Amazon Prime Day 2026 ad benchmarks

Marketplace Ad Pros publishes this study from its own customer base. The numbers below are aggregate and scale-free -- median per-brand percentage changes, rate metrics, and distribution statistics. No individual advertiser's data appears here, no totals are summed across brands, and cohort coverage is reported as a share of the cohort rather than as advertiser counts.

Last updated: July 25, 2026

TL;DR

We measured several hundred US advertisers across Prime Day 2026 (June 23-26) against a matched baseline. Six findings:

1. The typical brand grew. Median Sponsored Products ad sales rose +30%. Median total sales (Seller Central) rose +37%, and 76% of brands grew total sales.
2. Spend rose without budget increases. 75% of brands spent more on Sponsored Products, yet only 13.7% raised daily budget by 20%+ and 51% actually cut budget. Prime Day demand filled the budgets brands already had.
3. Raising budget didn't buy a better outcome. Budget-raisers saw the same median ad-sales lift (1.30x) as budget-holders (1.30x).
4. Auctions ran hotter, but not dramatically. Median SP CPC +10% ($1.19 → $1.31). Median ACOS moved 33% → 35%. SB CPC barely moved (+2%).
5. Sponsored Display was the exception. The median advertiser pulled back -- spend −17%, impressions −37% -- even as SD efficiency improved (ACOS 34% → 25%).
6. Prime Day is not a guaranteed win. SP spend ranged from 1.00x at the 25th percentile to 1.87x at the 75th; ad sales from 0.89x to 1.95x. Roughly a quarter of brands spent less, sold less, or both.

Methodology

• Cohort: several hundred US advertisers, with 97% of the intended cohort successfully processed. We report cohort size as a range rather than an exact count, and channel coverage as a share of the cohort.
• Event window: June 23-26, 2026 (Tuesday-Friday, 4 days).
• Baseline: matched Tuesday-Friday days from the two weeks before the pre-event ramp, compared as per-day averages. Using pre-ramp days matters -- brands start building in the two weeks prior, so a baseline drawn from the immediately preceding weeks would understate the lift.
• Marketplace: US only, USD. Other marketplaces would run as separate cohorts.
• Metrics: per-brand indexed change (event per-day ÷ baseline per-day), reported as medians with 25th-75th percentile spread, plus rate metrics and "% of brands" shares.
• Attribution maturity: pulled at roughly 95% maturity. June 26 clicks fully mature around July 10, so event ad sales are slightly understated -- the error runs in the conservative direction.

Per-channel results

Coverage is the share of the cohort with meaningful activity in that channel, which is also a rough measure of how much weight to put on each row.

* We deliberately omit Sponsored Display CPC and CPM. SD blends costType (CPC and vCPM in the same account), so a blended SD CPC is an artifact rather than a real bid movement. Read SD on spend, sales, and impressions only -- and be skeptical of any Prime Day report that quotes an SD CPC without splitting by cost type.

The budget finding

This is the result that most contradicts standard Prime Day advice.

Three-quarters of brands spent more, but only one in seven opened their budgets. The extra spend came from Prime Day traffic pushing more volume through budgets that were already in place.

And the brands that did raise budgets did not do better: median ad-sales lift was 1.30x for budget-raisers versus 1.30x for budget-holders.

Read this carefully -- it's correlational, not causal. Brands often raise budgets because they're already winning, so the comparison isn't a clean experiment. What the data does rule out is the simple version of the advice: raising your budget is not, by itself, what produces a Prime Day lift.

What 76% of brands did do was launch new campaigns, and 53% advertised new products. Catalog and campaign refresh was far more common than budget expansion.

Category leaderboard

Median Sponsored Products spend index, for categories with at least six advertisers in the cohort. Category is derived per advertiser from the top-spend advertised ASIN.

Electronics, Home & Kitchen, Baby, and Automotive leaned in hardest. Office Products barely moved.

Cell sizes vary a lot across this table. Health & Household, Home & Kitchen, and Beauty & Personal Care are among the best-represented categories here; Electronics, Baby Products, and Patio, Lawn & Garden sit near the six-advertiser floor, so read those as directional rather than precise.

What this suggests for next Prime Day

• Check budget caps, don't reflexively raise them. Most brands didn't raise budgets and still spent more. The useful work is making sure a cap doesn't truncate a good day, not bidding up the whole account in advance.
• Expect CPC inflation of roughly 10%, not 50%. Plan efficiency targets around a modest auction premium.
• Don't judge the event on ACOS alone. Median total sales rose faster (+37%) than median ad sales (+30%), so ad-attributed share of revenue falls for many brands even when the event goes well. If you only look at ACOS, a good Prime Day can read as a bad one.
• Decide about Sponsored Display deliberately. The median advertiser pulled back and their SD efficiency improved. That may be right for you -- but make it a decision rather than a default.
• Refresh campaigns and products. That is what the majority of the cohort actually did.

Limits of this study

• Correlational, not causal, especially the action-to-outcome comparison.
• US only. Other marketplaces behave differently and would need separate cohorts.
• Ad sales are slightly understated because attribution was ~95% mature at pull time.
• Sponsored Display coverage is the thinnest of the ad channels, and its CPC/CPM are omitted entirely.
• Small category cells. Categories are reported at a six-advertiser floor, which is thin at the extremes.
• This is one cohort of Amazon advertisers, not a census of the marketplace. It skews toward brands that use optimization software.
• We don't publish exact cohort or per-channel counts. Coverage is given as a share of the cohort instead. Medians and percentile spreads across several hundred advertisers are stable enough that the exact denominator doesn't change how you should read them.

Frequently asked questions

How much did Amazon ad spend increase on Prime Day 2026?

Across several hundred US advertisers measured by Marketplace Ad Pros, the median brand raised Sponsored Products daily spend 27% during Prime Day 2026 (June 23-26) versus a matched baseline, and 75% of brands spent more. The spread was wide: the 25th percentile brand spent the same as baseline (1.00x) while the 75th percentile spent 87% more (1.87x).

Did Amazon CPCs go up on Prime Day 2026?

Yes, but modestly. Median Sponsored Products CPC rose about 10%, from $1.19 to $1.31, across several hundred US advertisers. Sponsored Brands CPC barely moved, from $1.18 to $1.20 (+2%). Median ACOS was roughly flat to slightly up, moving from 33% to 35% on Sponsored Products.

Should I raise my Amazon ad budget for Prime Day?

The 2026 data does not support it as a lever on its own. Only 13.7% of brands raised daily budget by 20% or more, and 51% actually cut budget, yet 75% still spent more because Prime Day demand filled the budgets they already had. Brands that raised budget by 20%+ saw the same median ad-sales lift (1.30x) as brands that held budget steady (1.30x). This is correlational, not causal, but it suggests the lift came from demand rather than from opening budgets.

How much did Prime Day 2026 lift total sales, not just ad sales?

For the roughly 43% of the cohort with Seller Central data connected, median total sales rose 37% during Prime Day 2026, and 76% of those brands grew total sales. That is higher than the median Sponsored Products ad-sales lift of 30%, meaning organic sales rose faster than ad-attributed sales for the typical brand.

What happened to Sponsored Display on Prime Day 2026?

Sponsored Display was the exception. While Sponsored Products and Sponsored Brands both saw median spend rise 27%, the median advertiser pulled back on Sponsored Display: spend fell 17% and impressions fell 37%. Efficiency improved as they did so, with median SD ACOS dropping from 34% to 25%. Brands concentrated Prime Day dollars in Sponsored Products and Brands.

Which categories leaned into Prime Day 2026 the hardest?

By median Sponsored Products spend index, Electronics led at 1.78x, followed by Home & Kitchen at 1.53x, Baby Products at 1.51x, and Automotive at 1.50x. Office Products barely moved at 1.03x, and Toys & Games was also soft at 1.12x. Only categories with at least six advertisers in the cohort are reported.

Related guides

• Best Amazon ads management tools in 2026
• Best Amazon ads reporting & analytics tools in 2026
• Amazon advertising automation tools, compared

Want these cuts for your own account instead of the cohort median? Connect your Amazon Ads and Seller Central data and ask Claude or ChatGPT directly -- free integration.