Mobile Growth Glossary

Attribution based on strong matching signals such as eligible advertising IDs, device identifiers, click IDs, or install-referrer data.

Why it matters: It is generally more reliable than probabilistic matching, but consent and platform availability determine which identifiers can legally and technically be used.

A controlled campaign or ad group using broad match, automated matching, or exploratory targeting to find new search terms, audiences, or creative opportunities.

Why it matters: Winners should be promoted into controlled structures and excluded from discovery to protect reporting clarity and prevent internal competition.

A governed specification defining event names, triggers, properties, identity rules, owners, and business meaning across product, MMP, ad platform, and backend systems.

Why it matters: Without a shared taxonomy, teams optimize different definitions of the same outcome and cannot reconcile campaign performance reliably.

A keyword matching mode intended to serve ads for the selected search term and accepted close variants rather than broad semantic expansion.

Why it matters: It provides stronger bid and reporting control, but close variants and overlap still require search-term inspection and cross-negation.

The moment a user actually receives or experiences an experimental treatment, which can occur after variant assignment.

Why it matters: Tracking exposure rather than assignment alone prevents users who never saw the changed experience from diluting experiment results.

A granular SKAdNetwork conversion value, traditionally represented on a 0-63 scale, used when privacy conditions allow detailed post-install signaling.

Why it matters: The mapping should balance detail with crowd anonymity and prioritize signals that predict revenue or retained value within the available window.

The first recorded app session after download when the measurement SDK initializes and the app becomes observable to attribution tooling.

Why it matters: MMP install counts are commonly based on first open, while app stores may count downloads, creating legitimate reporting differences.

The average number of times each reached user was exposed to an ad during the selected period.

Why it matters: Rising frequency with falling response can indicate fatigue, but the acceptable level depends on objective, audience size, buying model, and creative variety.

Analysis of the proportion and time with which eligible users move through an ordered sequence of events, such as install, registration, activation, trial, and purchase.

Why it matters: It identifies where value is leaking, but entry criteria, event order, conversion window, and identity logic must remain consistent.

Android's resettable advertising identifier for eligible advertising and measurement uses under Google Play policy and user privacy choices.

Why it matters: It is not universally available and should not be treated as a permanent customer identifier or the only attribution mechanism.

An incrementality experiment that compares outcomes across matched geographic markets where advertising exposure is intentionally varied.

Why it matters: It is useful when user-level randomization is unavailable, but market similarity, spillover, seasonality, and statistical power require careful design.

A safety metric monitored alongside an experiment's primary outcome to detect unacceptable harm to retention, revenue, stability, trust, or another critical dimension.

Why it matters: A statistically positive primary metric should not ship when a material guardrail deteriorates beyond the pre-agreed tolerance.

A deliberately unexposed comparison group used to estimate the outcomes that would have occurred without a campaign, feature, or intervention.

Why it matters: A valid holdout turns attributed performance into a stronger causal estimate of incremental impact.

The opening visual, statement, demonstration, or tension that earns attention and frames why the audience should continue engaging with an ad or page.

Why it matters: A strong hook connects a specific audience insight to one clear promise; it is more consequential than minor styling changes.

Apple's resettable advertising identifier, available to an app only when the relevant user authorization and platform requirements permit access.

Why it matters: An all-zero or unavailable IDFA changes deterministic attribution and audience capabilities, but does not mean all measurement becomes impossible.

An iOS identifier shared by apps from the same vendor on a device, subject to reset behavior when the vendor's apps are removed.

Why it matters: It can support vendor-scoped analytics and SDK identity but should not be confused with a cross-company advertising identifier.

The percentage of eligible impressions an advertiser received out of the estimated opportunities available under the platform's auction and targeting conditions.

Why it matters: It helps distinguish limited coverage from conversion problems and prevents unnecessary bid increases when visibility is already saturated.

A time-bound activity promoted on an App Store product page and eligible discovery surfaces, such as a competition, premiere, challenge, or live event.

Why it matters: It creates an additional merchandising and re-engagement surface when the event is timely, accurately described, and tied to real in-app value.

The portion of observed installs, conversions, or revenue caused by an intervention beyond what would have happened without it.

Why it matters: It is the correct lens for cannibalization, brand campaigns, retargeting, organic uplift, and channels where deterministic attribution is incomplete.

A measurement event commonly recorded by an MMP when the downloaded app is opened and the SDK starts for the first time on that install state.

Why it matters: Store downloads, first opens, redownloads, reinstalls, and reattributions use different rules, so the source definition must accompany every install metric.

The percentage of store visits, taps, or another specified upstream action that result in an attributed install.

Why it matters: It diagnoses message and listing effectiveness only when the denominator and attribution source are clearly stated.

Store-provided metadata that passes eligible campaign and click information from an acquisition interaction into the installed Android app.

Why it matters: Referrer matching can provide strong deterministic attribution without relying solely on an advertising identifier.

The hidden iOS App Store metadata field used to supply relevant keyword terms within Apple's character and formatting rules.

Why it matters: It should complement title and subtitle semantics without duplication, while keyword selection remains grounded in relevance, demand, competition, and conversion potential.

A relative estimate of search demand for an App Store keyword, often derived from platform or third-party signals rather than exact query volume.

Why it matters: Popularity helps prioritize opportunities but cannot replace relevance, ranking feasibility, conversion quality, and downstream value.