Where can I see purchase lag in Google Ads?

Google Ads shows lag on three surfaces. In the Campaigns, Ad groups or Search keywords table, choosing Days to conversion under Conversions from the segment icon splits the conversion columns into lag rows; this is the distribution that can be narrowed to the purchase action. The Path metrics report under Attribution shows average days or hours by default, also gives day-level conversion rows, and its Measure from menu picks first or last ad interaction. In accounts using Smart Bidding, hovering over the Conversions column shows the average lag and the estimated number of conversions not yet reported. In all three, the report date range needs to end at least 30 days before today.

Is there a report in the GA4 interface that shows a lag distribution?

No. The top row of the Key event attribution paths report has the Days to key event and Touchpoints to key event metrics, but these are a single average; a distribution split into buckets such as 0 days, 1 day, 2 to 3 days cannot be produced in the interface.

Is there a way to get the distribution without BigQuery?

Partly, through the Data API. When the firstSessionDate user dimension and the date dimension go into the same query and the event name is filtered to the key event, each row carries a first session date and a key event date, and their difference is the lag in days. The GA4 connector in Data Studio sees the same dimension. It has two limits: rows are aggregated by date pair, so they need to be weighted by event count, and first and repeat orders are not separated; rows with a (not set) first session date drop out of the distribution, and the reason for that value cannot be determined on its own.

How do I translate the lag I measured into a setting?

The measured p90 becomes input to three separate decisions. Google recommends choosing the window based on the time lag report and at least 7 days; taking the smallest option above p90 is my rule, and it accepts leaving roughly a tenth of the tail outside. Since the current window also truncates the lag report, a p90 measured under a 7-day window cannot see beyond it; the measurement should be done under a longer window. The GA4 key event lookback window for purchase stays at the 90-day default, because changes apply only going forward and narrowing it changes the credit of early touchpoints and comparability with the past. For the earliest date a campaign can be evaluated, the rule I use is learning period plus 2 full conversion cycles; Google says roughly 50 conversions or 3 cycles for calibration and at least 2 full cycles for evaluation, but gives no formula that adds the two.

Why do Ads, GA4 and BigQuery show different lag for the same account?

There are three structural reasons. Ads writes the conversion to the click day, GA4 and BigQuery to the conversion day. Ads sees only Google touchpoints in the conversion path, GA4 sees all channels including organic and email, and BigQuery sees whatever the query defines. Ads includes cross-device modelled conversions, the GA4 interface includes consent modelling, BigQuery only observed events. The three are distributions of different customer sets; since the behaviour of the excluded users is unknown, there is no guaranteed ordering between them.

Measuring Purchase Lag: Google Ads Reports, the GA4 Gap and the Window Decision

Every window setting, conversion window and lookback window alike, depends on a store's purchase lag, yet that lag rarely makes it into conversion analysis. This post works through the three reports Google Ads provides, the distribution the GA4 interface does not provide, the Data API workaround and how a measured p90 translates into each setting.

2026-09-10 Updated: 2026-10-14

TL;DR

Purchase lag is a distribution, not a single number, and Google Ads and GA4 show different parts of it. Ads measures only from Google touchpoints; the GA4 interface gives one channel-agnostic average. Day-level percentiles come from Ads rows and the Data API; a distribution with a self-defined start event, identity and first order comes only from the BigQuery export. The measured p90 feeds three settings: Ads conversion window, GA4 lookback and the earliest campaign evaluation date.

Get Access Sign In

Subscribers Only

The full article plus 25 more subscriber-only posts come with a subscription.

analytics

Subscribers Only

RELATED

Linking a Banner Click to a Purchase: Promotion Events, the Carrier Layer and the Consent Boundary

Session Integrity from Landing to Purchase: Counting the Break in BigQuery

Why Google Ads and GA4 Conversions Never Match: Counting Unit and Time Axis

Joining Google Ads and GA4 at Row Level: A Shared Key and a Two-Layer Reading

Connecting a Shopify Order to an Ad Click: Choosing a Join Key for Offline Conversions

GA4 Attribution Reports: Direct Inflation, Model Selection, and Finding the Real Source

Tracing the Source of GA4 and Ads Signals in Shopify

Deriving ID and Slot for E-Commerce Homepage Banners

Google Ads Conversion Double Counting on Shopify: Diagnosis with the Webpages Report and the Checkout Token

Measuring Consent Rate: Where Accepts, Rejects and Later Changes Get Counted