Google Ads value rules: only one applies to a conversion
A lead generation client sets three value rules: mobile by 1.2, London by 1.5, Customer Match by 2. A London mobile lead from that list should be worth 3.6 times the base value. The account applies one rule and discards the other two, and the reports look normal while it happens.
That is the single most useful fact about conversion value rules. They do not stack. Google applies at most one rule to any given conversion, and which one wins is decided by a resolution order that nobody reads before creating the second rule.
What a value rule actually changes
A value rule adjusts the number Google Ads attaches to a conversion. It does not change what your tag sends, create extra conversions, or touch the count in the Conversions column. It changes the value beside each conversion, and that value does two jobs.
The first is reporting. The adjustment appears in the conversion value column, and the reported total is the adjusted figure, not the number your site passed. The second is bidding. At auction time, value based strategies read the adjusted value, so a conversion worth £10 that a rule has doubled bids like £20.
That second job is the reason to use them. If you sell through a shop, a call centre or a sales team, the value your website records and the value the business earns are different numbers. A form fill from a city where you have a service team closes at a higher rate than one from a town you cover badly. The tag cannot know that. A value rule can.
They apply to Search, Shopping, Display, Hotel and Performance Max campaigns. Travel campaigns are excluded, though itinerary rules exist for Hotel and Performance Max campaigns on travel goals.
Three operations, and only two of them are generally available
Every rule carries one action.
- Add puts a fixed amount on top of whatever value the conversion passed. The amount must be greater than zero. Add a value of 20 to lift a £5 lead to £25.
- Multiply scales the value by a factor between 0.5 and 10. A factor of 2 doubles it, 0.5 halves it.
- Set replaces the value with a fixed number. Google still labels this as beta, and the API restricts it to accounts on an allowlist or to shop visit and shop sale categories with a single no-condition rule.
For most accounts the choice is between Add and Multiply. Multiply keeps the rule proportional, so a £5 lead and a £50 lead from the same segment both scale by the same factor. Add flattens the relative difference, which is wrong if your values already vary per order. Add fits where the underlying value is constant and you want to correct a known bias, such as a fixed handling cost on phone enquiries.
Conditions: audience, location, device
A rule can carry a primary condition and an optional secondary condition, and at most two conditions in total. The three types are audience, location and device. A fourth option, no condition, appears only when you are setting rules for shop visits or shop sales, and it means the rule applies to that conversion category outright rather than to a slice of it.
One condition type cannot be used twice in the same rule. A location primary condition cannot have a location secondary condition. Multiple selections inside a single condition behave as OR, so picking Birmingham and Southampton gives you a rule that fires for either city.
Device rules are the least interesting of the three because device categories do not overlap. You can lift desktop and lower tablet, and each conversion hits one of them. Location carries most of the value for service businesses, since margins and closing rates vary by postcode and by country. Audience rules let you say that people on your Customer Match list behave differently from everyone else.
Some conditions may be unavailable to advertisers in housing, employment and credit, so a rule that refuses to save deserves a policy check before a bug report.
The constraint that forces a rebuild
Here is the rule that catches people on their fourth rule rather than their first. All value rules in an account must use the same type of primary condition, and the same type of secondary condition if they have one.
If rule one uses location as its primary condition with device as the secondary, then every other rule in the account must also use location as primary and device as secondary. You cannot add an audience primary rule alongside it. If you want to change the shape of your rules, you delete all of them and start again. There is no conversion path and no merge.
The API shows the same constraint differently. Rules live in rule sets, a set declares which dimensions its rules may use (at most two), and a rule can belong to only one set. Reuse a rule and the call fails with a data constraint error.
So the rule design is one account level decision. Choose the pair of dimensions you want to reason about, and write it down, because the next person to open the account will try to add a rule that does not fit.
Only one rule applies, and here is the order
Google's documentation states it plainly. No matter how many rules exist, only one is applied to a given conversion. When a conversion meets the conditions of several rules, the following order decides the winner.
- Campaign level rule sets are considered first. If any rule set is attached to the campaign that produced the conversion, account level sets are ignored entirely for that conversion.
- Within the sets under consideration, a set that names a conversion action category matching the conversion is preferred over a set that applies to all categories.
- Rules inside the chosen set are evaluated in order, and the first rule whose conditions match is the one applied.
- For location, the most precise match wins. A rule for a city beats a rule for the region that contains it, which beats a rule for the country.
- For audience, Google's attribution hierarchy decides. Customer Match sits above your data segments and similar segments, which sit above affinity and in-market, which sit above detailed demographics.
- If there is still a tie, Multiply wins over Add. If two Multiply rules tie, the higher multiplier wins.
- Device rules cannot collide, because the categories are exclusive.
- A rule with no condition catches impressions that match nothing more specific in the same set.
Two consequences follow. First, a narrow rule created last does not automatically beat a broad rule created first. Position in the set decides, not specificity, except where location precision or the audience hierarchy overrides it. Second, two rules of the same condition type that both match are not additive. One of them simply loses.
A worked example
A UK installer records every enquiry at £40. The sales team's own figures say enquiries from Southampton close at roughly double the rate of enquiries elsewhere, so the business treats those as £80. One rule, location primary, Southampton, Multiply by 2, and Google starts reporting £80 and bidding towards £80.
Now the installer adds a second rule. Mobile conversions get Multiply by 1.2 because mobile traffic converts slightly better. A Southampton enquiry on a mobile now matches both rules. Location precision resolves it first, so the Southampton rule applies and the mobile rule is ignored. Bidding targets £80, not £96, and the person who added the mobile rule believes otherwise.
The way to express both signals is a single rule with a primary and a secondary condition, location Southampton and device mobile, Multiply by 2.4. One rule is the only way to combine two dimensions, because two separate rules never combine.
Reading the adjustment in reports
The adjusted value flows into the conversion value metrics, including total conversion value and all conversions value, so any report built on those figures already carries the effect. To see the effect on its own, segment the campaign report by Conversions, then Value rule adjustment.
The segment splits the total into original value where a rule applied, original value where none applied, and the adjustment itself, broken out by audience, location, device or no condition depending on the set's dimension. Total conversion value is the sum of those, so the headline alone cannot tell you how much came from the rule.
Treat the segment as a reconciliation tool rather than a performance report. If the adjustment column is large, the rule is doing a lot of work and deserves scrutiny. If it sits near zero, the rule is either not matching or matching segments that behave like everyone else.
Which bid strategies actually use it
Value based strategies optimise to the adjusted value. Target ROAS, Maximise conversion value and Performance Max all read it at auction time, which is the whole point of the feature.
Strategies that do not optimise on value still see the adjustment in reporting, but bidding does not change. Manual CPC, Target CPA and Maximise conversions are in that group. If the account runs on Target CPA, a value rule makes your reports more accurate and leaves bidding alone. Worth doing, but do not expect a performance shift.
Rules created at campaign level affect only that campaign. The exception is a portfolio bid strategy, where a change to one campaign can carry a small effect into the others sharing the strategy, because the portfolio optimises as a group.
The caveat Google writes itself
Google's own guidance includes a warning that most articles about value rules skip. Smart Bidding already reads geography, device and first party audience lists as signals. If a segment converts better and that shows up in your reporting, bidding already accounts for it. A value rule is for information that is not in the account at all: offline close rates, margin differences, contract values, the fact that a lead from one region gets a service engineer within a day and a lead from another waits a week.
That is the test for whether a rule is worth creating. If the signal is already in the conversion data, a rule double counts it. If the signal lives only in the CRM or the finance spreadsheet, a rule is the right place for it.
Removing a rule returns bidding to your definition of value as it stands at that moment. Pausing and deleting both stop the adjustment. Nothing about past reporting changes, so an experiment is reversible in the sense that future bids revert, not in the sense that the history unwinds.
Before you create one
- Confirm the account runs a value based strategy. On Target CPA, expect reporting accuracy and no bidding change.
- Decide the condition pair once, at account level, and write it down. Changing it later means deleting every rule.
- Check whether the signal is already visible in the conversion data. If it is, skip the rule.
- Prefer one rule with two conditions over two rules with one each, because the second will never fire alongside the first.
- Check for cross-account conversion tracking. If it is in use, rules must be created at the manager account.
- Segment by value rule adjustment a month later and check the adjustment is the size you expected.
The feature is small, free and needs no retagging, which is why it accumulates quietly. Three rules that look complementary, one of which silently wins, is a common state in accounts handled by more than one person. Reading the resolution order takes ten minutes and explains numbers that otherwise look arbitrary.
Not sure which value rule is steering your bids? Book a free tracking audit and we will map every rule, check which one wins for each segment, and confirm your bid strategy actually reads the values you think it does.