Glossary of terms used in the article

Question

Accepted Answer

Product metric &#8211; a measurable indicator showing whether the product delivers value to the user and the organization. A good product metric changes team decisions instead of just looking good in a presentation.
Lagging indicator &#8211; a metric that confirms the result after the fact. Revenue, NPS, 12-month retention, or contract renewals talk about the past, so they are not suitable for steering the product in real time, only for validating whether the strategy worked.
Leading indicator &#8211; a metric whose change comes before the change in a business indicator. It is strongly correlated with a lagging indicator, but measurable every day. It gives the team a reaction window counted in weeks, not quarters.
Proxy metric &#8211; an indirect indicator, not identical to the business goal, but strongly enough correlated with it that its movement predicts movement in the goal. The classic example: &#8220;10 friends in 7 days&#8221; as a proxy for retention at Facebook.
North Star Metric (NSM) &#8211; the one metric that best captures the moment when value is delivered to the user. In Spotify it is minutes listened, in Airbnb nights booked, and in B2B SaaS the number of key actions performed in a week by an active user.
Activation rate &#8211; the percentage of new users who complete a key action in a defined period that shows they experienced value (for example adding the first project, inviting the first team member, or completing onboarding).
Cohort retention &#8211; the percentage of users from a given cohort (for example those who registered in the same week) who are still actively using the product after N days, weeks, or months. The strongest signal of product-market fit.
Time-to-value (TTV) &#8211; the time it takes a new user to reach the first experience of value in the product. The shorter the TTV, the higher the activation rate and cohort retention.
Vanity metric &#8211; a metric that looks good in a report but does not change any decision (for example &#8220;number of website visits&#8221; or &#8220;total registrations since the dawn of time&#8221;). In Lean Startup, Eric Ries contrasts it with an actionable metric &#8211; one that reveals the cause and shows what to do differently.
Outcome (in product) &#8211; a measurable change in user behavior caused by a release, for example &#8220;the user performs X 40% faster.&#8221; The opposite of output, which is just the &#8220;delivered feature.&#8221;
Product hypothesis &#8211; a written assumption about what change in user behavior or business indicator a given implementation will produce. It includes a specific metric and direction of change, so after release it can be verified instead of debated as opinion.
Cycle time &#8211; the time from when work starts on a task to when it is released. Shorter cycle time means faster feedback loops and faster validation of product hypotheses.
Lead time &#8211; the time from when a need is identified (for example when a ticket is added to the backlog) to when it is released. It includes cycle time plus waiting time in the queue.
Throughput &#8211; the number of tasks completed by the team in a unit of time (for example per sprint). Together with cycle time, it gives a picture of delivery predictability.
WIP (Work in Progress) &#8211; the number of tasks started at the same time and not yet finished. High WIP usually correlates with longer cycle time and lower throughput, in line with Little&#8217;s Law.
MRR (Monthly Recurring Revenue) &#8211; recurring monthly revenue in a subscription model. Together with churn rate (the percentage of customers leaving in a given period) and CAC (customer acquisition cost), it is the foundation of business metrics in SaaS.
NPS (Net Promoter Score) &#8211; a measure of customers&#8217; willingness to recommend a product, calculated as the percentage of promoters minus the percentage of detractors. A classic lagging indicator &#8211; it tells you how the product is perceived, but only after the fact.
Technical debt &#8211; the accumulated cost of technical shortcuts taken in the past (missing tests, inconsistent architecture, workarounds instead of a target solution). It does not send invoices, but it causes estimates to grow and makes the team spend more and more time firefighting instead of building.
Product-market fit &#8211; a state in which the product solves a real problem well enough and for a large enough group that users come back on their own, and sales stop requiring persuasion. It can be measured, among other things, through cohort retention and NPS.

Pragmatic Blog

product entries

Revenue, listening minutes, and a question about God. Which product metrics really show that a product is moving in the right direction

We're here to become the world's most effective software company.

Pragmatic Blog

product entries

Revenue, listening minutes, and a question about God. Which product metrics really show that a product is moving in the right direction

Newsletter

We're here to become the world's most effective software company.