On 14 October 2026, Google released Google Ads API v26 with Search Lift measurement, adding 15 new metrics, a read-only SearchLiftKeywordBundle resource, and two new segments that make Search Lift results queryable for the first time. Search Lift answers a question standard conversion reporting cannot: did your ads create additional searches for your brand or category, or did they only harvest demand that already existed? For B2B SaaS, that is the measurement that finally separates paid search’s cannibalization problem from its demand-creation value.
This post explains what Search Lift measures, how it differs from the Conversion Lift expansion Google shipped the same month, whether paid really cannibalizes your organic brand demand, and how to fold a search-lift read into a measurement stack without letting it contaminate your bidding. The release also marks the start of Google’s twice-monthly API release cadence, so measurement surfaces will now change faster than the old quarterly rhythm.
What Search Lift measures that attribution cannot
Search Lift estimates incremental searches — the additional brand or category queries that happened because people were exposed to your ads. It works like other lift studies: a portion of the audience is held back from seeing your campaigns, and Google compares search behaviour between the exposed and holdback groups. The gap is the lift. In API v26 that read is exposed through metrics such as relative_search_lift, normalized_incremental_searches, and search_lift_impressions, segmented by a SearchLiftKeywordBundle that represents the set of queries a study is tracking.
This matters because attribution is structurally blind to demand creation. When a Demand Gen or video campaign prompts a prospect to search your brand three days later and convert through an organic or branded-paid click, last-click and even data-driven attribution assign that upper-funnel campaign nothing — the conversion lands on a different touchpoint. Search Lift captures the created search directly, which is the missing half of the story for any SaaS running top-of-funnel spend. It is the counterpart to the holdout logic behind Conversion Lift for Search and PMax, applied one stage earlier in the funnel.
Search Lift vs Conversion Lift: which question each answers
The two studies are easy to confuse and measure different things. Conversion Lift isolates incremental conversions; Search Lift isolates incremental searches. For a long-cycle B2B SaaS account, that distinction decides which study is even feasible: qualified conversions are scarce, but searches are plentiful, so Search Lift can produce a statistically valid read where a user-based Conversion Lift study stalls below its volume floor.
| Dimension | Search Lift | Conversion Lift |
|---|---|---|
| Outcome measured | Incremental searches (brand / category queries) | Incremental conversions (leads, trials, sales) |
| Funnel stage | Upper funnel — demand creation | Lower funnel — demand capture |
| Best-fit campaign types | Demand Gen, Video, Performance Max | Search, Performance Max |
| Volume constraint for SaaS | Lower — searches are high-volume | High — needs ~1,000 observed conversions |
| What it proves | The ad created demand | The ad created a customer |
A mature stack uses both: Search Lift to justify upper-funnel budgets that attribution under-credits, and Conversion Lift to validate that the demand actually converts. Neither replaces the attribution-based conversion data that Smart Bidding and your board metrics still run on day to day.
Does paid search cannibalize your brand demand?
For some of your spend, yes. Branded search is the textbook over-crediting case: a large share of conversions credited to a brand campaign come from buyers who already knew you and would have clicked your organic listing for free. We quantify that dynamic in our analysis of branded-search conversion measurement, and the same effect inflates Performance Max brand cannibalization. Those are real cannibalization problems that a conversion-only view hides.
But cannibalization is only half the ledger. Upper-funnel campaigns frequently create brand searches that did not exist before, and standard attribution gives them no credit for it. Search Lift is the tool that separates the two: if a Demand Gen campaign shows a positive, significant lift in brand searches, it is generating demand, not just intercepting it; if it shows none, you are paying to reach people who were never going to search for you. The honest read for most B2B SaaS accounts is a mix — some spend cannibalizes, some creates — and the point of the v26 metrics is that you can finally tell which campaigns fall on which side instead of defending or cutting the whole line item on last-click credit.
Running a Search Lift study on a low-volume SaaS account
The practical appeal of Search Lift for B2B SaaS is volume. User-based Conversion Lift needs roughly 1,000 observed conversions to resolve, which a correctly-optimised SaaS account tracking SQLs or closed-won deals may take a year to accumulate — the same thin-data constraint we cover in running Google Ads on thin data. Searches are orders of magnitude more frequent than qualified conversions, so an account generating only a few dozen SQLs a month can still drive the thousands of impressions and searches a lift study needs to reach significance.
Eligibility still depends on campaign type, spend, and whether the study is available through a Google rep or beta for your account rather than a single published self-serve threshold, so confirm access before you plan around it. Design the study the way you would any experiment: pick the keyword bundle that represents genuine brand or category intent, run it on the upper-funnel campaigns most likely to create demand, and give it a window long enough to clear noise. The failure mode is treating a near-zero or non-significant lift as proof the campaign works — a flat result is itself a finding, and usually an argument to reallocate.
Where Search Lift belongs in the measurement stack
Search Lift is a validation metric, not a bidding signal, and the most common mistake will be wiring it into the wrong place. Smart Bidding optimises against your conversion actions, so Search Lift must not displace conversion tracking, offline conversion imports, or the enhanced conversions that feed the algorithm. Keep it at the budget-planning altitude: it tells you whether an upper-funnel campaign that looks weak on CPA is quietly manufacturing the brand searches that convert elsewhere, which is precisely the campaign attribution tempts you to cut.
Sequenced correctly, the stack reads like this: attribution-based conversion data steers daily bidding and optimisation; Search Lift validates that upper-funnel spend creates demand; Conversion Lift validates that the demand converts; and marketing mix modeling with Meridian provides the account-level read when campaign-level experiments are not feasible. The v26 release does not change what you optimise toward — it adds one more layer of causal evidence for the budget decisions where last-click credit is least trustworthy, and it does so for the upper-funnel spend that previously had no incrementality read at all.
Decide whether Search Lift is worth running for your account
The decision sequence is: identify which campaigns are genuinely upper-funnel and plausibly create demand (Demand Gen, Video, broad PMax), confirm Search Lift availability for your account, then run a study on the brand and category queries those campaigns should influence. If the lift is positive and significant, you have the evidence to protect that budget against a CPA-only challenge; if it is flat, you have a clean reallocation case. Running no study at all leaves upper-funnel spend defended or cut on attribution that is structurally blind to demand creation.
If you want help deciding which studies your account can actually support, a Google Ads audit establishes whether your conversion tracking and campaign structure can carry incrementality measurement, and our Google Ads management service builds the full stack — attribution to steer, Search Lift and Conversion Lift to validate — around a B2B SaaS sales cycle. To check whether your tracking is clean enough to measure lift at all, start with the free 10-point audit checklist.