Marketing budgets across a matched group of European advertisers rose while the price of a click fell, according to benchmark data published on 4 August 2026 by Billy Grace, an Amsterdam-founded marketing intelligence company. The clearest structural shift in the dataset sits in channel allocation: search lost close to five percentage points of budget share in twelve months, and the money moved into a widening set of smaller platforms.

The findings come from a benchmark report titled "A start to a year that rewards conviction, not caution," which compares the first half of 2026 against the first half of 2025 across a matched cohort of several hundred advertisers. Every business in the group was an active paid-media investor in both periods, a constraint that excludes advertisers present in only one half and is intended to isolate genuine year-on-year movement from changes in who is being measured.

According to Billy Grace, the typical advertiser in the cohort lifted paid media investment by 17.5% while paying 10.8% less for each click. Click-through rate improved 13.0%, moving from 1.25% to 1.41%. The cost of reaching a thousand people rose 1.3%, a figure the company describes as essentially flat. Roughly two in three advertisers, 64% of the group, spent more than they had a year earlier.

Efficiency moved without triggering inflation

The combination is what makes the dataset unusual. Rising budgets across an advertiser population normally bid up auction prices, and much of the commentary in European performance marketing over the past two years has assumed that outcome. In this cohort it did not happen at the aggregate level. Median cost per click fell from EUR 0.53 to EUR 0.47. Clicks purchased per unit of spend rose 12.1%.

Cohort-wide measured investment, a spend-weighted figure rather than a median, grew about 14%. That gap between roughly 14% and 17.5% is not a contradiction. The larger advertisers in the group, who dominate a spend-weighted total, grew investment and outcomes broadly in step, while the sharper efficiency gains sat further down the size curve among small and mid-sized advertisers. The report states this explicitly as one of its five closing observations.

One allocation number moves alongside the cost picture. The share of investment directed at mid- and upper-funnel activity rose 3.9 percentage points, indicating that the additional budget was not simply poured into the same bottom-funnel placements.

A front-loaded year that cooled in June

The two halves followed different rhythms. In 2025, spending built gradually from January toward a June peak. In 2026, advertisers front-loaded: January budgets came in well above the prior year, and the pace slackened in the closing month. The 2026 line sat above 2025 in every month of the half when each year is indexed to its own January.

June 2026 is the exception that carries the most weight for anyone modelling the second half. It recorded the weakest year-on-year investment growth of the six months and the highest cost of reach in the entire two-year window, with median CPM reaching EUR 7.22. That single month is what pulled the half-year CPM average up to its otherwise flat +1.3%. Cost per click also began climbing toward the prior year's level in June after five months of sitting below it. The widest gap between the two years fell in May, when 2025 costs peaked at EUR 0.586.

The sequencing matters: the one month in which advertisers eased off spending was also the month in which reach became most expensive.

Search share slips as challenger channels absorb the difference

Google remains the anchor of nearly every media plan in the cohort, but its share of spend-weighted investment moved from 62.1% to 57.3% year on year, a fall of 4.8 percentage points. Meta picked up 0.9 points, moving from 19.7% to 20.6% and holding its position as the second anchor. Everything else combined rose 3.9 points, from 18.2% to 22.1%.

A share decline is not a spend decline. Alphabet reported second-quarter 2026 results on 22 July showing Google Search advertising revenue rising 17% to $63.3 billion while the Google Network segment slipped 1%. Both things can hold simultaneously: search budgets grow in absolute terms while occupying a smaller proportion of a media plan that is itself expanding.

Growth rates on the smaller platforms are steep, and the report is explicit that they start from a small base. Redditinvestment across the cohort grew 650%. TikTok grew 65%, Pinterest 42%, Bing 34% and YouTube 15%. Reddit's own second-quarter figures, reported on 30 July, put advertising revenue at $762 million and growth at 64%, which places the cohort's much larger percentage movement in perspective: a small number of advertisers moving onto a platform for the first time produces a far higher ratio than a platform-level revenue line.

Adoption ran net positive on every emerging channel

Underneath the spend figures sits a measure the report calls adoption momentum: the number of advertisers who added a channel for every one who dropped it. A ratio above one indicates net inflow. Bing recorded 4.6 new advertisers for each departure, TikTok 4.5, Reddit 4.4, YouTube 3.5, Pinterest 3.0 and Snapchat 2.3.

That measure separates genuine adoption from concentrated budget increases by a handful of existing accounts. Every emerging channel in the dataset ran a clear net inflow, which is the mechanism producing the 3.9-point gain in the "other" bucket rather than a small number of large advertisers making outsized bets.

The direction aligns with separate research on channel breadth. Funnel data published on 28 July found TikTok entering the top five advertising platforms in EMEA for the first time, with regional spend rising from $2 million monthly in January 2024 to $12 million in April 2026, while Social moved from 18% to 33% of total EMEA spend over the same window. Concentration remains the baseline against which these shifts are measured: UK market data covered in May showed two of every three pounds in advertising going to Google, Meta and Amazon.

Ecommerce grew on volume with cost per outcome holding

Product sellers in the cohort lifted investment 9.9% and won 26.2% more clicks as cost per click fell 10.4%. Click-through rate rose 11.7%. Cost per thousand impressions rose 2.7%, the one traffic metric in the segment that did not clear the report's significance threshold.

On the outcomes side, conversions rose 9.8% and conversion value 9.6%. Return on ad spend improved 6.0%. Cost per acquisition fell 1.6%, average order value rose 0.9%, and conversion rate per click softened 4.7%. Of those four, only return on ad spend cleared statistical significance; cost per acquisition, order value and conversion rate did not.

The softening conversion rate alongside rising total conversions is the arithmetic of a widening funnel. More clicks entering at a slightly lower average yield produces exactly that pattern, and the report treats it as the texture of scaling rather than a deterioration, on the grounds that order value held and acquisition cost stayed flat.

The comparison against other European ecommerce datasets is not uniform. Channable's benchmark, drawn from EUR 1.38 billion in verified Google Ads spend across more than 10,000 European merchants and published in July, found advertisers losing 46% of return on ad spend as Google click costs rose. WordStream's May study of 13,474 US campaigns recorded cost per click climbing to $5.42 even as cost per lead fell for the first time since 2020. The three datasets measure different populations, different geographies, different channel mixes and, critically, different attribution models, which is the most likely source of the divergence.

Lead generation recorded the steepest cost fall

Lead-generation advertisers scaled harder than any other group in the study. Investment rose 25.7% and clicks rose 49.3%, while cost per click fell 21.1%, the steepest decline of any segment in the report. Click-through rate gained 6.8%. Cost per thousand impressions eased 11.8%, a directional movement that did not reach significance.

The mechanics behind the click-cost fall run through engagement rather than sacrifice. Click-through rate improved by a statistically significant margin while reach costs eased by a larger but non-significant amount. In other words, the price of a click dropped because audiences engaged more readily on broadly cheaper inventory, not because advertisers traded quality of placement for volume.

Billy Grace flags the downstream exposure directly. Cost per click carries no information about lead quality, and a funnel expanding at nearly 50% more clicks can fill with traffic that never reaches a pipeline. The report frames the risk as sitting after the media buy rather than inside it.

Verticals diverged, and the boldest numbers sit on the thinnest samples

Investment growth by industry ranged from +68.2% to -38.4%. Public sector and nonprofit recorded the largest increase at 68.2%, followed by travel, transport and hospitality at 63.4%, financial services at 61.9%, health, wellness and life sciences at 47.5%, food, beverage and agriculture at 35.9%, retail and consumer goods at 28.6%, and media, entertainment and education at 17.7%. Industrial and energy was the only sector to cut budgets, down 38.4%.

Only four of those movements cleared the significance threshold: public sector and nonprofit, travel, retail and media. Financial services, health, food and industrial are marked as directional. The company's own accompanying release describes travel as posting the sharpest rebound at 63%, a characterisation that sits slightly below the public sector figure shown in the report's chart, and the divergence is not reconciled in the published material.

Cost per click by industry moved in both directions. Public sector and nonprofit clicks fell 34.1%, retail and consumer goods 16.3%, media and education 10.0%, industrial and energy 5.9%, and travel 4.6%. Three sectors paid more: food and beverage clicks rose 15.8%, financial services 23.5%, and health and wellness 24.2%. Only the retail and media declines are marked as significant.

Retail and consumer goods carries the largest and most stable sample in the study, and it is the only vertical for which Billy Grace publishes ratio measures. Retail return on ad spend rose 13.1%, from 7.72 to 8.73, and cost per acquisition fell 3.7%, from EUR 15.06 to EUR 14.51. For every other vertical, the company states that attribution gaps in H1 2025 left too few advertisers with valid values in both periods to compute a reliable ratio.

Method, and what the figures do not cover

All comparisons run through Billy Grace's Unified Marketing Measurement model, which combines marketing-mix modelling with multi-touch attribution and is applied consistently across both periods. The company argues this captures impression-led and view-through demand that last-click reporting misses, which matters for a report whose central finding concerns budget moving into channels that generate exactly that kind of demand.

That claim is not unique to this dataset. Kochava research first published in September 2025 and restated in the company's second-quarter bulletin found marketing-mix modelling credited TikTok campaigns with 35% higher incremental impact than last-touch attribution recorded. Survey work from TransUnion and EMARKETER found platform-provided attribution still the most common methodology at 65.8%, with 46.9% of marketers planning to increase marketing-mix modelling investment.

Several limits sit in the published material. Headline figures describe the median advertiser rather than the market, and where a spend-weighted market-level figure is reported it is labelled as such. Significance is set at a probability below 5%. The smallest verticals are suppressed. Cohort size is disclosed only as several hundred advertisers, without a stated count. The report figures are denominated in euros, while the accompanying release describes the tracked adspend in pounds, a currency inconsistency the source material does not resolve. Billy Grace measures advertisers on its own platform, which is not a random sample of the European market.

Founded in Amsterdam in 2022 and backed by Fortino Capital, the company expanded into the United Kingdom in 2026. Mitch Voskuilen, co-founder and chief executive, said: "When budgets are rising and costs are falling in the same market, standing still is the one thing advertisers can't afford. The businesses that got ahead this year weren't necessarily the ones who spent the most. They were the ones who treated a flat budget as a real cut, tested new channels while they were still cheap, and judged results on the full picture rather than whichever number a platform hands back."

Why this matters for the marketing community

The practical weight of the report sits in the interaction between two of its numbers. A budget held flat in a market where clicks cost 10.8% less buys measurably less growth than the same budget bought a year earlier, because competitors holding the same euro figure are acquiring more traffic for it. That is an arithmetic point about relative position rather than a recommendation, and it applies with more force in lead generation, where click costs fell 21.1%.

The channel-share movement carries a separate implication for media planning. Search losing 4.8 points of budget share while Alphabet's search revenue grows 17% describes a market in which the dominant channel is expanding in absolute terms while shrinking as a proportion of the plan. Planning teams working from platform-reported growth alone would not detect that shift. It only appears in advertiser-level allocation data.

The flat aggregate CPM also runs against several adjacent datasets. US programmatic CPMs rose 33.9% year over year in April 2026, according to DataBeat figures covering more than 35 billion monthly impressions. A cohort of European search and social advertisers experiencing near-flat reach costs is measuring a different inventory pool than an open programmatic exchange, and the divergence is a reminder that "the cost of reach" is not a single market price.

Finally, the measurement dependency is worth stating plainly. A finding that impression-led channels deserve more budget rests on a model built to credit impression-led channels. That does not invalidate the result, but it does mean the conclusion and the methodology point in the same direction, and any advertiser comparing these benchmarks against last-click platform reporting will be comparing outputs of two systems that count conversions differently. Debate over platform concentration has run through the industry all year, with Rory Sutherland arguing in July that routing most budget through two platforms hands them structural control over an advertiser's business. This dataset supplies allocation figures to a discussion that has largely been conducted without them.

Timeline

Summary

Who: Billy Grace, a marketing intelligence company founded in Amsterdam in 2022, backed by Fortino Capital and expanded into the United Kingdom in 2026. Mitch Voskuilen, co-founder and chief executive, provided commentary. The dataset covers a matched cohort of several hundred European advertisers spanning ecommerce, retail and lead-generation businesses.

What: A benchmark report comparing the first half of 2026 against the first half of 2025. Median paid media investment rose 17.5%, cost per click fell 10.8%, click-through rate rose 13.0% and cost per thousand impressions rose 1.3%. Google's spend-weighted share of investment fell from 62.1% to 57.3%, Meta gained 0.9 points to 20.6%, and all other channels combined gained 3.9 points to 22.1%. Reddit spend grew 650%, TikTok 65%, Pinterest 42%, Bing 34% and YouTube 15%. Lead-generation advertisers recorded the steepest click-cost decline at 21.1%.

When: The measurement window covers 1 January to 30 June in both 2025 and 2026. Billy Grace published the report and accompanying release on 4 August 2026.

Where: The cohort consists of advertisers active on the Billy Grace platform across Europe, with figures reported in euros.

Why: The report documents a period in which rising advertiser budgets did not produce the auction inflation that typically accompanies them, and in which search lost close to five percentage points of budget share to a widening set of smaller platforms. Both findings bear on how media plans are constructed and how flat budgets are evaluated in relative terms. All figures rest on a unified attribution model combining marketing-mix modelling with multi-touch attribution, which credits impression-led channels that last-click reporting does not, meaning the results are not directly comparable with platform-reported performance data.