Conversion Value Rules: Telling Google Which Leads Matter

Conversion value rules and ROAS need commercial evidence

A campaign producing 200 leads can be commercially worse than one producing 100. Conversion value rules can teach Google that difference, but only after CRM closed-won data proves it over at least one complete sales cycle. They are a lead-quality instrument, not a bidding cheat code; writing rules from intuition trains the bidder on your guesses.

Google Ads can add to or multiply a conversion value when a conversion matches one of three practical dimensions: location, audience or device. The adjusted value can then inform Maximise conversion value or target ROAS bidding.

That mechanism is useful when the standard conversion value hides a genuine commercial difference. A London enterprise enquiry might consistently produce more closed-won value than an overseas small-business enquiry. Treating both as identical gives Google the wrong objective.

The common abuse is different. Someone increases the value assigned to a preferred territory, reported ROAS rises, and the account appears healthier. Cash has not changed. The advertiser has edited the numerator.

That is why value rules belong downstream of measurement. First establish the distinction between reported Google Ads ROAS and commercial return. Only after the value basis is stable should industry ROAS benchmarks with proper caveats enter a budget conversation.

Our position is blunt: if the CRM cannot identify which segments close, no value rule should exist.

Commercial evidence should lead; Google Ads should follow.

Measure before the bidder sees a multiplier

The version we see in audits is usually a spreadsheet of hunches. Sales prefers one region, the founder dislikes mobile leads, and an agency assumes larger companies must be worth more. None of those opinions establishes value.

The order of operations is non-negotiable: measure close rate by segment for at least one full sales cycle before writing a single value rule. Writing rules from intuition trains the bidder on your guesses.

Use this five-step order:

1. Lock one dependable lead conversion and one commercial outcome, such as closed-won.
2. Define segments that both Google Ads and the CRM can identify consistently.
3. Let every lead mature through at least one complete sales cycle.
4. Calculate close rate and closed-won value per raw lead for each segment.
5. Apply the sample, materiality and stability thresholds before creating any rule.

A short reporting window creates a predictable distortion. On an illustrative 90-day sales cycle, comparing a 30-day cohort with a fully matured 90-day cohort means unresolved opportunities are being treated as failures. Waiting is part of the measurement, not administrative delay.

Segment labels must also reflect actual buying groups. The same discipline appears in our case-study library: Lanteria’s broad HR platform required routes for multiple stakeholder audiences, while Lake Erie Shores separated stay and ownership audiences under one site. The lesson is architectural rather than numerical: meaningful groups must exist before they can be valued.

Offline conversion import must remain the route by which downstream outcomes reach Google Ads. A value rule is not a substitute for that evidence. If the underlying action or revenue record is unreliable, repair the inaccurate ROAS data at source before applying weights.

Evidence must precede every conversion value rule.

A three-segment UK B2B worked example

Take an illustrative UK consultancy spending £8,000 a month. These figures demonstrate the arithmetic and are not measured client results.

The six shared inputs are:

• Monthly Google Ads spend: £8,000.
• Observation window: six acquisition months.
• Sales cycle: 90 days.
• Lead-count basis: all non-spam enquiries.
• Value basis: signed first-year fees from closed-won deals.
• Stability condition: the segment ranking repeated across two matured three-month cohorts.

The three observed segments produce the following comparison:

The ad-spend input is straightforward:

£8,000 × 6 = £48,000 total ad spend

Closed-won value for each segment is calculated independently:

15 × £8,000 = £120,000 for London enterprise

12 × £5,000 = £60,000 for rest-of-UK mid-market

12 × £5,000 = £60,000 for Ireland SME

The blended value per raw lead becomes:

(£120,000 + £60,000 + £60,000) ÷ (100 + 150 + 200) = £533.33

The illustrative closed-won ROAS is:

£240,000 ÷ £48,000 = 5.00× closed-won ROAS

Each multiplier compares segment value per raw lead with the £533.33 blended baseline:

(£120,000 ÷ 100) ÷ £533.33 = 2.25× for London enterprise

(£60,000 ÷ 150) ÷ £533.33 = 0.75× for rest-of-UK mid-market

(£60,000 ÷ 200) ÷ £533.33 = 0.56× for Ireland SME

Raw volume points in the wrong direction. Ireland generated 200 leads versus London’s 100, but produced £60,000 closed-won versus £120,000. Its close rate was 6% versus 15%, while value per raw lead was £300 versus £1,200.

The honest limit here is that geography can be confounded by sales-representative allocation, discounting and product mix. These figures cannot prove that location caused the value difference. If the best closer receives London enquiries, changing sales routing may be more defensible than changing bids.

Closed-won value, not lead volume, earns the multiplier.

If the path from ad click to signed revenue still has gaps, Actualyse will trace each hand-off with you — book a call

Four decision rules for conversion value rules and ROAS

A segment deserves bidding weight only after passing all four gates.

1. Maturity gate: wait at least one full sales cycle before assessing the segment. An unresolved cohort receives no rule.
2. Volume gate: require a minimum of 10 closed-won deals in the exact segment. At nine or fewer, merge it into a parent segment or leave it at 1.0×.
3. Materiality gate: require closed-won value per lead to differ from the baseline by at least 20% across two matured cohorts. A calculated 1.19× remains 1.0×; a stable 1.20× can justify a rule.
4. Drift gate: suspend the rule when close rate or value per lead moves by more than 25% between consecutive full-cycle cohorts. Recalculate before restoring it.

The four gates apply differently across three rule dimensions.

Geography

Use a geographic rule when a real commercial boundary affects closing or contract value. Sales coverage, serviceability and market-specific pricing are defensible boundaries. A region selected merely because management wants growth there is not.

Audience

Use an audience rule when the cohort maps consistently between CRM records and a supported Google Ads audience. Enterprise prospects, existing customers and partner-referred companies may warrant separate analysis. Broad labels such as “high intent” deserve no multiplier unless closed-won results validate them.

The audience logic should also survive a proper B2B Google Ads lead-quality diagnosis. A segment producing more qualified leads but identical closed-won value per raw lead has not earned a different rule.

Device

Device is usually the weakest B2B business segment because poor mobile performance often reflects form friction, page speed or call handling. Fix those faults before concluding that mobile users are commercially inferior. A device rule still requires 10 closed deals, a 20% value difference and stable results across two cohorts.

In our Google Ads management for B2B companies, each rule needs an owner, evidence period, calculation and withdrawal trigger. Undocumented multipliers become permanent folklore.

Our claim is falsifiable: a useful segment will retain the same above-or-below-baseline rank across two completed sales cycles; one rank reversal proves it too unstable for a rule.

Stable segment evidence deserves bidding weight; unstable segmentation deserves none.

What conversion value rules do not fix

Google Ads can weight an input; it cannot manufacture a commercial fact.

Three boundaries are absolute: value rules do not change reporting truth, do not fix a broken conversion action, and do not survive a segment whose close rate moves.

Four specific failures matter:

1. They do not change reporting truth. Adjusted conversion values may change Google Ads value and ROAS columns, but they do not change customer payments. An adjusted value remains a bidding instruction, not audited revenue.
2. They do not fix a broken conversion action. Three false signals commonly cause trouble: thank-you-page reloads, unqualified phone clicks and spam submissions. Multiplying any of them merely gives the error greater bidding influence.
3. They do not replace offline conversion import. Without downstream outcomes, Google still knows that a lead occurred but not whether the opportunity became revenue. A multiplier cannot supply the missing event history.
4. They do not survive segment drift. On an illustrative comparison, a segment falling from a 14% close rate to 7% no longer deserves a multiplier based on the earlier cohort. Sales staffing, proposition changes or seasonality can invalidate the original relationship.

Value rules also cannot make a weak offer persuasive, repair a slow landing page or create demand in an unsuitable market. Those interventions change the underlying outcome; weighting only changes how Google values the observed signal.

No multiplier can repair false, broken or unstable conversion evidence.

FAQ

Four implementation decisions settle the common edge cases.

Should I use an additive or multiplicative adjustment?

Choose multiplication when the segment difference scales with the base lead value. Choose addition only when the segment contributes a genuinely fixed £ amount. On an illustrative £500 base value:

£500 × 1.5 = £750

Adding £250 reaches the same result only at a £500 base, while multiplication preserves the relative 50% difference when base values vary.

Do conversion value rules require target ROAS bidding?

No platform requirement forces target ROAS. The named decision is to use rules for bidding only when the strategy consumes conversion value: Maximise conversion value or target ROAS. Maximise conversions pursues conversion count, so changing values does not give that strategy a value objective.

When should target ROAS change after publishing a rule?

Hold the target steady for one complete conversion-lag window. If the account’s own lag report shows 80% of outcomes settling by day 21, wait at least 21 days and then change only one bidding variable. Simultaneously changing the rule and target removes any useful comparison.

Can Search and Performance Max share the same rules?

Use one rule set only when three conditions match: the primary conversion action, the value basis and the eligible geography. If one condition differs, separate the rule logic rather than forcing unlike campaigns into one valuation model.

Bidding settings should follow the value model, never obscure it.

Summary

The five operating rules are:

• Wait one complete sales cycle before assessing any segment for a value rule.
• Require at least 10 closed-won deals in the exact segment.
• Apply a rule only when value per lead differs by at least 20% across two matured cohorts.
• Calculate each multiplier as segment closed-won value per lead divided by the blended baseline.
• Suspend any rule when segment value moves by more than 25% between full-cycle cohorts.

Rules earn authority through stable closed-won evidence.

Actualyse builds measurement and attribution setups that tie B2B ad spend to real revenue. Book a call to talk through where yours stands.