CPM trends: 2025 vs 2026

Why no deltas, precisely

Gupta Media's trackers chart their own time series and are the right place to read Meta and Instagram movement; republishing their historical curve secondhand without their granularity would manufacture precision we do not have. TheB2BHouse and AdvertiseCast publish current aggregates without a frozen prior-year snapshot we can cite against. Every channel excluded from delta treatment is excluded by name in the table on /rates, which marks 13 of 19 rows quote-only. When our own quarterly snapshots accumulate (first review cycle: October 2026), this page becomes the delta ledger; the source ledger logs each snapshot date.

The structural forces, which ARE sourced

• Retail media's rise.dentsu's forecasts place retail media among the fastest-growing spend categories; budgets migrating into Amazon-shaped closed loops tighten demand there and relieve it elsewhere (context).
• Signal loss and the fee chain.The CMA's market study documented ~35% intermediary capture in open display, and the DOJ's adtech suit keeps restructuring pressure on the pipes that set effective prices (the chain).
• Macro spend. The IAB/PwC internet advertising revenue report is the neutral series for total demand growth, the tide all channel-level prices float on.

What a small advertiser should expect for late 2026

Mechanically safe expectations, no crystal ball required: Q4 will reprice your auctions upward as it does every year (the surge, worked); US targeting will stay premium to global averages (visible in the $13.48 vs ~$23 Meta spread today); and any specific percentage forecast you read without a dated source attached belongs in the folklore bin with the rest. Watch your own account's trend monthly; it is the one series nobody can misquote to you.

Build note, on the record: this slug hard-codes its years by design and gets re-pointed for 2027 at the annual refresh rather than silently redated.