Product management glossary

App store review analysis is the systematic collection and classification of public reviews and ratings from the Apple App Store and Google Play. Product and support teams use it to spot defects, regressions, and feature complaints as they emerge. Reviews carry metadata such as star rating, app version, country, and device. But the review text itself is unstructured data, so topic and sentiment classification has to happen before the volume becomes readable. Example: a release-over-release view surfaces a login failure concentrated in a single app version, which prompts root cause analysis. Booking.com uses OnClarity to spot bugs in customer feedback and get the right engineers the information they need.

A feature request is a customer-originated ask for new or changed product functionality. These arrive through support tickets, sales calls, app reviews, NPS comments, and community forums. Product teams rarely act on raw counts alone. Requests are weighted by frequency, revenue at risk, and the segment asking, since one enterprise account’s blocker can outweigh a hundred casual mentions. A stated request often masks the underlying job the customer is trying to get done, so teams probe past the literal ask before scoping work. At volume, near-duplicate requests worded differently need consolidation to avoid double-counting demand. OnClarity’s Voice of Customer classifies requests by topic across 100+ sources automatically. Booking.com uses this feed to spot bugs in feedback quickly and get the right engineers what they need to fix them.

Jobs To Be Done (JTBD) is a framework holding that customers “hire” products to make progress in a specific situation. This is different from buying products for their features or demographics. Clayton Christensen popularized the concept in “The Innovator’s Solution,” defining a job as an action that enables progress in a given context. Tony Ulwick later developed Outcome-Driven Innovation, a structured method for identifying customers’ desired outcomes. A job statement typically names the situation, the underlying motivation, and the expected outcome, rather than a persona or age bracket. For brand health, this matters because customers associate brands with the jobs those brands do well. That association drives which brand comes to mind when the need arises.

A north star metric (NSM) is the single measure a company uses to capture the value its product delivers to customers. It is chosen because it predicts long-term revenue better than any other number available. A valid NSM passes three tests: it reflects real customer value, it moves before revenue does, and a team can actually influence it through their work. Airbnb tracks nights booked. Spotify tracks time spent listening. The common failure mode is choosing a volume metric like signups or pageviews, which can rise while retention quietly falls. Beneath the NSM sit input metrics, the specific actions that drive it. These give teams a measurable, causal chain from daily work to the number that matters.

Product analytics is the measurement of how people behave inside a product. It is built on event data: the individual actions users take, such as clicks, feature use, or completed steps, as opposed to page traffic. Standard analyses include funnels (where users drop off in a sequence), retention cohorts (whether users return over time), feature adoption (which capabilities get used and by whom), and paths (the routes users take through a product). This sets it apart from web and marketing analytics, which measure acquisition: traffic sources, campaigns, and conversions. The limit is that behavioral data shows what people did, not why. That is why teams pair it with feedback data, such as OnClarity’s Voice of Customer, to explain the patterns event data surfaces.