Analytics & Measurement

Busted

• Acquisition channelThe broad bucket for how visitors arrived, such as organic search, paid search, email, referral, social, or direct.
• Activation eventThe first action that proves a new signup actually got something out of your product. The 'aha moment', written down as an event.
• Anomaly detection / alertingAn automatic warning when a metric moves outside its expected range, so a real spike or drop is noticed before the next dashboard check.
• attribution modelThe rule for splitting credit for one sale across every ad, email, and link the buyer touched on the way there.
• Attribution windowThe time limit for credit: how long after an ad, email, or link a conversion can happen and still count toward that touchpoint.
• Bounce rate vs. engagement rateVisits where someone did nothing and left. In GA4 it is exactly the flip side of engagement rate; older tools defined bounce differently.
• CDP (customer data platform)A central hub that collects customer events once, then routes them to analytics, marketing, support, and warehouse tools.
• Churn rateThe share of customers or revenue you lost in a period. Retention's evil twin: if 95% stayed, 5% churned.
• Cohort analysisGrouping people by when they joined, then comparing how each group behaves over time instead of blending everyone into one number.
• Consent mode / privacy-preserving measurementMeasurement that respects a person's tracking choice, then uses limited signals and aggregate modeling to estimate the missing totals.
• Control groupThe randomly assigned group that keeps the old experience, giving you a fair baseline for what would have happened without the change.
• Conversion eventA tracked action promoted to the finish line, such as a completed purchase or booked demo, so analytics can count success.
• Conversion funnelThe ordered steps of a flow (visit, signup, pay) measured for survivors at each stage, so you can see exactly where people fall off.
• Conversion rateThe percent of people who did the thing you wanted, out of everyone who had the chance. Completions divided by eligible people.
• Cross-domain trackingKeeping one visitor and session intact when a journey crosses related domains, so checkout does not look like a brand-new referral.
• DashboardA saved screen of the same charts in the same places, checked on a schedule, so a change in the numbers jumps out.
• Data freshness / latencyThe wait between something happening and its appearance in a report, including collection, processing, and dashboard cache time.
• Data layerA shared, structured queue where the app publishes tracking facts once, so analytics tags stop scraping the page or inventing their own values.
• DAU/WAU/MAUCounts of unique people who did something in your product in the last day, week, or month. What counts as 'something' decides everything.
• Dimension vs. metricDimensions are the labels you slice by (country, device, plan). Metrics are the numbers you measure (users, revenue, rate).
• Event deduplicationMaking retries and repeated deliveries count as one event, so one purchase, signup, or click does not appear two or five times.
• Event taxonomyThe naming rules for your analytics events, so 'signup', 'sign_up', and 'Signed Up' never become three different metrics.
• Event trackingLogging named user actions with details attached, so you can count what people actually did instead of guessing from pageviews.
• Experiment sample-size calculationThe pre-test calculation of how many users each variant needs to reliably detect the smallest change worth acting on.
• First-party analyticsAnalytics served from your own domain into data you own, instead of shipping visitors off to someone else's tracker. Fewer blockers eat it.
• Funnel analysis vs. path analysisFunnels measure survival through steps you chose; path analysis reveals the routes people actually took when you do not know the steps yet.
• Funnel leakageThe one funnel step where far more people disappear than expected, pointing to a likely bottleneck, bug, or tracking gap.
• Funnel step conversion rateThe percent of people who make it from one funnel step to the very next one, instead of all the way through to the end.
• Funnel windowThe time a person is allowed to take from the first funnel step to the last before the report counts them as a drop-off.
• Guardrail metricThe metric you watch so a win does not hide a loss: signups went up, but did refunds, load time, or support tickets go up with them?
• heatmapA color wash over a page screenshot showing where people clicked, hovered, or stopped scrolling. Red is busy, blue is dead.
• Holdout groupA slice of eligible users deliberately left on business as usual, so their results reveal the baseline without the campaign or rollout.
• Identity resolution / user stitchingGluing someone's anonymous browsing before signup onto their real account after it, so the data reads as one person, not two.
• Incrementality testA holdout test that asks whether a campaign created extra conversions, not whether it appeared somewhere before conversions that would happen anyway.
• Last-click / first-touch attributionTwo all-or-nothing rules: give a conversion entirely to the final touchpoint, or entirely to the one that started the journey.
• Lifecycle stageA label on each user for where they are right now: new, current, resurrected, churned. It turns one active-user number into a story.
• Metric definition driftWhen two reports use the same metric name but different formulas, filters, time zones, or identity rules, so both produce different answers.
• Multi-touch attribution (MTA)Splitting one conversion's credit across the ads, emails, searches, and links that appeared along the path, not just the last one.
• North Star metricThe single number a whole team steers by, chosen because it goes up only when customers actually get value.
• Novelty effectA temporary rise or dip caused by a change being unfamiliar, which fades after people stop exploring it or learn the new behavior.
• Peeking problemThe false-winner problem caused by checking an ordinary test repeatedly and stopping the first time the result turns significant.
• Power user curve (L30)A histogram of how many days each person used the product in the last 30, separating occasional users from the heavy regulars.
• Product analyticsAnalytics that measures what people do inside your product, action by action, instead of which pages they loaded.
• Retention curveThe line showing what percent of a signup group is still active N days later, and whether it flattens out or slides to zero.
• SegmentA saved filter that names a subset of your users, like 'paid accounts on mobile', so any chart can be re-run for just them.
• Server-side trackingSending confirmed events from the backend, where ad blockers, closed tabs, and flaky browser requests cannot swallow the outcome.
• session replayA replay of one real visit, cursor and clicks and all, so you can watch exactly where a person got stuck and gave up.
• Source / mediumThe paired traffic label that says who sent a visit and how: google / organic, newsletter / email, or partner / referral.
• statistical significanceThe check on whether your winning variant really won or just got lucky. A low p-value means the data fit badly with 'no difference'.
• Stickiness ratioThe share of monthly or weekly active users who show up on a typical day, revealing whether use is occasional or habitual.
• Survivorship biasThe mistake of learning only from users who stayed, while the people who quit or failed are missing from the evidence.
• Tag management systemA controlled container for shipping analytics scripts and pixels through rules and versions, often without a full application release.
• Time to value (TTV)How long it takes a new user to get their first real benefit, measured from signup to the moment the product pays off.
• Tracking planThe document that pins down exactly which events you fire, what properties they carry, and what each one means. The cure for event-name chaos.
• User properties vs. event propertiesTraits that stick to a person (plan, country, signup date) versus details attached to a single action (button, price). Two different buckets.
• User vs. session vs. pageviewThree ways of counting the same traffic: a person, one visit by that person, one page they loaded. Same day, three numbers.
• UTM parametersTags stapled onto a link (?utm_source=...) so analytics can tell which post, ad, or email actually sent that visitor.
• Vanity metricA number that looks impressive but has no clear decision behind it, like total downloads with no activation, retention, or revenue context.

The territory

30 core terms mapped for this field, ranked by how often builders reach for them. Each one is a future entry. Want to bust one? One entry, one file, one pull request.

• event trackinglogging named user actions with attached properties"record when someone clicks the thing" · "log what people do"
• conversion funnelordered steps measured for drop-off between each stage"where people fall off" · "the step-by-step drop chart"
• conversion ratepercentage of eligible users completing a desired action"how many actually sign up" · "what percent convert"
• user vs. session vs. pageviewperson, visit, and viewed page counted as different units"is that people or visits" · "why are visits higher than users"
• cohort analysisgrouping users by signup period, comparing behavior over time"compare people who joined in March vs April" · "the triangle chart"
• retention curvepercent of users still active N days after signup"how many come back" · "the curve that flattens or doesn't"
• North Star metricsingle number best proxying delivered customer value"the one number that matters" · "our main metric"
• attribution modelrules assigning conversion credit across touchpoints"which ad gets the credit" · "what actually caused the sale"
• UTM parametersURL query tags identifying traffic source, medium, campaign"the ?source= stuff on links" · "campaign tags in the link"
• session replayrecorded playback of a real user's screen and cursor"watch someone use my site" · "screen recording of visitors"
• heatmapcolor overlay showing where users click, move, or scroll"the red-blob click map" · "where people look"
• A/B testrandomized comparison of two variants on one metric"try two versions and see which wins" · "split test"
• statistical significanceconfidence the observed difference isn't random noise"is this result real or luck" · "enough data yet"
• event taxonomynaming and property scheme keeping events consistent"rules for naming my events" · "stop the mess of event names"
• product analyticsbehavior-level analytics inside the product, not just pages"analytics for in-app actions" · "not just pageviews"
• dimension vs. metriccategorical breakdowns versus numeric measurements used in reports"the labels vs the numbers" · "rows vs values in a report"
• activation eventfirst meaningful action proving a new user got value"when they actually start using it" · "the aha moment" · "activation"
• time to value (TTV)elapsed time from signup to first real benefit"how long before it's useful" · "time until they get something out of it"
• churn rateshare of users or revenue lost in a period"people leaving" · "how many cancel"
• funnel step conversion ratepercent advancing from one funnel stage to next"how many made it to the next step"
• identity resolution / user stitchinglinking anonymous visits to a known user after login"connect the visitor to their account later"
• dashboardfixed panel of saved charts tracking core metrics"the screen with all the numbers"
• segmentsaved filter defining a reusable subset of users"just the paying users" · "a saved group"
• DAU/WAU/MAUactive-user counts per day, week, month"how many people use it daily" · "active users"
• lifecycle stagelabeling users as new, active, dormant, resurrected, churned"which stage a user is in" · "new vs dormant users"
• user properties vs. event propertiestraits attached to a person vs. to a single action"person info vs action info" · "what goes on the user record"
• bounce rate vs. engagement ratesingle-action visits vs. meaningfully engaged visits"people who leave right away" · "bounce rate vs engaged sessions"
• first-party analyticstracking from your own domain, without third-party cookies"privacy-friendly analytics" · "cookieless analytics"
• tracking plandocumented spec of every event and its properties"the spreadsheet of what we track" · "event schema"
• guardrail metricsecondary metric watched so a win doesn't break something"make sure we didn't hurt something else"

Deeper in the field

• conversion event tracked action designated as a desired outcome
• acquisition channel traffic grouping such as organic, paid, direct, or referral
• source / medium labels identifying traffic origin and delivery method
• multi-touch attribution (MTA) splitting credit across all touchpoints, not just last click
• last-click / first-touch attribution crediting the final or initial touchpoint entirely
• attribution window period after a touchpoint during which conversions receive credit
• cross-domain tracking preserving user and session identity across related domains
• data layer structured browser data supplying consistent values to analytics tools
• tag management system tool for deploying analytics tags without code releases
• server-side tracking sending events from backend, bypassing browser blockers
• CDP (customer data platform) central hub collecting and routing events to tools
• event deduplication preventing double-counted or spammed events
• control group unchanged comparison group estimating a variant's causal effect
• holdout group users deliberately excluded to measure baseline
• incrementality test holdout comparison proving a channel caused lift
• experiment sample-size calculation users needed to detect a given effect reliably
• peeking problem inflated false positives from checking test results early
• novelty effect temporary metric lift because a change is simply new
• survivorship bias conclusions skewed by only measuring users who stayed
• vanity metric impressive number that does not guide a meaningful decision
• stickiness ratio DAU-to-MAU or DAU-to-WAU ratio indicating habitual use
• power user curve (L30) frequency distribution separating casual from heavy users
• funnel analysis vs. path analysis fixed-step vs. free-path exploration of user routes
• funnel leakage disproportionate drop at one specific step
• funnel window time limit within which steps must occur to count
• consent mode / privacy-preserving measurement modeling metrics when users decline tracking
• data freshness / latency delay between an action and its appearance in reports
• metric definition drift same metric name computed differently in two tools
• anomaly detection / alerting automatic flags when a metric deviates from expected