Maximise Conversion Value is a Google Ads Smart Bidding strategy that sets bids to bring in the highest total value of conversions your budget allows, rather than the highest number. It needs a value attached to each conversion, such as order revenue for a shop or an estimated worth for each type of lead.
How Maximise Conversion Value works
For every auction, Google predicts two things: how likely this search is to lead to a conversion, and how much that conversion is likely to be worth. It then bids more where the expected value is high and less where it is low and, when no target is set, aims to spend your full daily budget. A search for a premium product range might attract a higher bid than a search for a bargain accessory, even if both convert equally often.
You can add an optional target ROAS. Without one, the strategy simply chases the most value it can find within the budget, which can mean a low return on the last pounds spent. With one, it tries to hit that return on average and may spend less than the budget if it cannot find enough auctions that meet it. Google has folded the old standalone Target ROAS strategy into this one as an option.
The values come from your conversion values: transaction revenue passed by your ecommerce tracking, fixed values you assign to lead types, or values adjusted by conversion value rules for particular locations, devices or audiences.
Why it matters
Not all conversions are equal. A Surrey kitchen fitter who counts a brochure download and a design consultation booking as one conversion each will see Maximise Conversions chase the cheap downloads. Give those actions honest values and Maximise Conversion Value will bid towards the consultations, because that is where the value is.
For online retailers it is usually the natural strategy for Shopping and Performance Max campaigns, because revenue varies so much from order to order. The catch is that it is only as good as the values it receives. If those values are wrong, the algorithm will be efficiently wrong.
Common mistakes
- Passing revenue including VAT and delivery while comparing ROAS against margins calculated without them.
- Using revenue when product margins differ sharply, so the strategy pushes low-margin best-sellers. Value based on profit, or profit on ad spend, fixes this.
- Switching on a campaign with only a handful of conversions a month, where predictions have too little to learn from.
- Setting a target ROAS far above recent performance on day one, which starves the campaign of auctions.
- Including a secondary action, such as add to basket, in the conversions the strategy optimises towards, so value is counted twice.
- Changing the target every few days, which keeps resetting the strategy’s learning.
How to act on it
Check the values first: make sure purchase revenue is accurate, decide whether you report net or gross of VAT and use that consistently, and give lead actions values that reflect how often each type becomes paying work. Then run Maximise Conversion Value without a target for a couple of weeks to see what return it achieves naturally, and introduce a target close to that level.
Move targets in modest steps and judge results over at least one full conversion cycle. This sits within the wider approach of value-based bidding, which I set up and monitor as part of monthly PPC management.