BFCM Attribution: Why Your Ad Platforms Will Claim More Revenue Than You Think
· Tutorials/Guides

Every Black Friday Cyber Monday (BFCM), your ad platforms will report a record week. Add Meta's attributed revenue to Google's attributed revenue, then compare the sum to what Shopify booked, and the platforms will have claimed credit for more sales than your store processed.
That gap is the attribution problem at full volume. The 5 days with the most purchase intent of the year are also the 5 days when every platform has the most purchases available to claim, so each one reports its best numbers at the exact moment its numbers are least trustworthy.
The common reaction is to find the platform with the highest ROAS and push budget toward it. We think that reflex costs brands more in Q4 than any bidding or creative mistake. Platform ROAS during BFCM tells you how many buyers a channel touched before they purchased, which is a different figure from how many buyers your spend created.
We run Google, YouTube, and paid social for 8-figure ecommerce brands, with measurement as the third pillar of the stack. This guide covers why platforms over-claim during BFCM, how to build a source of truth before the promo goes live, which settings to check inside each platform, and how to read the results in December without fooling yourself.
Why Every Platform Over-Claims During BFCM
Platforms grade their own homework, and BFCM is the week when the grading gets most generous. Each pixel sees only its own touchpoints, so each platform assigns full or partial credit for the same order with no deduplication against the others.
Spend spikes, so touchpoints spike
Across 12 client accounts in Data Druid, our account intelligence platform, the median account's peak spend day in the 2025 season ran at 3.78x a normal August day. The worst 30 consecutive days ran at a median of 1.79x. More spend buys more impressions, which means a larger share of your buyers have seen or clicked an ad before checkout, whether that ad changed their decision or not.
The result is predictable. Attributed conversions rise in step with spend because the platforms are counting exposure to the buyer, which climbs whenever budget climbs.
View-through windows absorb buyers who were already coming
Meta's default setting of 7-day click and 1-day view credits any purchase made within 24 hours of seeing an ad. During BFCM, your email list, your SMS subscribers, and every shopper who saved your product in October will see an ad on their way to buy. Meta counts all of them.
Those buyers would have purchased from the email alone. A view-through credit for them inflates reported ROAS without adding a single incremental dollar.
Long click windows reach back into October
Google Ads conversion actions default to a 30-day click window. A shopper who clicked a Shopping ad on October 31 to research and then bought on Black Friday gets credited to that click, even if a Meta retargeting ad, a promo email, and a branded search all happened in between. Meta, Google, and your email platform can each record that single order as their own.
Branded search and retargeting harvest existing demand
Branded search volume rises sharply in BFCM week because your promo is driving people to look for you by name. Performance Max will happily serve on those branded queries unless you apply a brand exclusion list, then report the revenue as its own. Retargeting does the same on Meta by concentrating spend on people who already sit in your cart and site visitor audiences.
Both tactics post the highest ROAS in the account during promo week. They also have the weakest claim to having caused the sale.
Do not add platform revenue together
Summing Meta, Google, and YouTube attributed revenue produces a number that describes no real outcome. When that sum exceeds store revenue, the excess is double counting. A sum below store revenue still tells you nothing about which channel drove which order.
Build Your Source of Truth Before the Promo Goes Live
You cannot fix attribution in the middle of BFCM week. The numbers you will trust in December have to be defined, baselined, and instrumented by early November, while there is still time to collect a clean comparison period.
Make MER the scoreboard
Marketing efficiency ratio (MER) is total store revenue divided by total ad spend. It ignores channel credit entirely, which makes it the one efficiency number no platform can inflate. We track it daily through BFCM next to new customer MER, which uses first-order revenue only, because a promo that sells to existing customers can post a strong blended MER while acquiring almost no one.
Platform ROAS stays on the dashboard as a checkpoint. Budget decisions during promo week follow MER and new customer CAC.
Recalculate breakeven at promo pricing
A discount shrinks contribution margin, which raises the ROAS you need to break even. As an illustration, a product with 60% contribution margin at full price has a breakeven ROAS of 1.67. Take 25% off that same product and margin falls to roughly 47%, which pushes breakeven to 2.14.
Most teams carry their October ROAS targets into November unchanged. Reset tROAS and your internal guardrails to the promo margin, or the account will scale spend that loses money on every order while the dashboard shows green.
Set up the measurement stack in order
Pull 4 to 6 weeks of pre-promo baselines for MER, new customer CAC, new customer share of orders, and AOV.
Launch a post-purchase survey through Fairing or Triple Whale that asks customers where they first heard about you, with channel options that match your media mix.
Design a geo holdout for your largest prospecting channel, using matched regions with no spend, and get the first read before Black Friday so you have an incrementality factor to apply during the promo.
Confirm your first-party pixel or Triple Whale setup deduplicates orders across channels, so one view of the funnel counts each order once.
Write down the decision rule before launch, naming which number wins when the platforms disagree with MER.
That last step matters more than any tool. A team that decides in advance to follow MER and holdout data will not get talked into moving budget by a 9x retargeting ROAS on Cyber Monday.
What to Check Inside Each Platform
Each platform inflates BFCM credit through a different mechanism, so each one needs a different correction. The goal is to read every channel on the terms closest to incrementality that its own reporting allows.
Meta: separate click credit from view credit
Use the Compare Attribution Settings columns in Ads Manager to show 7-day click and 1-day view side by side for every campaign. If the 1-day view share of purchases climbs sharply in promo week while click-attributed purchases stay flat, Meta is collecting credit from buyers your email and SMS already sent to checkout.
Make promo-week budget calls on 7-day click. Cap retargeting and existing customer spend with Advantage+ audience controls or an existing customer budget cap, because those segments carry the most inflated ROAS during BFCM.
Google: wall off your brand
Apply a brand exclusion list to every Performance Max campaign and run branded search in its own campaign. Branded search ROAS during BFCM measures how many people your promo sent to Google looking for you, which belongs to the promo, the email, and the prospecting that built awareness.
Open the PMax channel performance report weekly through November. If the share of spend and conversions sitting in Search climbs during promo week, PMax is leaning on branded and high-intent queries that would have converted through your search campaigns at a lower cost.
YouTube: judge it on lift
YouTube engaged-view conversions count a purchase after someone watches 10 seconds of your ad, then buys within the conversion window. In a week when buyers are already primed, that rule credits YouTube for a lot of shopping it did not cause.
We judge YouTube during BFCM on what it does to branded search volume, new customer MER, and geo holdout lift. A YouTube campaign that drove awareness in October will show up as branded search demand in November, which is where its real contribution lives.
Platform | Inflation source | Setting to check | Number to trust in promo week |
Meta | 1-day view credit on buyers already headed to checkout | Compare 7dc 1dv against 7dc only. Cap retargeting and existing customer spend | 7-day click purchases and new customer CAC |
Google Search and PMax | Branded queries and long click windows | Brand exclusion list on PMax. Separate branded campaign. Channel performance report | Non-brand ROAS at promo-adjusted tROAS |
YouTube | Engaged-view conversions on primed buyers | Engaged-view share of conversions. Branded search trend | Geo holdout lift and branded search volume |
Use CPMr to Catch Over-Claiming Before ROAS Shows It
CPMr, or Cost Per Mille Reached, is the cost to reach 1,000 unique people with your ads. Phil Kiel, our Director of Paid Social, popularized it internally as a leading indicator for paid social, and BFCM is where it earns its keep. You calculate it as CPM x frequency, and Meta reports it in Ads Manager as Cost per 1,000 Accounts Centre accounts reached.
CPM measures the cost of 1,000 impressions. If one person sees your ad 5 times, that is 5 impressions and 1 person reached, so CPM stays flat while CPMr climbs.
Why it matters for attribution
When CPMr rises week over week, your budget is cycling through the same people instead of finding new ones. Those repeat viewers are often your hottest prospects, which means they convert at a high rate and collect view-through credit on every purchase. Reported ROAS holds steady or even improves while the pool of new buyers your spend can reach shrinks.
This shows up in CPMr weeks before CTR drops and well before ROAS moves. By the time platform metrics flag a problem, the account has usually spent most of the promo window paying to re-reach its own customers.
Diagnose the cause before you act
A rising CPMr has 2 possible causes with 2 different fixes. Frequency rising while CPM holds flat means audience saturation, which you fix with exclusions and campaign consolidation. CPM rising while frequency holds flat means auction pressure, which is expected in BFCM week as every advertiser bids for the same inventory.
How we track it through BFCM
Benchmark CPMr against your own account average from September and early October.
Track it weekly through the ramp, then daily from the Monday before Black Friday.
Split every CPMr move into its CPM and frequency components before changing budget or targeting.
Treat a frequency-driven climb in prospecting as a signal that attributed revenue from that campaign is drifting toward existing demand.
Read the Results in December Without Fooling Yourself
The BFCM read that matters happens after the attribution windows close and the returns come back. A read taken on the Tuesday after Cyber Monday captures the platforms at their most generous and your margin at its most optimistic.
Wait for the windows to close
Google's 30-day click window keeps adding conversions to November clicks through most of December. Pull the final platform numbers once the windows have closed, then net out promo returns against store revenue before you calculate MER for the period.
Convert platform ROAS into incremental ROAS
Your geo holdout produces an incrementality factor, the share of platform-attributed conversions that the test shows your spend created. Apply it to platform ROAS for every channel you tested.
\text{Incremental ROAS} = \text{Platform ROAS} \times \text{Incrementality factor}
As an illustration, a prospecting campaign that reported a 4.0 ROAS with an incrementality factor of 0.5 delivered an incremental ROAS of 2.0. Against a promo breakeven of 2.14, that campaign lost money on every order even though the dashboard called it a winner.
Judge the BFCM cohort on payback
Customers acquired at a discount behave differently from full-price customers. Track the BFCM cohort's repeat purchase rate and contribution margin at 60 and 90 days, then compare its LTV/CAC and payback period against your Q3 cohorts before you decide how hard to push next year's promo.
What not to do
Do not add Meta, Google, and YouTube revenue together and report the total to your board.
Do not move December budget toward retargeting and branded search because they posted the highest ROAS in promo week.
Do not cut YouTube or upper-funnel Demand Gen based on platform ROAS alone. Check branded search volume and holdout lift first.
Do not set January targets from BFCM platform ROAS. Reset to full-price margin and your pre-promo baselines.
Your ad platforms will claim more BFCM revenue than your store booked, because each one counts every buyer it touched during the busiest buying week of the year. View-through windows, 30-day click windows, branded search, and retargeting all pull credit from demand your promo and your email list created.
The brands that come out of Q4 with an accurate read do the work in early November. They make MER the scoreboard, reset breakeven to promo margin, launch a holdout and a post-purchase survey, then write down which number wins before the first discount code goes live. In December, they wait for the windows to close and convert platform ROAS into incremental ROAS before a single 2027 budget gets set.
If you do one thing this week, pull your pre-promo MER baseline and recalculate breakeven ROAS at your Black Friday discount.