Data freshness is how up to date the figures in a report are: the delay between something happening on your site or in an ad account and that event appearing, fully processed, in the numbers you look at. Every analytics and advertising platform has some delay, and the most recent days are usually incomplete.
How data freshness works
When a visitor buys something, the event reaches GA4 within seconds and shows up in the Realtime report. Standard reports are a different matter. GA4 processes data in batches, joins events into sessions, applies attribution and, where consent mode is in use, adds modelled figures. At the time of writing (October 2026), Google says processing for standard properties can take 24 to 48 hours, and figures for recent days can change during that time.
Other platforms have their own delays:
- Google Ads updates most statistics within a few hours, but conversions keep arriving for days or weeks after the click as people come back to buy. This is conversion lag, and it means last week’s cost per conversion usually looks worse than it will once all the sales are counted.
- Meta Ads Manager shows spend almost immediately, while conversions inside the attribution window can keep being credited for several days afterwards.
- Search Console usually runs a couple of days behind for its standard date ranges.
- Dashboards add another layer. Looker Studio caches data and refreshes it on a schedule set for each data source, so a dashboard can be hours behind the platform it reads from.
- Raw exports vary too. GA4’s BigQuery export offers a daily export and an optional streaming export, which arrives faster but adds cost.
Why it matters
Decisions made on unfinished data are often wrong. A Bristol plumber checking Google Ads on a Monday morning might see Sunday with spend and no enquiries, and pause a campaign that, once Sunday’s late conversions arrive, was doing well. An online retailer comparing this week with last week on a Wednesday is comparing two and a half days with seven.
Automated bidding is affected as well. Google’s Smart Bidding already allows for conversion delay, so manual changes made in reaction to the last 48 hours can undo a strategy that was working. For that reason, in ongoing PPC management I judge performance on complete periods and treat the most recent few days as provisional.
Common mistakes
- Judging a campaign, landing page or SEO change on yesterday’s figures.
- Comparing a partial period with a complete one, such as month to date against the whole of last month.
- Setting dashboard date ranges to end today, which always shows an apparent drop at the end of the chart.
- Expecting the Realtime report and standard reports to match. They are processed differently and often will not.
- Pulling figures for a board or client report on the first of the month, before the previous month has finished processing.
How to act on it
Set a simple house rule: no decisions on the last two or three days of data, and monthly reports produced no earlier than the third or fourth of the following month. Change the default date range on dashboards to end a few days ago, or label recent days as incomplete.
In Google Ads, check the conversion lag for your own account by segmenting the conversions columns by “Days to conversion”, and use the result to decide how long to wait before judging a change. Ad accounts selling considered purchases, such as kitchens or private medical treatment, often need longer than those selling everyday items. For important reports, record the date the data was pulled, so anyone reading it knows how complete it was.