How to See What Users Actually Do: The Dot Plot Method

Key Insights

• Aggregate metrics hide individual behavior: DAU/MAU numbers mask how real users interact with your product and whether they actually find value
• The dot plot visualizes user patterns: A simple 2D grid (users vs. time) shows which features drive engagement and reveals onboarding or retention problems
• Early warning system for churn: Spotting low activation or inconsistent usage across paying customers can prevent contract loss before renewal
• Works at any scale: From 10 users to billions—just sample strategically when you reach massive scale
• Pairs with cohort retention: Use cohort curves to see if users stay; use dot plots to see how they use your product

Understanding the Problem With Aggregate Data

Most founders rely on Daily Active Users (DAUs) and Monthly Active Users (MAUs) to gauge product health. The problem? These numbers lump all users together, hiding critical truths about individual behavior.

Even with just 10–20 users, monitoring logs becomes overwhelming. Aggregate graphs can trend upward even when users aren't genuinely enjoying your product—a misleading signal that everything is working. You need a way to see what each individual user is actually doing.

What Is the Dot Plot?

The dot plot is a two-dimensional grid that puts individual user understanding within reach. Think of it like a spreadsheet:

• Rows = individual users (Dave, User 2, User 5, etc.)
• Columns = time periods (typically days, though you can adjust based on your product's rhythm)
• Dots = a specific value-generating event (listening to a song, sharing a photo, processing an invoice)

For a music streaming app, you'd mark a blue dot on the day a user listens to a song. You can add symbols—like a ring around the dot—to mark the day they first onboarded.

As you fill the grid, patterns emerge that aggregate charts never reveal. You might notice "Dave" listens mostly on weekdays while "User 2" and "User 5" prefer weekends. "User 4" might try once and disappear—a pattern repeated across rows that signals an onboarding problem.

Real-World Example: Why a B2B Contract Failed

One YC company landed an $80,000 annual contract with a prominent customer. They onboarded 10 seats. Later, the customer churned. A dot plot would have shown the warning signs:

• Only 3 of 10 seats were activated
• Those 3 users engaged less than twice per week
• Usage was sporadic and inconsistent
• The champion who pushed the deal internally had left the company

When the new decision-maker arrived, they asked "Why are we paying for this?" and chose not to renew. The company could have spotted this risk early and intervened—but aggregate numbers (e.g., "contract signed") hid the reality that most users weren't gaining value.

Using Dot Plots at Scale

When you have over a billion users, the solution is sampling. Print dozens of dot plots representing different user segments:

• iOS users in France
• U.S. web users earning $80k+
• Users from a specific onboarding cohort

Sit with your team, review these visualizations, and ask: "What patterns do we see? What's different about these users?" Your brain naturally identifies anomalies and behavioral clusters that algorithms might miss.

This approach originated from Max Levchin (PayPal co-founder), who used similar visualizations to spot fraud. Human observers couldn't explain why they saw fraud, but they could say "that thing happening there is different and probably fraud." Dot plots work the same way—your intuition catches patterns hidden in raw data.

Common Mistakes With Dot Plots

Charting the wrong event: Don't track "app opens" or "logins." These actions don't prove users derive value. Instead, track events that represent real value—songs listened to, photos shared, invoices processed.

Using time periods that are too broad: Don't collapse data into weeks or months. Use daily granularity (or hourly, depending on your product) so you see actual usage pacing. Wider time windows hide critical churn signals.

Dot Plots + Cohort Retention Curves

Cohort retention curves answer one question: Do users stay? Dot plots answer another: How do they use it?

Cohort curves are essential—they show aggregate retention trends. But they don't reveal why users churn or which features drive engagement. Dot plots provide those insights. Together, they form the most powerful toolkit for understanding user behavior:

• Cohort retention: Big-picture health signal
• Dot plots: Granular, actionable insights for product decisions and user conversations

Why Dot Plots Scale

The beauty of dot plots is simplicity. There's no complex math—just log parsing and 2D grid visualization. Modern AI coding tools can build one in 10 minutes. For early-stage founders with under 100 users, a dot plot can be your entire dashboard.

Conclusion

Aggregate metrics feel safe but mislead. The dot plot cuts through noise to reveal how real users behave—who engages consistently, who tries once and vanishes, whose usage is deteriorating. For founders, it's the bridge between "we have users" and "our users actually love this." Pair it with cohort retention data and you'll catch problems early, build the right features, and ask smarter customer questions before contracts slip away.

---

Original source: See How Customers Actually Use Your Product

powered by osmu.app