Analyzing 1,000 Ads: Spend, CPM, and Format Performance in Spring 2026

Analyzing the Spring 2026 Ad Cohort

Between February 6, 2026, and May 7, 2026, we examined an aggregate sample of 1,000 distinct ads. This cohort represents 270 ad accounts and a combined total spend of exactly €186,776.36. Looking at aggregate spend alone often masks the operational reality of how media buyers actually deploy capital. By breaking down the median distributions, a very specific methodology emerges.

The overall numbers point to an environment where buyers favor high-velocity testing over sustained delivery. We see ad accounts heavily fragmented into hundreds of micro-campaigns.

The Dominance of Micro-Testing

Spend Distribution

The most telling metric in this cohort is the median spend per ad, which sits at just €18.14. Even when we look at the 75th percentile, the spend only rises to €23.78.

This tight clustering at the lower end of the budget spectrum reveals a ruthless approach to creative testing. Media buyers are not launching ads and letting them run for weeks to gather mature data. Instead, they are deploying automated rules. An ad goes live, spends a few euros to test initial viability, and is promptly paused if it does not hit a strict performance threshold.

When 1,000 ads share roughly €186,000 in spend, the mathematical mean sits much higher than the median. This indicates an extreme rightward skew. A tiny fraction of ads scale up to consume the bulk of the budget, while the vast majority are killed in their infancy.

Reach and Saturation

This hyper-fragmentation is also reflected in the reach metrics. The median total reach within the European Union for an ad in this cohort is a mere 85 users.

You cannot build broad top-of-funnel awareness when your median ad reaches fewer than one hundred people. These metrics confirm that the buyers in this dataset are optimizing heavily for direct response. They are fishing for an immediate conversion or click signal. When the 75th percentile for daily reach is 1,428.57 users, it becomes clear that delivery is heavily constrained. These constraints are likely enforced either by granular audience targeting or by the micro-budgets allocated to each specific creative.

CPM Realities and Pricing Compression

One of the most unusual findings in this 90-day window is the extreme stability of the CPM. The median CPM across the entire cohort is €0.329.

What is remarkable is the lack of variance. The 25th percentile CPM is €0.32. The 75th percentile CPM is €0.3298. There is almost zero price fluctuation across the middle fifty percent of the data.

In a typical varied account structure, we expect to see a wider spread based on different audience segments, placement quality, and bidding models. The fact that pricing is locked tightly between €0.32 and €0.33 suggests that these campaigns are heavily reliant on fixed bid caps or cost-cap bidding strategies. Alternatively, the algorithms may be aggressively routing impressions to the cheapest available inventory pools, such as secondary feeds or right-column placements, to sustain the high volume of low-cost creative testing.

Format Variations and Performance

When we segment the cohort by creative format, the disparity in how buyers treat different assets becomes stark.

Video Ads

Video ads represent 271 units in this dataset. Astonishingly, the median spend for a video ad is just €1.60. The median CPM for these video assets aligns with the broader cohort at €0.32.

A median spend of €1.60 means that buyers are cycling through video creatives at a rapid pace. They are clearly demanding immediate engagement. If a video does not capture attention within the very first few impressions, the automated rules terminate it. The production cost of video might be higher, but the media tolerance for underperforming video assets is effectively zero.

Image Ads

Image ads make up 129 units in the sample. Here, the behavior shifts slightly. The median spend for static images is €18.14, matching the overall cohort median. The median CPM for image assets is €0.3298.

Buyers appear slightly more patient with image assets, allowing them to spend ten times more than the median video ad before making a pause decision. This might be due to the fact that static images require less initial engagement data to prove their conversion viability compared to video, which relies heavily on watch-time metrics.

Dynamic and Unknown Formats

The largest segment belongs to the unknown format category, encompassing 599 ads. These are typically dynamic creative structures, platform-specific multi-asset carousels, or catalog ads that do not report as a single static file. The median spend for this group is €23.78, with a median CPM of €0.329. The higher spend indicates that algorithms are being given slightly more budget to test the various permutations within these dynamic ad units.

Implications for Media Buyers

The data from this Spring 2026 cohort paints a clear picture of modern ad operations. High-volume, low-spend testing is the dominant strategy.

To operate in this manner requires flawless naming conventions, robust tracking, and a heavy reliance on automation. No human media buyer can manually monitor and pause a video ad exactly when it hits €1.60 in spend. Success in this environment depends on setting up strict, emotionless rules that cut the losers instantly and funnel remaining capital into the rare assets that prove their worth within the first hundred impressions.