Dot Plots: How to Exactly See What Your Users Are Doing

Aggregate user metrics hide critical behavioral insights behind rising trendlines. This chapter explores Dot Plots—a powerful visual analytics tool introduced by Max Levchin at PayPal and refined at companies like Google—which allows founders and product teams to visualize individual user actions over time, detect hidden usage patterns, prevent enterprise churn, and evaluate whether people are actually getting real value from their product.

1. The Core Problem with Aggregate Metrics

Many founders rely entirely on macro-level usage charts like Daily Active Users (DAUs) or Monthly Active Users (MAUs). While these metrics give a high-level view of scale, they suffer from fatal analytical flaws:

• The Aggregation Blind Spot: They clump every user together, erasing individual context, usage frequency, and behavioral variety.
• The Vanity Growth Illusion: If a product has a continuous influx of new signups, aggregate charts will naturally slope upward to the right—even if older users are completely miserable, churning out, and getting zero value.
• Total Ignorance of "How": Aggregate charts cannot answer crucial product-market fit questions: How are they interacting with the product? What features are they using? What is their true rhythm and engagement frequency?

Note: As Dave emphasizes, while cohort retention curves tell you if users stick around over time, aggregate metrics completely fail to tell you what those users are actually doing inside your product.

2. What is a Dot Plot? Core Mental Model

A Dot Plot is a two-dimensional spreadsheet-like grid that maps individual users against discrete time intervals. It bridges the gap between raw user logs and human pattern recognition.

Structure of a Dot Plot

• Rows: Each individual user gets their own dedicated row in the grid.
• Columns: Each column represents a discrete time period (typically Days work best for most software products).
• Cells: A dot is placed in a cell whenever that user performs a designated value-creating action on that specific day.

3. Choosing the Right Value Event

A foundational rule of creating effective dot plots is choosing the correct action to plot. Founders often make the mistake of tracking trivial events because they are easy to measure.

4. Pattern Recognition: What Dot Plots Reveal

When you visualize data at individual density, human pattern-recognition kicks in to reveal stark behavioral realities that summaries mask.

Key Insights Unlocked by Dot Plots:

• Usage Segmentation (Weekday vs. Weekend): You might discover distinct user archetypes—such as "weekday office power users" versus "weekend casual listeners"—allowing you to redesign or target features more accurately.
• Onboarding Drop-Offs: If you see rows (like User #4) with a dot on Day 1 (circled to denote signup day) and zero subsequent dots, it signals an immediate onboarding friction or failure.
• Feature-Driven Behavior: Upgrading dot plots with custom symbols (e.g., an "S" for search usage, a "P" for joining a public playlist) lets you correlate specific feature adoption with sustained retention.

5. Visualizing the Contrast: Aggregate DAU vs. Dot Plots

To understand why aggregate metrics lie while dot plots tell the truth, consider a product with small-scale usage.

The Aggregate DAU View

If you looked strictly at a daily active user count chart, you would see a flat, stagnant trendline:

DAU Count:
Day 1: 2 users
Day 2: 3 users
Day 3: 2 users
Day 4: 2 users
Day 5: 2 users
Day 6: 2 users
Day 7: 1 user
Day 8: 0 users
Day 9: 1 user

Conclusion from DAU: "We are flatlining. No growth is happening."

The Dot Plot Reality

When you look at the underlying dot plot for those exact same users, you instantly realize:

• Certain users are heavily engaged Monday through Friday in office environments.
• Feature dependencies become clear (e.g., users who engage with playlist features stay active multiple days in a row).
• You understand user context and lifestyle without needing complex data science models.

6. Scaling Dot Plots: From 10 Users to 1 Billion

Dot plots are not just for early-stage startups with 20 beta testers. They scale effectively to massive enterprises when sampled correctly.

Scaling Strategies

• Early Stage (< 100 users): Plot every single user on a single screen or printout. This should act as your primary operational dashboard.
• Enterprise / Mass Scale (Millions to Billions): At Google (where Dave used this with over 1 billion users), teams take representative cohorts and user samples and print or render targeted dot plots:
  • Sheet 1: iOS users based in France.
  • Sheet 2: US-based web users earning over $80k/year.
• Teams spend hours analyzing these printed samples together to extract qualitative product intuition.

7. Saving Enterprise Deals Before They Churn

Dot plots are equally vital for B2B and enterprise SaaS companies where large corporate contracts mask internal engagement rot.

Real-World B2B Scenario

• The Situation: A prominent enterprise brand signs an annual contract worth $80,000, purchasing 10 user licenses.
• The Aggregate Illusion: The company paid, the invoice cleared, and management assumes everything is healthy.
• The Dot Plot Reality Check:
  CODE

  Copy

  License 1 (Departed Sponsor): • • (Active early, then quit)
  License 2 (Inactive Employee):  
  License 3 (Intermittent User):  •     •
  Licenses 4-10:                  (Zero activity)
  

• The Root Cause: An influential stakeholder in the company bought the software, championed it, and then left the organization. A successor arrived, saw zero active utilization, and immediately flagged the contract for cancellation at renewal.
• The Saving Grace: By monitoring B2B dot plots, the account management team would have seen that only 3 out of 10 licenses were active, intermittent usage was dropping, and renewal was at risk months before cancellation occurred.

8. Common Pitfalls and How to Avoid Them

When implementing dot plots, avoid these two major traps:

1. Plotting the Wrong Event: Never plot page views or logins. Always anchor dots to actions that deliver genuine end-user value.
2. Choosing Too Wide a Time Horizon: Do not plot by weeks or months early on. Keep your time granularity at the daily level (or finer) until you achieve massive scale.

9. Summary and Key Takeaways

• Aggregate metrics conceal truth: DAUs and MAUs blend all users together, often sloping upward due to new signups even while core users churn.
• Dot plots reveal individual behavior: By structuring a 2D grid of users (rows) against time intervals (days in columns), you can visually map actual product usage.
• Choose core value events: Only plot actions that represent true user value creation (e.g., sharing a photo, listening to a song), not vanity metrics like logging in.
• Uncover hidden patterns: Dot plots instantly expose weekday versus weekend usage splits, onboarding drop-offs, and feature-driven engagement loops.
• Scale through sampling: As your user base grows from dozens to billions, use targeted sampling (e.g., specific OS platforms, regions, or demographics) to maintain visual clarity.
• Save B2B contracts early: In enterprise software, dot plots reveal unactivated licenses and dying engagement long before a corporate client cancels their annual renewal.
• Zero complex math required: Dot plots are simply visual representations of your existing database logs—easily generated by modern AI and query tools in minutes.
• Combine with cohort retention: Use cohort retention curves to track high-level macro retention, and use dot plots to understand the micro-behavioral "why" behind those retention numbers.