Liam Walsh
Liam is a Co-Director at Intelligency and heads up the agency's Digital Intelligence & Paid Social activity. Over the last decade, he has worked with brands from the world of sports such as Premier League clubs to entertainment such as Channel 4 and Disney.
If you run Google Ads campaigns using automated bidding, you will be familiar with the learning period. It is the window after a significant campaign change during which Google’s system recalibrates before performance settles. Google has just updated its official guidance on how long that period lasts, shifting the focus from time elapsed to something more meaningful: how many conversions the system has seen. It is a subtle but important distinction with real implications for how you judge campaign performance after making changes.
What the New Guidance Says
Previously, Google described the learning period primarily in terms of time and conversion cycles. Its updated documentation now states that Smart Bidding calibration can take up to around 50 conversions or three conversion cycles, whichever comes first. Campaigns with strong historical data may learn considerably faster. A campaign enters a learning state when an advertiser creates or reactivates a bidding strategy, changes its settings, or makes significant alterations to campaign structure. The update explicitly extends this guidance to Search, Shopping and Performance Max campaigns, noting that Performance Max can take longer when most traffic comes from channels outside Search or Shopping.
Why This Matters More Than You Might Think
The shift from a time-based to a volume-based benchmark changes how you should interpret early campaign data. Previously, many advertisers used calendar time as their reference point. The new guidance reframes the question entirely: it is not how long the campaign has been running, but how much conversion data the algorithm has accumulated. For businesses generating conversions more slowly, perhaps a considered purchase category or a B2B lead generation campaign, this explains why the learning period can stretch across several weeks and why intervening too early can restart the clock.
What You Should Do Differently
After any major bidding change, resist judging performance before the system has gathered sufficient data. Avoid further adjustments to budgets, targets or campaign settings during the learning window, as each change can reset the process. For Performance Max campaigns in particular, Google cautions against frequent changes during the initial period. If your campaigns convert slowly, build the learning period into your planning and stakeholder expectations from the outset. A campaign that looks like it is underperforming in week two may simply be learning.
