV is critical to gaining a deeper understanding of value across your products and leverages machine learning to provide more accurate value estimates. In addition, it can help you better understand how users may adopt new features, especially those rooted in AI.

How to measure CLTV

CLTV is simple to calculate: Take a customer’s annual value, multiply that by the customer’s average lifetime (in years subscribed or licensed), and subtract any acquisition or maintenance costs (such as general customer acquisition costs, time-to-value costs, or other metrics).

2. AI feature adoption rate

Feature adoption is essentially a user activation metric, and feature adoption for an AI product is a clear demonstration of the immediate and ongoing value your AI products bring to users.

This KPI measures how deeply users engage with AI features like chatbots, automated reporting, or predictive analytics. A low adoption rate may signal a misalignment between user needs and feature design. For instance, a CRM platform might discover that its AI-driven sales forecasts are underutilized due to a lack of education, prompting redesign of the in-app onboarding experience.

If your team understands how your user base engages with AI features, they can better support users and encourage continued adoption.

Why is AI feature adoption rate important?

Tracking AI adoption ensures ROI on investments. Low adoption signals misaligned feature design or poor onboarding, while high adoption correlates with retention and revenue growth. Measuring engagement with chatbots or predictive analytics helps teams refine user experiences and justify AI development costs.

How to measure AI feature adoption rate

The adoption rate is expressed as a percentage of your total user base with access to the feature:

How to increase feature adoption

One of the core benefits of measuring adoption is understanding how to drive customers towards the functionality that would benefit them, making them more likely to stay engaged with your product. When it comes to driving adoption of key features, it’s most effective to do so within the product itself. In-app tactics to boost feature adoption include:

• Using in-app messaging to announce new features.
• Using in-app walkthroughs to guide users to specific features and help with onboarding.
• Using segmentation to target communications to customers who would find value in the feature.

Real-world spotlight: Okta

Measuring user experiences involves understanding your user base at a granular level. Okta found that Pendo helped them understand their existing customer experience and identify improvement opportunities.

“We can segment, target, and re-target users across channels based on the data that we have at our disposal so that we can effectively coach them, guide them, or nudge them through critical core pieces of configuring Okta.”

Tom Witczak, Director of Digital Experience, Okta

3. User growth

Growth might be the single most important metric to the health of your organization. Think of growth as a reflection of your brand’s outreach efforts: How well are you attracting new users? How well are your efforts working to expand usage among your existing user base?

Why is user growth important?

As the old saying goes, you’re dying if you’re not growing. Changes in growth metrics can help teams identify problems in outreach, user experience, and other concerns. “Growth” itself can vary as a category, but for software platforms, growth will directly tie to expanding a user base and, following that, revenue expansion.

That being said, you can also measure growth based on other factors, like how much time these users spend on your platform or using certain features.

Measuring user growth over time

The key to measuring user growth is to understand essential measures of time. The most critical time frames for most subscription-based software platforms are months and years. More specifically, it measures the growth of users month over month and year over year.

The formula to measure Monthly Active Users (MAU) growth can vary based on the timeframe. For example, to measure growth from one month to the next would mean finding the growth percentage of that period:

This expresses that growth as a percentage. To measure a larger time period, simply extend this formula. For example, to measure growth from the start of the year to the end of that year, the formula would look like:

4. Product Engagement Score (PES)

PES is a broad indicator of your product’s overall health. It combines metrics that measure stickiness, adoption, and growth. PES is meant to provide an at-a-glance pulse check, but you can dig deeper into the metrics it encompasses to find and fix trouble spots.

Why is PES important?

PES is a guiding light for how users adopt and continue to use a product or feature over time. It’s also a great “big picture” way to get an idea of how a product or feature is performing, giving you an idea of where you might start digging in to understand individual KPIs better.

How to measure PES

PES is a composite of stickiness, adoption, and growth. Once you’ve calculated values for adoption, stickiness, and growth, you can calculate your PES by simply taking the average of the three numbers and multiplying by 100:

You can choose to measure your PES based on visitors or accounts. If your product is only used by individuals (and not teams), then you’ll likely only want to measure each PES component at the visitor level. If your company is focused on new logo acquisition, measuring growth (and overall PES) at the account level will best reflect those efforts. When in doubt, try to align your PES configuration with your organization’s current overarching priorities.

Better PES = better business outcomes

The Pendo data science team challenged itself to predict customer churn, usage, and growth rates using PES alone. They found that accounts with higher PES in the months leading up to the end of a contract were more likely to renew, while lower scores correlated with churn. Just six months of PES scores created a reliable indicator of renewal.

5. Time to value (TTV)

In competitive markets, users expect immediate ROI. TTV measures how quickly users achieve their first “core event”—a pivotal action demonstrating value, such as completing data analysis in a BI tool or deploying code via a DevOps platform. Slow TTV often correlates with high abandonment rates. For example, a no-code web design platform might reduce TTV by guiding users through templated workflows, enabling them to launch a landing page within minutes.

Some key components to consider when looking at TTV include:

• User onboarding efficiency: How quickly can users navigate the product effectively?
• First meaningful outcome (FMO): When does the user complete their first foundational task (creating a report, scheduling a meeting, etc.)?
• Full value realization: What process does a user undergo to fully realize a project, artifact, or outcome with your product?

Why is time to value important?

Users demand access to value as soon as possible. Shortening TTV through streamlined onboarding reduces abandonment and drives competitive differentiation, especially in crowded SaaS markets.

How to measure TTV

There isn’t a general formula to calculate TTV—understanding this metric means clearly defining what the “value” proposition is (use of a feature, deployment of a particular process or practice, etc.) and the start point of your measurement (from the first registration, from first login, or another milestone). Based on these criteria, TTV could be best expressed in days, minutes, or even seconds.

6. Product performance

User tolerance for performance issues has plummeted, and delays of even one second can erode trust. This KPI quantifies uptime and responsiveness with benchmarks like 99.9% uptime and sub-2-second load times for critical actions. For example, an e-commerce platform experiencing checkout delays during peak traffic is at risk of cart abandonment and reputational damage.

Some critical performance metrics to understand include:

• Average system uptime
• Page load times
• Response times for user tickets
• User satisfaction scores

Why is product performance important?

Performance metrics (uptime, response times) will be non-negotiable in 2025. A one-second lag can slash conversions by 7%, making near-instant load times and 99.9% uptime essential for retaining users. This is especially true in sectors like e-commerce or fintech.

Wie schlägt sich Ihr Produkt im Vergleich?

Product performance benchmarks don’t exist in a vacuum. They are best understood when compared to other internal metrics or data from peers in your industry. With that in mind, Pendo compiled data from over 6,800 applications across 2,500 of our customers to help teams benchmark their products based on region, company size, and industry. See how your product stacks up with the interactive benchmarks.

7. Revenue uplift from personalization

This KPI measures the revenue uplift from personalized experiences, such as dynamic pricing or tailored content recommendations. For example, an AI-powered e-learning platform that customizes course suggestions might see higher completion rates and subscription renewals.

Determining ROI as a function of personalization means linking the incremental benefits of personalization directly to revenue gains and cost efficiencies. Approaches can include:

• Identifying key metrics related to user uptake and retention, such as feature adoption, lifetime value, and churn. It’s essential to establish a baseline with these criteria.
• Linking personalization approaches with metrics, like using generative AI to create in-app messages or tracking behavioral analytics to deliver timely support for users interacting with key features.
• Experimenting with A/B tests to help link specific personalization tasks with positive business results.

Why is ROI from personalization important?

AI-driven hyper-personalization boosts engagement but requires measurement to justify the associated expense. Measuring revenue uplift from tailored experiences (e.g., dynamic pricing) ensures that efforts align with business outcomes.

Real-world spotlight: Alarm.com

Alarm.com wanted to ensure its customer service reps could meet users’ needs with tailored guides and support features. With Pendo Analytics, the company used paths and funnels to understand better what customers were thinking and feeling so that it could tailor content to their specific needs.

“It made such an impact on agent operations and overall sentiment during the support center launch. And equally important, it’s allowing us to grow in scale because we’re spending much less time developing asynchronous training by creating on-the-job resources.”

Mary Kidd, Systems Manager, Customer Operations at Alarm.com

8. Predictive churn rate

Churn remains a critical metric, but predictive models now identify at-risk users before they leave. ML models assign churn risk scores (0–100) by analyzing engagement drop-offs, payment histories, and sentiment trends. For example, a streaming service might flag users who reduce watch time after a price hike, triggering targeted retention offers.

Why is predictive churn rate important?

Predictive models identify at-risk users before they leave by analyzing engagement drop-offs and payment trends. Proactive retention strategies (e.g., targeted offers) reduce attrition, safeguarding revenue in subscription-based models.

How to measure predictive churn

Traditionally, churn can be expressed as the formula:

However, predictive churn is more forward-looking. It means using analytics and monthly trends to determine potential churn in upcoming months. Machine learning and AI can provide deeper analysis.

9. Net promoter score (NPS)

Would your users recommend your brand? If not, you might use NPS to understand their sentiments about their experience with your products.

Typically, an NPS is gathered by surveying users and customers, separating “promoters” from “detractors” based on their responses, and then expressing that relationship as a ratio of the two.

These scores are a great glimpse into users’ hearts and minds, especially when collected directly in software. The responses are contextual and timely, providing ways to address issues proactively rather than reactively. This makes NPS especially useful as an early warning system for customer dissatisfaction and experiences.

Why is NPS in feedback loops important?

Understanding customer sentiment is always worthwhile, and tracking NPS consistently helps you see how customers will advocate for you now and down the road. However, NPS is a qualitative measure, which means it works best in conjunction with quantitative data—like product usage—to support it.

Real-world spotlight: Cision

Cision wanted to know how new customers felt about the brand and, specifically, how higher NPS scores emerged from different customer behaviors. Using Pendo, they used targeted NPS surveys to discover which indicators made customers more likely to promote the brand.

“The ability to roll up the data to the account level is a big differentiator for us. None of our other analytics tools can do this.”

Becky Banasik, Vice President of Customer Success, Cision

10. Feedback-to-feature velocity

This KPI measures how quickly teams translate user input into product improvements. For example, a project management tool might prioritize feature requests from power users and release updates in quarterly sprints.

Why is feedback-to-feature velocity important?

Your platform or product is continuously being developed using a SaaS development model. This means that user feedback will be critical in addressing software limitations and providing a product that meets the needs of an established (and, ideally, loyal) user base.

How to measure feedback-to-feature velocity

This metric is expressed as an average period between initial user feedback or alpha reviews and the implementation of features related to that feedback. You can draw from your comments and feedback platforms to see how feedback loops result in incremental change and document how many production features resulted from direct or indirect user feedback.

Real-world spotlight: Filevine

Legal platform Filevine had established internally that they needed to embed rolling feature updates in their development cycle. They needed to keep the software in development while maintaining a long-term roadmap rooted in customer feedback, one that reflected the industry’s demands while still ensuring regulatory compliance. They used Pendo Feedback to collect user responses with targeted, in-context feedback requests.

“Using Pendo’s behavioral analytics and in-app guidance in conjunction with Pendo Feedback made it a perfect package. The ability to have all of the usage contexts, right alongside the feedback, was pivotal.”

Mary Lyon, Director of Product Feedback, Filevine

Evergreen KPIs

While you start measuring, analyzing, and working to improve the AI-augmented KPIs we’ve covered above, don’t neglect the well-established (and still-relevant) KPIs. Here are three that will never go out of style:

Net revenue retention (NRR)

Net revenue retention (NRR) remains a cornerstone for subscription-based businesses, partly because it supports ongoing revenue retention. According to McKinsey, companies with NRR of at least 120% could expect 20% growth yearly without acquiring new customers.

NRR is the difference between expansion and churn. NRR is generally expected to have a value above 100%, assuming that you’ll lose some customers but also find ways to expand the value of the customers you keep.

How to measure NRR

As the definition suggests, you can measure NRR by subtracting churn from expansion, which requires four values:

• Monthly recurring revenue of the last month
• Revenue generated through upsells and cross-sells
• Revenue lost due to down-sells
• Revenue lost due to churn

Companies usually track NRR monthly or yearly, but we advise looking at your monthly net revenue retention to take timely action when there are fluctuations.

Akzeptanz

Adoption metrics validate product-market fit but now require granularity to distinguish between general product usage and engagement with AI-driven features. In 2025, products will increasingly rely on embedded AI tools—such as automated workflows or predictive analytics—to differentiate themselves, and (according to Accenture) 97% of executives see this move as a transformative one for their industry.

Leaders must track overall adoption rates and how users interact with different features. For example, a project management platform might monitor its AI-powered task prioritization feature’s weekly active users (WAUs), correlating this data with upsell rates for premium tiers.

How to measure adoption

Product adoption can be expressed over time by the number of monthly active users (MAU), weekly active users (WAU), or daily active users (DAU). You can also measure product adoption as a rate relative to new user signups for a given period. How you measure product adoption will largely depend on what it means to be an active product user. If your software is one customers should access daily, a DAU metric might make the most sense.

Stickiness

Stickiness—the measure of habitual product usage—has evolved beyond tracking daily logins. AI features like personalized recommendations or predictive insights now deepen user engagement within an app. For example, a fintech app using AI to forecast spending habits may see users return daily to check budget predictions (directly impacting retention).

Modern stickiness metrics must isolate AI-driven engagement, comparing retention rates between users who adopt these features and those who rely on traditional functionalities.

How to measure stickiness

Stickiness is reflected in the ratio of active users at a specific point in time against active users over a period of time. For example, you can see the stickiness of your software on a given day using DAU and MAU:

Essential KPIs, enhanced

AI is transforming how product leaders track and interpret KPIs, making product success more measurable, predictive, and actionable than ever before. From real-time user insights to automated trend analysis, AI-driven analytics empower product managers to move beyond static dashboards and embrace dynamic, adaptive decision-making.

As AI continues to evolve, those who integrate its capabilities into their KPI frameworks will gain a significant competitive edge, ensuring their products meet current market demands and anticipate future trends.

Pendo helps businesses measure and improve on the KPIs discussed throughout this guide. Use product analytics and replay data to understand user behavior and spot friction, and AI-powered analysis to identify trends. By combining this quantitative data with qualitative feedback, teams get a complete view of the customer experience. They can then easily take action with no-code in-app guides that increase product adoption, boost user engagement, and grow revenue.

Ready to see how Pendo can help you measure and take action on key product KPIs? Get a personalized demo today.