Reading BFCM MER Trends 2022-2025: A Methodology for DTC Operators

Black Friday / Cyber Monday is the highest-spend, highest-volatility window of the DTC year, which makes it both the best and worst time to read your blended Marketing Efficiency Ratio (MER). Best, because the absolute revenue numbers are large enough to be statistically meaningful. Worst, because four years of post-2021 structural shifts — iOS 14.5 ATT, server-side CAPI adoption, Apple Mail Privacy Protection, the retail-media surge — have made year-over-year BFCM MER comparisons treacherous if you don't normalize.

This post is a methodology piece. It walks through the major structural shifts in DTC measurement from 2022 through 2025, why year-over-year MER comparisons mislead without adjustment, and how to read your own historical BFCM data correctly. We deliberately do not publish cross-brand benchmarks — every account's mix is different enough that aggregate numbers obscure more than they reveal.

The technical reality — what changed across 4 BFCMs

BFCM 2022 — first full BFCM under iOS 14.5 ATT

Apple shipped App Tracking Transparency (ATT) in iOS 14.5 on April 26, 2021. BFCM 2021 was already affected, but BFCM 2022 was the first full BFCM where every brand's iOS-targeted Meta and TikTok campaigns ran under the post-ATT signal regime. Per Apple's own privacy materials, ~75-80% of iOS users decline the ATT prompt when shown.

What changed in BFCM 2022 vs prior:

• iOS view-through windows on Meta capped at 1 day (down from 28-day click + 1-day view pre-ATT).
• Meta's Aggregated Event Measurement (AEM) protocol replaced full event-level reporting for iOS-opted-out users — only the top 8 events per domain reported back.
• Lookalike audience seeds for iOS were materially smaller. Brands that hadn't yet adopted server-side CAPI saw match rate drop into the 45-60% band.

BFCM 2023 — first full BFCM with widespread CAPI adoption

By BFCM 2023, server-side CAPI (Meta's Conversions API, Google's Enhanced Conversions, TikTok's Events API) was mainstream rather than cutting-edge. Shopify's official Conversions API for Meta app launched in mid-2022 and was used by a meaningful share of merchants by Q4 2023. Brands that wired CAPI correctly recovered match rates into the 75-90% band — see our server-side vs browser pixel post for the mechanics.

What changed in BFCM 2023 vs BFCM 2022:

• Meta-attributed revenue recovered for CAPI-enabled brands, often by 20-40% of the prior year's "lost" volume.
• Meta's optimizer had richer signal — CPAs softened on prospecting for accounts where CAPI was correctly deduplicated via event_id.
• Brands that didn't wire CAPI fell further behind: the gap between properly-instrumented and under-instrumented accounts widened.

BFCM 2024 — first full BFCM after Mail Privacy Protection's full effect on email

Apple's Mail Privacy Protection (MPP) shipped in iOS 15 (September 2021), but BFCM 2024 was the first where the operational consequences — open-rate inflation, broken smart-send, broken re-send segments — were fully baked into most brands' Klaviyo workflows. See our iOS 15 MPP post for the full breakdown.

What changed in BFCM 2024 vs BFCM 2023:

• Brands that hadn't migrated A/B testing from open rate to click rate kept scoring "winners" that were noise. Aggregate email-channel performance looked similar year-over-year on the dashboards, but the underlying decisions were lower quality.
• Send-time-optimization tools that hadn't been retrained for the post-MPP world degraded campaign performance for Apple-Mail-heavy lists.
• The honest engagement metric — click rate, not open rate — became the only reliable cross-year comparison.

BFCM 2025 — first full BFCM under the retail-media surge

The shift here is less platform-driven and more competitive-environment-driven. Amazon Ads, Walmart Connect, Target's Roundel, and Kroger Precision Marketing all materially scaled their off-platform retail-media products through 2024-2025. Combined with TikTok Shop's continued US growth and Meta's Advantage+ Shopping Campaigns hitting maturity, the competitive ad-auction landscape in Q4 2025 was meaningfully different from Q4 2022.

What changed in BFCM 2025 vs BFCM 2024:

• CPMs in the Q4 peak weeks ran higher across most DTC verticals as retail-media spend pulled performance budget into auction overlap with social-platform spend.
• Brands that had diversified into Amazon Ads / TikTok Shop / Pinterest saw their channel mix shift — the share of BFCM revenue attributed to Meta dropped vs prior years not because Meta got worse but because the alternatives got bigger.
• Blended MER calculations that include retail-media spend on the cost side and retail-media revenue on the revenue side became more important.

The cumulative consequence: BFCM 2022 vs BFCM 2025 is not a clean year-over-year comparison. The same brand running the same products at the same blended MER number across the four years is, mechanically, running a fundamentally different program at the underlying measurement layer.

Why it matters for DTC attribution

The temptation every BFCM is to compare this year's MER to last year's and declare "we improved" or "we regressed." Without normalization, both conclusions can be wrong:

• An apparent MER improvement from 2.4 (2022) to 2.9 (2023) might be entirely CAPI-driven — the same conversions happened, but more of them were tracked, so Meta-attributed revenue rose without any underlying business change.
• An apparent MER regression from 3.1 (2024) to 2.7 (2025) might be entirely retail-media-driven — your competitors moved more spend into ad auctions you compete in, pushing your CPMs up, without any change in your own creative or product offering.

A budget decision based on a misread year-over-year MER trend is worse than no comparison at all.

Methodology — how to read your own BFCM data correctly

Step 1 — Pull blended MER for each BFCM year on consistent windows. BFCM technically runs the week of Thanksgiving (US) through the Monday after. We recommend pulling the 7-day window from Wednesday before Thanksgiving through the following Tuesday as the canonical "BFCM week." Pull the same 7 days each year, with Shopify's order_total as the revenue numerator and total paid-media spend (Meta + Google + TikTok + Pinterest + Amazon Ads + any other paid) as the denominator. Blended MER = Shopify revenue / total paid spend.

Step 2 — Annotate each year with its measurement context. Build a small table:

Year | Blended MER | CAPI status        | Email MPP-adjusted | Retail-media share of spend
2022 | (yours)     | not deployed       | no                 | <5%
2023 | (yours)     | partial            | no                 | 5-10%
2024 | (yours)     | full               | yes                | 10-20%
2025 | (yours)     | full + first-party | yes                | 20%+

If your CAPI status, email-attribution methodology, or retail-media share changed materially between two years, the year-over-year MER is not directly comparable — it's a structural comparison plus a performance comparison stacked.

Step 3 — Separate the structural from the performance shift. A useful exercise: for each year, estimate what that year's MER would have been if you'd had the prior year's measurement stack. For 2023 vs 2022, that means: what would your 2023 MER have looked like with no CAPI? You can approximate this by pulling 2023 Shopify revenue and dividing by 2023 paid spend, but using only Meta-attributed-revenue-equivalent-to-2022's match rate as the numerator. The exact math depends on your accounting, but the principle is: try to isolate the "we did better at marketing" component from the "we instrumented better" component.

Step 4 — Compare BFCM week to the rest of Q4 within the same year. This is more honest than year-over-year. BFCM week MER / non-BFCM Q4 MER shows whether BFCM was a productive use of the demand spike. The ratio is comparable year-over-year because the measurement stack within a year is consistent.

Step 5 — Look at Shopify revenue per Shopify customer (AOV) and Shopify orders separately. Attribution noise doesn't affect Shopify's order count or AOV. If your Shopify orders rose 30% year-over-year and your AOV held flat, you genuinely sold more — the only question is whether marketing or organic was the driver. If your Shopify orders held flat and your blended MER rose 10%, you didn't sell more; you measured better.

Step 6 — Avoid cross-brand BFCM benchmarks. They're tempting and they're almost always misleading. A "DTC BFCM benchmark MER of 3.2" mashes together apparel, supplements, furniture, and food in proportions you don't share. Your honest benchmark is your own prior years adjusted for the structural shifts above.

Illustrative scenario — reading a 4-year MER trend honestly

Imagine a DTC accessories brand looking at their BFCM week numbers:

Year | Spend | Revenue | Blended MER | CAPI? | Retail-media spend share
2022 | $200K | $480K   | 2.40         | no    | 3%
2023 | $235K | $640K   | 2.72         | yes   | 7%
2024 | $260K | $755K   | 2.90         | yes   | 15%
2025 | $310K | $810K   | 2.61         | yes   | 22%

The naive read: 2022 → 2024 improved steadily, 2025 regressed.

The structural read:

• 2022 → 2023 +0.32 MER jump: most of this is the CAPI instrumentation step. Meta-attributed revenue mechanically rose because match rate jumped from ~60% to ~85%. Underlying business probably improved by a smaller amount.
• 2023 → 2024 +0.18 MER jump: smaller and likely more real. CAPI was steady-state at this point. The improvement is more likely genuine creative + offer + LTV gains.
• 2024 → 2025 -0.29 MER drop: the retail-media spend share more than doubled (15% → 22%), which means more of total paid spend went to channels with thinner historical optimization. Year-over-year auction competition was also higher. A genuine regression on the within-channel performance may be small or zero; the headline regression is dominated by mix shift.

A more honest summary: "we genuinely improved BFCM performance 2022 → 2024, with most of the apparent 2022 → 2023 jump being instrumentation rather than performance, and the 2025 dip is mostly mix-shift and competitive auction effects rather than a real regression. Within-Meta MER actually held flat across 2024-2025."

This kind of read informs a fundamentally different budget conversation than "MER dropped, we need to cut spend." The numbers above are illustrative; the methodology of separating instrumentation, mix, and competitive effects from underlying performance is the transferable point.

What we do at Admaxxer

Admaxxer's attribution models documentation walks through the math of blended MER vs platform-attributed ROAS, and the tracking health view shows your CAPI match rate so you know which years are comparable on instrumentation. The marketing acquisition surface ranks channels by blended MER contribution with retail-media spend included in the denominator — the right denominator for a brand that's actually running retail media. For brands looking to run year-over-year BFCM analysis, we expose the underlying Shopify orders / AOV / spend / match-rate signals separately so you can construct the honest comparison yourself.

FAQ

What's a "good" BFCM blended MER?

There isn't one universally. The honest answer depends on your AOV, repeat-purchase economics, gross margin, and category. Supplements brands with high LTV can profitably run BFCM at 2.0-2.5x. Furniture brands with low repeat need 3.5-5x+ to pencil out. Don't pick a benchmark from a generic DTC source — back out the MER target from your unit economics.

Should I include free shipping promo costs in my BFCM spend denominator?

Yes, if the free shipping was a BFCM-specific incremental cost (you wouldn't have offered it otherwise). If free shipping is your year-round default, treat it as cost-of-goods, not marketing. The principle is: every BFCM-specific incremental cost should be in the spend denominator.

What about discounts — do they affect MER?

Discounts reduce Shopify revenue (the numerator). A 25% sitewide BFCM discount means the same gross orders count as 75% of revenue in the MER numerator, so MER falls mechanically. Some brands track "gross MER" (pre-discount revenue / spend) and "net MER" (post-discount revenue / spend) separately. Net MER is the truth; gross MER is comparable year-over-year if your discount depth is consistent.

Is BFCM still worth running given rising CPMs?

For most DTC brands, yes. BFCM's organic demand surge dwarfs the auction-cost increase — the opportunity cost of not running is high. But the math of how much you should scale BFCM spend vs your non-BFCM baseline is where most brands over-extend. A simple sanity check: BFCM MER / non-BFCM Q4 MER should be at least 1.2 to justify the additional spend. If your BFCM MER is worse than your Q4 baseline, you're overspending.

How early should I look at BFCM data?

Don't make scale-up or scale-down decisions in the first 36 hours. Conversion lag (especially under 7-day click windows) means BFCM week's MER doesn't fully settle until 5-10 days after Cyber Monday. Pull the BFCM week MER on the second Tuesday of December for the cleanest read.

Does cyber Monday + Black Friday split show anything useful?

Yes — the split between BF and CM tends to reveal channel mix changes. Black Friday usually skews more discovery-channel (Meta prospecting, TikTok); Cyber Monday usually skews more retargeting-channel (Google, email, retargeted Meta). If your CM revenue weight rose YoY, you're more dependent on retargeted demand — generally a sign of healthy creative on the discovery side over the prior weeks.

What's the right way to handle gift purchases in BFCM data?

Gift purchases inflate AOV (multiple recipients per order), distort LTV cohorts (the buyer ≠ the eventual repeat customer), and skew geographic data (ship-to is often different from buyer's home zip). Most brands don't bother adjusting for it in BFCM dashboards, but if you're trying to read LTV signal off BFCM cohorts, the gift-purchase confound is the biggest hidden variable. Worth filtering separately.