Conversion lag (time-to-conversion)
Conversion lag is the time between an interaction and the resulting conversion. Some conversions happen minutes after a click; others take days or weeks. Because of lag, recent activity always looks under-performing at first — conversions for recent touches have not happened yet — and the lookback window must be long enough to capture them. It is a core reason attribution reports change as data matures.
What this means
Conversion lag is the elapsed time from a touchpoint to its conversion. The distribution varies enormously by business: an impulse purchase may convert in minutes, while a considered B2B or high-value purchase may take weeks of research before the conversion fires.
The practical consequence is that today's numbers are incomplete. Conversions for clicks that happened recently have, in many cases, simply not occurred yet, so any report of the most recent window understates performance until those conversions arrive.
Why it matters for attribution
Conversion lag interacts directly with lookback and reporting windows. If your lookback window is shorter than your typical lag, you will systematically miss conversions whose initiating touch falls outside the window, mis-crediting them. If you read reports too soon, recent campaigns look worse than they are.
Understanding your lag distribution tells you how long to wait before trusting a period's numbers and how wide a lookback window needs to be. It is also why the same period's attribution can shift over the following days — that is maturation, not error.
- Lag varies by purchase consideration length
- Recent windows under-report until conversions mature
- Lookback window should exceed typical lag
How it appears in analytics and logs
Recent periods showing weak conversions that improve over following days are exhibiting conversion lag, not necessarily poor performance — the conversions are still maturing.
Diagnostic use case
Account for conversion lag before judging recent campaigns, and set lookback and reporting windows long enough to let delayed conversions land.
What WebmasterID can help detect
WebmasterID timestamps first-party events, so you can observe the distribution of time-to-conversion in your own data and avoid judging fresh periods prematurely.
Common mistakes
- Judging recent campaigns before conversions have matured.
- Setting a lookback window shorter than the typical lag.
- Mistaking maturation-driven changes for measurement errors.
Privacy and accuracy notes
Conversion lag is a timing distribution over events, not a personal attribute. Measuring it needs event timestamps, not identity.
Related pages
- Lookback and conversion windows explained
A lookback (or conversion) window is the period before a conversion in which earlier touchpoints are eligible for credit. Touches outside the window are ignored entirely. Because every attribution model only sees touches inside this window, its length quietly governs which channels can ever receive credit.
- Decay half-life in time-decay attribution
In time-decay attribution, credit declines exponentially the further a touchpoint is from the conversion, and the half-life is the parameter that sets how fast. A touch one half-life before the conversion gets half the credit of one at conversion time; two half-lives back, a quarter. Choosing the half-life decides how strongly recency is rewarded — a model choice, not a measured fact.
- Attribution window vs reporting window
The attribution (lookback) window decides which past touches can earn credit for a conversion; the reporting window is the date range you are viewing. They answer different questions, and confusing them is a frequent cause of numbers that 'do not add up' between tools or between dates.
- Attribution analytics
Observe time-to-conversion in first-party data.
Sources and verification notes
- Google Ads Help — Conversion windows and conversion delayDocuments conversion windows and the delay between click and conversion.
Last reviewed 2026-06-24. Facts are checked against primary/official sources where available; uncertain specifics are marked “Data not yet verified” rather than guessed.