Performance Max drives 45% of Google Ads conversions for advertisers using it. In January 2026, Google expanded the per-asset-group search theme cap from 25 to 50 and shipped a usefulness indicator labelling each theme Useful, Limited Usefulness or Not Useful. Most WooCommerce stores filled the new cap with brand variants and category terms. The indicator is now exposing that the majority of those themes duplicate queries PMax would have found anyway — and worse, route paid budget against branded queries the store wins organically for free.

Read your usefulness indicator before you add another theme. The 50-slot cap is capacity, not a target.

What the Usefulness Indicator Actually Shows

The indicator is a three-state label sitting next to each search theme inside an asset group. The states are straightforward.

• Useful. The theme is surfacing queries that PMax’s keywordless targeting wasn’t already covering. Incremental coverage. Keep it.
• Limited Usefulness. The theme is doing some work, but most of its query coverage overlaps with what PMax would have found on its own. Marginal value — review against the data underneath.
• Not Useful. The theme is duplicating queries PMax already finds without it. Remove it. The slot is wasted.

The companion change shipped alongside it: a new source column in search terms insights. For the first time, Google tags each surfaced query with whether it came from PMax’s keywordless targeting or from an advertiser-provided theme. That column is the audit trail. The usefulness indicator is the summary score.

Where to Find It

Open the Performance Max campaign, click into the asset group, and scroll to the Search themes panel. The label sits beside each theme. To see the queries underneath, open the campaign’s search terms insights and look at the new source column — it tells you exactly which themes are pulling their weight on which queries, and which themes are pulling queries the campaign was already going to get.

This is the first time Google has shipped a built-in cannibalisation signal inside the Performance Max interface. Use it.

The 50-Slot Trap

When the cap expanded from 25 to 50 in January 2026, the natural response across the WooCommerce advertiser community was to fill it. More themes felt like more coverage. The usefulness indicator is the post-hoc check on that instinct, and it tends to land hard for stores that filled the slots quickly with two common patterns.

Pattern one: brand variants. Misspellings of the store name, abbreviated forms, product-line subbrands. These are the queries the store wins on the organic free listing for almost zero acquisition cost. Adding them as themes routes paid budget against the same searches.

Pattern two: broad category terms. Generic descriptors of the product catalogue. PMax’s keywordless targeting was already finding most of these queries through the product feed alone. The theme adds no coverage — it just labels the spend differently.

If your store’s PMax campaigns lean on either pattern, the usefulness indicator probably flags 60-70% of the asset group’s themes as Limited Usefulness or Not Useful once the campaign accumulates enough conversion data to compute the score.

You may be interested in: Google Finally Opened the Performance Max Black Box — But Your WooCommerce Store Still Can’t Spend the Insights — the broader frame on why PMax visibility upgrades don’t help if the underlying conversion data is incomplete.

The Brand Cannibalisation Pattern

This is the dynamic most WooCommerce store owners don’t see until the usefulness indicator surfaces it. A search for the store’s brand name triggers two listings on the same results page — the paid PMax ad and the organic blue link the store already ranks for. A click on either counts as a conversion. A click on the paid ad costs money. A click on the organic listing doesn’t.

PMax’s reporting attributes the converting clicks to PMax regardless of whether the searcher would have clicked the organic listing for free. The result is a campaign that appears to be performing well on branded queries — high ROAS, low CPA — while actually displacing free traffic with paid traffic at zero incremental gain.

The usefulness indicator can surface this. It cannot quantify the lost organic traffic. For that, you need conversion data with click-id and organic-vs-paid provenance preserved — and Google Ads alone doesn’t give you the second half of that pair.

The Data Integrity Caveat

Here’s the part the announcement coverage doesn’t mention. The usefulness score is computed on the conversion data Google sees flowing into your Google Ads account, which on most WooCommerce stores means a mixture of GA4 imported conversions and direct conversion tracking. SR Analytics audit data suggests 73% of GA4 implementations have silent misconfigurations causing 30-40% data loss.

If your GA4 is missing a third of your conversions, Google’s usefulness scoring is computed on the two-thirds it does see. A theme flagged Not Useful might genuinely be Not Useful — or it might be Useful in a way the conversion pipeline isn’t recording. Either way, the indicator is a starting point, not the verdict.

This is the same incomplete-conversion-base problem that affects PMax channel reporting and asset group ROAS. For the deeper context on that pattern, see Google’s New Performance Max Channel Reporting — Your Asset Group ROAS Is Calculated on Half Your Conversions.

The BigQuery Incrementality Check

The reliable answer to “is this theme adding incremental traffic” is a query you run against your own conversion data, not Google’s internal score. The query has the same shape every time.

For each search theme, isolate the queries the source column tags as coming from that theme. Compare the conversion rate and revenue on those queries against the organic baseline for the same query class over the same window. If the paid conversion rate exceeds the organic baseline by a meaningful margin, the theme is adding incremental traffic. If it doesn’t, the theme is displacing organic clicks with paid clicks.

This query requires conversion data with the source/medium and click-id preserved alongside the WooCommerce order. Google Ads Reports alone can’t run it. GA4 alone usually can’t either — by the time the data lands in GA4, the click-id provenance has typically been lost or sampled.

Transmute Engine™ is a dedicated Node.js server that runs first-party on your subdomain (e.g., data.yourstore.com). The inPIPE WordPress plugin captures WooCommerce conversion events with full source/medium and click-id provenance and sends them via API to your Transmute Engine server, which streams them into BigQuery alongside your Google Ads extracts. The incrementality query runs against that warehouse — independent of Google’s own usefulness score and independent of GA4’s misconfigurations.

Key Takeaways

• The usefulness indicator labels each theme Useful, Limited Usefulness or Not Useful — read it before adding more themes.
• 50 slots is capacity, not a target. PMax’s keywordless targeting already finds most of the queries you’d add themes for.
• Brand themes route paid budget against queries you win organically — the indicator surfaces this, but the lost organic traffic isn’t in the score.
• The new source column in search terms insights is the audit trail — use it to see which themes drive which queries.
• Google’s score is computed on GA4 data, and 73% of SMB GA4 implementations leak 30-40% of conversions — treat the indicator as a starting point, not the verdict.
• The reliable check is an incrementality query against your own BigQuery conversion data with click-id and source/medium preserved.

Frequently Asked Questions

Audit your PMax search themes usefulness indicator this week and exclude every theme flagged Not Useful — then build the BigQuery incrementality query to verify Google’s score against your own conversion data — seresa.io.