The initial phase of Demand Gen campaigns are critical for overall campaign success. During this period, the campaign or portfolio undergoes a calibration process to align with the newly established goal. This article provides insights into the duration of this learning period for campaigns, the various factors that can influence this initial phase and where to get more details about the status of your Demand Gen campaign's learning phase.
Before you begin
This is a Beta version. Help center articles may not be available in all languages.
To ensure optimal performance, it's crucial to avoid making significant changes to your campaign settings during the learning phase. This can disrupt the system's learning process and delay its ability to reach optimal performance. It's best to follow the recommendations provided by the Google Ads UI.
As the system learns, you might notice fluctuations in your campaign's performance metrics, such as click-through rate (CTR) and conversion rate. These fluctuations are a natural part of the learning process as the algorithm gathers more data. Once your campaign reaches a stabilized performance, these variations should even out.
Why campaigns fluctuate during the learning phase
After you create a new campaign or make a change to your bids, budgets or creatives, the campaign or portfolio needs some time to calibrate towards the campaign goal. When this occurs, a "Learning" status may be shown in your bid strategy status. Learn more About bid strategy statuses.
Google Ads operates on a sophisticated machine learning algorithm that requires a certain amount of data to understand and optimize your campaigns effectively. This initial phase, often referred to as the “learning phase” (or "learning period”), spans up to 2 weeks when your campaign is set up correctly.
Where to learn more about my campaign’s learning phase
During campaign construction
When building a Demand Gen campaign, after adding a campaign budget advertisers can see a campaign forecast card titled “Learning phase estimates.” These estimates explain what kind of performance an advertiser can expect during the learning phase, including:
- Days to first conversion: Expected amount of days it will take until a campaign achieves its first conversion.
- First week conversions: Expected amount of conversions a campaign will achieve within the first week
- Days in learning phase: Expected amount of days a campaign will take to receive enough conversions to exit the learning phase
Advertisers will be notified if a campaign’s learning phase estimates meet a typical successful campaign’s performance expectations:
- Campaigns that meet or exceed a typical successful campaign’s performance expectations will be indicated by a checkmark icon next to the estimate.
- Campaigns that do not meet a typical successful campaign’s performance expectations will be indicated by an information icon next to the estimate.
- Advertisers can hover over the icon to learn more how to improve estimates, including making changes to bid, budget, conversion goal, audience or creative.
These performance estimates are based on a variety of factors, including campaign settings (bid and budget, conversion goal, audience selections) and account performance history. Factors not taken into account that may impact campaign performance include correct conversion pixel setup and eligibility of creative assets. These estimates also don’t account for potential delays in conversion reporting, but rather reflect eventual reporting.
When managing a live campaign
You can confirm whether your campaign is still learning by consulting its bid status. If it indicates “Learning,” it is recommended that you don’t make additional changes while a campaign is in learning in order to reach full campaign performance more quickly.
Tip:
You can hover over the ‘Learning’ status to learn why your bid strategy is in the "Learning" phase. Learn more About bid strategy status.
