The UK retail benchmark cohort we run at Bidfloor is now 42 accounts strong, up from 28 when we started tracking in Q1 2024. The accounts span home, apparel, electronics, and health and beauty, with a small number of specialist retailers in categories like garden, pet, and outdoor. Aggregate quarterly paid-search spend across the cohort is approximately £84m — meaningful enough to say something about the market, small enough that the numbers can still be checked by hand.
The headline read
Median non-brand cost per click, weighted by account spend, rose 4.2% quarter on quarter. The rise is broadly consistent with what we saw in Q1 2026 (4.5% QoQ) and Q4 2025 (3.8% QoQ). If you stretch back to Q2 2025, the trailing four-quarter increase is approximately 18%, and the trailing eight-quarter increase is approximately 34%. Non-brand retail search in the UK has become materially more expensive over the last two years, and the trend has not, on our data, slowed.
Brand CPC, on the same accounts and the same window, rose faster: 6.1% QoQ, and approximately 27% over the trailing eight quarters. The brand line is now, for the median account in our cohort, roughly 22% of total paid-search spend, up from about 16% in Q2 2024. The absolute count of brand impressions has risen slightly. The average bid required to defend a brand click has risen substantially. Someone is competing for the brand terms.
Who is bidding on your brand
The obvious next question is: who? The answer, on the impression-share reports we can access across the cohort, is disappointingly consistent. In roughly 78% of the accounts, the largest source of upward brand CPC pressure is a small number of category-adjacent competitors — often the two or three most direct rivals — using their own broad-match or automated bidding strategies in ways that pull them onto the target brand's terms.
The word "using" is doing some work in that sentence. In our conversations with the paid leads at the rival brands, most of them are not deliberately bidding on the target brand. They are running broad match, or Performance Max, or dynamic search ads, in configurations that permit the platform to serve their ads against brand queries where the platform's model believes the intent matches. From the target brand's perspective, this is indistinguishable from deliberate competitor bidding; from the rival brand's perspective, it is a byproduct of an automated setup they have not audited closely.
The Q2 story, more specifically, is that a small number of accounts in our cohort — five of them, to be exact — expanded their Performance Max spend meaningfully during the quarter without also adding campaign-level brand exclusions on their competitors' terms. The impressions those five accounts absorbed against competitor brand terms drove approximately 40% of the sector-wide brand CPC increase for the quarter, by our decomposition. The impact is that concentrated.
What this costs a typical account
For the median account in our cohort, running roughly £200,000 of paid search a month, the brand-side CPC inflation over the last twelve months has cost approximately £14,000 a month in additional brand-defence spend for the same set of brand impressions. Annualised, that's £168,000 the median account is now spending simply to hold the ground it held a year ago.
The number is small compared to total paid search spend at the account level. It is large compared to the incremental value the brand-side spend is producing. Brand search has always been the paid line item with the weakest incrementality case; when it becomes more expensive without becoming more incremental, the marginal economics degrade quickly.
"Every account in our cohort could reduce its brand-side spend by 30-40% tomorrow without measurable revenue impact, if it were willing to accept the impression-share loss. Almost none of them are willing to. That reluctance is what's driving the CPC line."
What to do about it
Three practical moves are worth considering, in roughly the order the accounts we advise have found them useful.
The first is to reduce brand-side aggression. Most retail accounts in our cohort are running exact-match brand campaigns with a target impression share of 90-100%. That target is defensible if the incremental cost of defending each impression is low; it is not defensible if the marginal impression is now costing three times what it did in 2023. Dropping target impression share to 70-80% typically cuts brand spend by 25-35% while losing only a small share of high-intent clicks (the rest are absorbed by organic listings, sitelinks, or the merchant's own shopping placements).
The second is to add competitor-brand exclusions at the campaign level, where the platform allows it. This does not stop competitor accounts bidding on your brand; it does stop your own PMax and broad-match campaigns from bidding on their brands, which reduces the reciprocal pressure that drives the arms race.
The third is to run a honest brand-search incrementality test at least once a year. The test costs 4-6 weeks of one geographic region's brand spend and produces a defensible number for the incremental value of the line. In our sample of accounts that have run this test, the median finding is that brand-side incremental ROAS is roughly 30% of the platform-reported ROAS. Once that number is on the table, the internal argument about brand-side aggression shifts substantially.
Looking ahead
Our forecast, based on the current trajectory and the composition of the cohort, is that non-brand retail search CPC will rise a further 3-5% in Q3 2026, driven primarily by the seasonal ramp in home and apparel. Brand CPC will likely rise 4-6% in the same window, driven by the same PMax dynamics that produced Q2's result. The trailing four-quarter figures will, if the trend holds, print at roughly 20% non-brand inflation and 26% brand inflation.
These are not comfortable numbers to defend to a CFO. They are the numbers the UK retail search auction is currently producing, and pretending otherwise does not change them. The teams that have been most effective in the last two quarters, on the accounts we watch, are the ones who have stopped optimising the auction and started managing their exposure to it — accepting a lower impression share here, a narrower keyword footprint there, a more conservative bid strategy in the third place — in order to keep the aggregate cost of the line under control while the auction re-prices around them.