metricData, experimentation & growth

Conversion Rate

Conversion rate is the share of an eligible population that completes a defined action within a specified journey and time window.

Download Markdown (.md)

01What it means

Conversion rate divides the number of converting units by the number of eligible units and expresses the result as a proportion or percentage. The unit may be users, sessions, accounts, leads, or transactions, so the denominator and conversion event must always be stated.[1][2][3]

A defined action divided by its eligible opportunity Conversion rate is only comparable when event, unit, window, and denominator stay consistent
Conversion rate = converting units ÷ eligible units × 100
Conversion event
One precisely defined qualifying action
Eligible denominator
Users, sessions, accounts, leads, or transactions
Identity + deduplication
Rules preventing unintended multiple counts
Window + attribution
When and how a conversion is assigned

Pair the percentage with volume, downstream quality, guardrails, and causal evidence before attributing movement to a product change.

02Why it matters

Conversion rate connects product or marketing changes to observable actions such as purchase, activation, application completion, or renewal. A precise definition makes funnel performance comparable over time and helps teams identify where customers fail to progress.

03When to use it

Use conversion rate when there is a clearly defined desired action, eligible population, attribution rule, and observation window. It is useful for funnel analysis and experiments, but the chosen action should represent genuine value rather than a convenient click.

04How it works

  1. Define the conversion event, eligible denominator, identity unit, attribution logic, and observation window.
  2. Validate that events are deduplicated and that one user or session cannot create unintended multiple conversions.
  3. Divide converting eligible units by all eligible units and multiply by 100 when reporting a percentage.
  4. Segment by meaningful cohort, source, device, or journey stage while guarding against tiny samples and multiple-comparison stories.
  5. Pair the rate with volume, downstream quality, guardrail metrics, and experimental evidence before attributing a change to a design decision.

05A practical example

Example

A trial product defines activation conversion as the percentage of new eligible workspaces that import data and successfully publish a first report within 14 days. A simplified import flow raises the rate, but the team also checks report accuracy, support contacts, and later retention before scaling it.

06Where it breaks

Conversion rate is highly sensitive to denominator, time window, tracking quality, and traffic mix. A higher rate can coincide with lower volume or poorer downstream outcomes. Observational changes do not establish causality, and aggressive optimization can create coercive or misleading experiences.

Common mistakes

  • Reporting conversion without naming the denominator, event, and time window.
  • Counting events rather than unique eligible units when the question is user conversion.
  • Celebrating an upstream increase without checking downstream quality or guardrails.

07Sources

  1. Conversion rate: DefinitionGoogle Ads Help · Authoritative analytics documentationConversion rate: Definition
  2. ConversionGoogle Analytics Help · Authoritative analytics documentationConversion
  3. Default calculated metricsAdobe Experience League · Authoritative analytics documentationDefault calculated metrics
View saved topics

Start typing to search titles, definitions, categories, and frameworks.