More clean, recent conversion data usually gives Smart Bidding a better chance, but 30 bad conversions are not better than 10 useful ones. Google recommends looking at at least 30 conversions over 30 days when evaluating Target CPA, while some strategies have specific eligibility rules, such as 15 conversions in 30 days for Target ROAS on Search and Shopping. Those numbers are guides or requirements, not a guarantee of good performance.
People often want one safe threshold: wait for 15 conversions, switch at 30, or never automate below 50. I do not think the account suddenly becomes intelligent when it crosses one of those lines.
The practical question is whether the campaign has enough recent, relevant feedback for the strategy and target you are asking it to use.
Why several different numbers get repeated
Different numbers refer to different things. A platform eligibility requirement is not the same as a performance recommendation, and neither proves that a campaign will hit your target.
For example, Google says Search and Shopping generally need at least 15 conversions in the previous 30 days to use Target ROAS. Its Target CPA guidance recommends evaluating the last 30 days with at least 30 conversions. Maximise Conversions can run with less history, but that does not mean every low-volume campaign will be stable.
Conversion quality matters before conversion quantity
I would rather give Google 20 genuine enquiries than 80 events made up of page views, duplicate forms and accidental calls.
Smart Bidding is trying to predict the outcome you feed it. If the goal is easy to trigger or commercially meaningless, more volume can simply help the system become better at finding the wrong people.
Conversion delay changes what recent data means
A purchase that normally happens on the same day gives faster feedback than a B2B opportunity that closes three months after the click.
I look at the conversion cycle before judging a recent change. Google recommends allowing one or two conversion cycles after target adjustments because the conversions still arriving today may belong to clicks from before the change.
Do not split limited data into too many campaigns
A common problem is not that the account has too few conversions overall. It is that the useful data is divided between too many campaigns, locations, match-type experiments or nearly identical ad groups.
I would consolidate where the campaigns share the same goal, economics and strategic purpose. I would keep them separate when budgets, targets, locations or product economics genuinely need different control.
An unrealistic target can make good data look useless
Even a mature campaign can struggle if the Target CPA is set far below what the market and website can support. The system responds by entering fewer auctions and may produce less volume, not a miracle efficiency gain.
The starting target should make sense against recent performance and the business numbers. I would tighten it gradually when the evidence supports that.
What I would do with a low-volume account
- Make sure the primary conversion is real and consistently recorded.
- Combine campaigns that do not need separate control.
- Avoid changing budgets, targets and conversion goals at the same time.
- Use a higher-volume meaningful stage when the final sale is too rare.
- Judge over a period that includes the normal conversion delay.
Low volume does not automatically mean manual bidding forever. It means I would be more cautious about tight targets and confident conclusions.
Common questions
Does Maximise Conversions need 30 conversions first?
Google does not present 30 conversions as a universal requirement for Maximise Conversions. Low-volume campaigns can use it, but performance may be less stable and the conversion goal still needs to be trustworthy.
How many conversions does Target ROAS need?
For Search and Shopping, Google currently lists at least 15 conversions in the previous 30 days at the conversion-tracking level. Other campaign types have different requirements.
Should I wait until the learning period ends before making changes?
Give the strategy enough time to receive conversions through the normal conversion cycle. Fix genuine errors immediately, but avoid reacting to a few weak days with several simultaneous changes.
Official platform sources: Google Ads overview of Smart Bidding, Google Ads Target CPA guidance, Google Ads Target ROAS requirements, Google Ads guidance on target changes and conversion cycles. The practical recommendations and interpretation above are my own.
Not sure whether your bidding setup has enough data?
I can review the conversion quality, campaign structure, targets and recent volume together instead of applying one generic threshold to the account.
Start a Conversation