Feed customer lifetime value back into search bidding so the algorithm chases high-value acquisition, not just cheap conversions
Search bidding optimizes toward conversions, and by default it treats them all as equal. A free-trial signup that never converts and a customer who will spend thousands over two years are, to the bid algorithm, the same conversion at the same value, so it bids to acquire whichever is cheaper. In categories with wide customer-value dispersion, this quietly steers the budget toward low-value, easy-to-acquire conversions, because those are cheaper per conversion even though they are worth far less per customer.
The information that would fix this lives in the CRM and the warehouse: which customers turned out valuable, and what the early signal of value looks like. Search platforms can consume that information through offline conversion import and value rules, but most accounts never close the loop, so the algorithm keeps optimizing on a click-to-conversion signal that knows nothing about what the customer was eventually worth.
This recipe feeds customer value back into search bidding, so the algorithm weights high-value acquisition rather than chasing the cheapest conversion.
Search bidding that optimizes for customer value rather than conversion count, shifting acquisition toward higher-worth customers. The metric is the high-value share of search-acquired customers, which should rise as the value signal reaches the bidding, ideally without blended CAC running away (a higher CAC for genuinely higher-value customers can be the correct trade, which is why the metric is value share rather than raw cost). The size of the effect tracks how much value dispersion exists in the customer base and how predictive the early value signal is; a subscription business with wide LTV variance gains more than a flat-priced one where every customer is worth about the same.
The second outcome is a closed loop: the search platform finally learns from what customers were actually worth, not just from whether they converted.
The search platform's offline conversion import and value-based bidding (Google's offline conversion import plus value rules and target ROAS) is the consumption side. The supply side is a warehouse join: GCLID captured at click, bound to the customer record, joined to the LTV score, and exported back to the platform. This is lighter than the full cross-channel value recipe because it is scoped to search and to one import mechanism, which is why it sits at medium-low readiness. The capability that matters is preserving the click identifier from acquisition through to the value join, because a GCLID that is lost or unbound breaks the reconciliation and the value never reaches the right click.
Compare the tools on Martech Stack Builder
Marketing ops owns the value rules and conversion definitions inside the search platform and the mapping from LTV to bid signal. Data engineering builds the offline conversion import keyed on the click identifier and the join to the LTV score. Analytics validates that the high-value acquisition share actually rose and that blended CAC moved in a way the business is comfortable with, since the goal is better customers rather than cheaper ones. The team is small because the recipe is scoped to one channel and one import path. The recipe ships in weeks because the mechanism is well-documented and bounded to search. The work is in the click-to-value join and in having an LTV score worth importing; the platform side is configuration.
Work in this order. Capture the click, observe the value, join them, import, then let value bidding optimize.
Server-side event collection with client-side fallback covers steps 1, 2 and 4, inferred attribute generation steps 3, 5 and 6, and signal quality monitoring steps 7 and 8.
The first failure is the broken click-to-value join. If the GCLID is not captured at acquisition or is lost before the value is known, the LTV score has nothing to attach to, and the import either fails or attaches value to the wrong click. The discipline of preserving the click identifier from acquisition through to the eventual value event is the unglamorous core of the recipe.
The second is importing a value signal the bidding cannot act on well. Offline conversion import has timing constraints, and a value that arrives too late, or that is too sparse because only a fraction of customers have a confident LTV score yet, gives the algorithm too little to learn from. The recipe works best where a usable early value signal exists within the platform's attribution window, and is weaker where true value is only knowable months later.
The third is optimizing for value while watching only cost. If the team imports value but keeps judging the account on blended CAC, a correct shift toward higher-value, higher-cost customers looks like a regression. The scorecard has to be the value of acquired customers, not the price, or the recipe gets switched off for doing exactly what it was meant to do.
The Workshop works out with your team which of these matter for your stack right now, and what to do first: a 90-minute session with the people who own the decision.
The LTV signal worth importing, the click-to-value join that has to survive from acquisition to outcome, and the scorecard shift from cost to customer value: those are the decisions that turn search bidding from a cheapest-conversion engine into a high-value-acquisition one.
Did this recipe match your situation?Anonymous response. Sign up to leave a longer note tied to your account.
Activate the target account list with one identity resolution and one exclusion list across paid display, paid social, direct mail, and BDR outreach, so the same account stops getting contradictory treatment
Aggregate the signals from every contact in the account into one account score, because the buying committee is buying together and scoring leads in isolation misses the picture that matters
Lower acquisition cost by feeding the bidding algorithms better conversions, deduplicated, consent-valid, and weighted by value, rather than just more of them