Glossary

Every term the playbook leans on, defined once. Terms are linked from the chapters, so you never have to leave a chapter to find out what something means.

A/B test: Splitting users between two versions and comparing the result. The textbook version needs more traffic than most startups have, which is an argument for running a smaller test rather than for running none.
Activation: The point where a new user reaches the first moment of real value in your product, rather than merely creating an account. Every product defines its own activation event, and that definition decides what your activation rate means.
Anonymous ID: The random identifier an analytics tool gives a visitor before it knows who they are. It survives until identify runs, and everything done before that point hangs off it.
Business analytics: Analytics answering questions about business objectives rather than user behaviour: revenue by country, cohorts against targets, reconciling revenue with finance data. It sits alongside product analytics under data analytics.
Causation: One thing making another happen. Your existing data cannot show it; an experiment, a look at the users who converted without the behaviour, or a conversation with them can.
Correlation: Two things showing up together. It is where an insight starts, not where it ends: the users who do the thing may simply be your most motivated ones.
Counter metric: A metric watched alongside a north star to catch gains taken from somewhere else. If sign-ups rise while activation falls, the counter metric is what tells you the bump is not progress.
Counterfactual: A made-up period built to answer a what-if: what this week would have looked like if only one thing had changed. It is the device that separates a rate effect from a mix effect.
Data analyst: The umbrella title for someone who analyses data. It says nothing about which kind: a strong data analyst can reconcile revenue perfectly and still be a weak product analyst, so hire for the kind of analytics you need.
Data analytics: The broad term: taking data, pulling insight out of it and understanding what is happening. Product analytics and business analytics both sit under it.
Data warehouse: A separate database built for analysis rather than for running the product. It is where every source lands together, which is what makes it the number you trust when two tools disagree.
Event: One thing a user did, recorded with a name and a set of properties. A sign-up, a click, a purchase. If it cannot be attributed to a specific user, it is not a product analytics event.
Funnel: The steps a user goes through in order. Every product is one, and so is every feature inside it. Counting who reaches each step is what shrinks a whole-product problem to a single step.
Holdout: A group deliberately kept away from a campaign so the difference between them and everyone else shows what the spend actually caused.
identify: The call that tells your analytics tool a visitor is a particular signed-up user. It runs at login and signup, after authentication, with your own internal user ID.
Identity stitching: Joining all of one person's activity into a single profile, whatever device or browser it arrived from. Get it wrong and every per-user number is wrong in the same direction.
Incrementality: The share of an outcome that only happened because of the spend. An incremental sign-up is one that would not have arrived anyway through the product, content or word of mouth.
MCP: A connector that lets an AI tool read directly from another tool, such as your analytics platform or your database. It is what makes an agent able to answer questions about your own numbers.
Media mix modelling: Estimating what each marketing channel contributed from weekly spend and outcome data rather than from tracked clicks. It is how influencer, offline and brand spend get measured at all.
Mix effect: The part of a change in a rate caused by volume moving between segments rather than by any segment performing differently.
MMP: A mobile measurement partner: the service that attributes an app install to the campaign that caused it, which UTM parameters cannot do because they do not survive the app store.
North star metric: The one measurable number that matches what the business is trying to do right now, and whose movement you can always explain.
Product analyst: An analyst who works on product analytics: how users move through the product, what you can measure along the way, and how to read those numbers. The question they answer best is "where are users dropping off?"
Product analytics: The analysis of what users do inside your product. The test is whether you can point at a row of data and name the user who did it.
Property: A piece of detail attached to an event or a user: which plan, which country, which surface the click came from. Properties are what let you break a number down far enough to act on it.
Qualitative analysis: The people half of the work: watching session recordings and talking to users. It tells you why something is happening, which the numbers never will.
Quantitative analysis: The numbers half of the work: charts, funnels, conversion rates. It tells you what is happening, and how widespread it is.
Rate effect: The part of a change in a rate caused by segments performing differently, with the mix between them held still.
Raw tables: Your data exactly as it lands, before anyone has cleaned, joined or organised it. Querying it directly costs compute on every question and time on every join.
Retention: The share of users who come back. Bounded retention counts people who returned on exactly that day; unbounded counts that day or any day after, and the two give very different numbers.
Segment: A group of users who share something: a country, an acquisition channel, a signup method, a first-session length. Splitting a metric by segment is what makes it actionable.
Semantic layer: A written guide between your data and anyone querying it, human or AI, saying what each table or event contains, how things join, and what each core metric actually calculates.
Sense check: Holding a number roughly against what you already know about the business, to see whether it is even in the right ballpark, before anyone acts on it.
Surface: Any place in the product an event can fire from: the home page, the pricing page, a popup, the app. The broken ones are usually the surfaces nobody checked.
Tracking plan: One sheet listing every event you track, when it fires and the properties it carries. Written before implementation, owned by someone who is not a developer.
UTM: Tags added to the end of a link that record where a click came from. Enough on their own for a web-only product, and useless the moment an app install sits in the path.
Vanity metric: A number that feels good when it goes up but does not tell you why it moved or what to do next.