Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Glossary term

Conversion Lag

The time between an ad interaction and the conversion it leads to, which leaves the most recent reporting days incomplete.
AnalyticsconceptUpdated

Definition

Every conversion has two dates: the date of the ad interaction that is credited with it and the date the conversion actually happened. Conversion lag is the gap between them. For an impulse purchase it may be minutes; for a considered purchase, a booking made well ahead or a business deal it can be weeks.

The lag matters because of how platforms date their reports. Ad platforms attribute each conversion to the date of the interaction, while analytics tools usually record it on the day it occurred. The comparison of platform attribution and analytics attribution sets out that difference and its consequence: a platform's figures for any past period keep rising for as long as conversions from that period's clicks are still arriving.

Why recent reporting days are incomplete

Take any day's clicks. Some of the people behind them convert the same day, more convert over the following days, and a few convert near the end of the window in which the platform will still credit them. On the morning after, only the first group has been counted. The day's conversions, conversion rate and return all look low, and they rise as the rest come in.

Three things lengthen the lag. Considered purchases take longer to decide. Leads that count only once they qualify depend on how quickly sales teams review them. Conversions uploaded from a CRM through offline imports arrive when the import runs, and are then dated back to the click. Each moves more of the eventual total into the period after the report is read.

In practice: reading tests and bidding around the lag

The practical rule is to judge periods that have finished converting, not periods that have finished spending. Teams measure their own lag from the platform's time-to-conversion reporting or from their CRM, note how long it takes for most conversions from a day's clicks to arrive, and treat anything more recent as provisional.

  • Tests. A test read the day it stops spending favours whichever arm converts faster. The broad match expansion test reads its result only after the conversion delay has passed, because an arm reaching new, less familiar searches may attract people who take longer to buy.
  • Automated bidding. Bid strategies learn from conversions as they arrive, and their targets are averages over time. The comparison of target CPA and target ROAS warns that teams who react to lag-depressed recent figures by changing targets keep the system relearning, and recommends a review rhythm that respects the lag. The Smart Bidding strategies guide explains which conversion data each strategy learns from, and so which delays reach it.
  • Cross-engine bidding. The SA360 bid strategies guide evaluates a strategy against its target only over a window long enough for Floodlight conversions to have arrived.
  • Planning. When a quarter ends far from forecast, the annual and quarterly marketing planning guide points to a diagnosis that tests conversion lag among the first explanations, before the next plan repeats the assumption.

Lag, lookback windows and sales cycles

Three ideas sit close together. Conversion lag is how long customers actually take. A lookback window is how long the platform will wait for them: conversions that arrive after it closes are not credited at all. Sales cycle length is the business's own measure of how long a buyer takes from first contact to purchase, which may include many interactions the ad platform never sees. A lag that is often longer than the window means conversions are being lost from ad reporting rather than merely delayed, and the sales cycle is the reason to check.

Illustrative example — not a real result

A lodge-rental company stops cutting budgets on Mondays

Pinewhistle Lodges spent about £600 a week on search ads. One Monday the report for the previous seven days showed 6 bookings, against a usual 14, and the manager cut the budget by a third. A fortnight later the same week showed 15 bookings: most guests booked three to ten days after their first click, after comparing lodges with their party. The company exported time-to-conversion data and found that about 55% of bookings arrived within three days of the click and about 90% within twelve. Weekly reviews now judge only the week ending fourteen days earlier, label the most recent fortnight as provisional, and leave budgets unchanged unless a completed week falls below its usual range.

A Lookback Window is the limit on how long a platform waits for a conversion; lag is how long customers actually take. Smart Bidding learns from conversions as they arrive, so long lags slow its feedback and make recent results look weaker than they are. Sales Cycle Length measures the whole journey from first contact to purchase, of which the ad-to-conversion lag is one part. Attribution decides which interaction receives the credit, and therefore which date the conversion is reported against.

Common confusions

  • Recent days show real performance. They show performance so far; the rest arrives later.
  • Conversion lag is a tracking fault. It reflects how customers decide; only very long or irregular lags point to a tracking or import problem.
  • Lag and the lookback window are the same thing. One is customer behaviour, the other a platform setting.
  • Analytics and the ad platform will agree on yesterday. They date conversions differently, so recent days differ most.
  • A short test with fast results is the safer read. It favours whatever converts quickly, not whatever converts most.

More terms to look up

Where this term is used

Related