Funnel Analysis: How Specs Quietly Change Conversion
Funnel analysis conversion rates shift with ordering, windows, and entry rules. See how one dataset yields four honest numbers, plus a spec checklist.
All articles
24 published articles, newest first.
Product Analytics Benchmarks 2026 by Category
Product analytics benchmarks for 2026: ~37% activation, 31–33% B2B SaaS stickiness, and why definitions decide every retention number you compare.
Cohort vs Segment vs Audience: Untangling Three Terms
Cohort vs segment vs audience confuses PMs, marketers, and analysts. Learn what anchors membership in each, with a worked example and a decision flowchart.
Acquisition vs. Behavioral Cohorts in Cohort Analysis
Acquisition and behavioral cohorts answer different questions in cohort analysis. Learn which axis fits your question before you draw the retention chart.
Identity Resolution: Stitching Users Without Corrupting Data
Identity resolution links anonymous and identified users, but retroactive merges can inflate new-user counts. How the tools differ and where stitching breaks.
Sessionization Explained: What Counts as a Session
Sessionization groups events into visits using a 30-minute timeout. See how GA4, Amplitude, and Mixpanel differ and why session counts diverge.
How to Build a Tracking Plan (with a Free Template)
Build an analytics tracking plan that stays clean: define your first 20 events, use properties over new events, and enforce it. Free five-column template included.
Event Taxonomy 101: The Object-Action Framework
An event taxonomy that survives growth: the object-action framework for naming events, snake_case rules, a good-vs-bad table, and a cleanup path.
Simpson's Paradox: When Your Aggregate Numbers Lie
Simpson's paradox is when a rate rises in every segment yet falls overall. Here's a worked mix-shift example and a rule to catch it before you panic.
Sample Ratio Mismatch (SRM): The Silent A/B Test Killer
Sample ratio mismatch quietly invalidates A/B tests. Here's the chi-square SRM check with real numbers, the p-value threshold to use, and ranked root causes.
Product-Qualified Leads: Scoring, and PQL vs. MQL
A product qualified lead is a user whose in-product behavior predicts conversion. Here's a PQL vs MQL table, a scoring rubric, and the PQA distinction.
Leading vs. Lagging Indicators: Metrics You Can Move
Leading indicators predict outcomes; lagging ones confirm them. A practitioner's method for deriving movable leading indicators from a lagging metric.
Vanity Metrics vs. Actionable Metrics: A Two-Question Test
Vanity metrics look great and change nothing. Here's a two-question test to tell them from actionable metrics, plus 20 rewrites for your dashboard.
Growth Accounting and the Quick Ratio, Explained
Growth accounting decomposes MAU into new, resurrected, and churned users. Here's the quick ratio formula, a worked example, and how to read the number.
Google's HEART Framework: Measuring UX With Product Data
The HEART framework turns UX quality into trackable numbers. How Google's Happiness, Engagement, Adoption, Retention and Task-success grid works, with an example.
Feature Adoption Metrics: Breadth, Depth, Time-to-Adopt
Feature adoption measured three ways: breadth (% of users), depth (frequency), and time-to-adopt. Formulas, benchmark bands, and the adoption funnel.
Power Users and the L28 Engagement Model
Power users and the L28 engagement model explained: read the days-active histogram, pick L7 vs L28, and stop letting averages hide a bimodal reality.
The Sean Ellis Test: Measuring PMF with the 40% Rule
A practical guide to the product-market fit survey: the exact wording, the 40% threshold, who to ask, and how to mine the 'very disappointed' answers.
Time to Value: How to Measure It and Shorten First Value
Time to value (TTV) is the gap between signup and first real payoff. Here's how to measure it as a duration between two events, plus benchmark ranges.
Retention Curve Shapes Decoded: Smiling, Flattening, Decaying
What a retention curve's shape tells you: smiling, flattening, and decaying curves explained, plus why the plateau height beats day-1 retention.
What Is the Aha Moment? Find Yours With Data, Not Guesses
The aha moment is the early action that predicts retention. Here's a data-driven method to find yours, plus the causation trap most guides skip.
The DAU/MAU Ratio: Benchmarks and Why 20% Isn't Universal
The DAU/MAU ratio converts straight to visit frequency, so 20% isn't a universal target. Formula, category benchmarks, and a natural-cadence correction.
AARRR Metrics, Redrawn as a Loop for Product-Led Growth
The AARRR metrics framework still works, but not as a funnel. Here's how to re-sequence activation and referral for freemium and self-serve products.
North Star Metric vs. Input Metrics: The Metric Tree
A practical guide to the North Star Metric vs. input metrics: how to build a metric tree teams own, act on, and wire to a live dashboard.