I have spent two decades inside DTC and CPG operating teams — scaling brands through Shopify Plus, retail expansion, and the inventory-led growth pressure tests that come with crossing the $20M, $50M, and $100M revenue lines. I use the public DTC and CPG cohort as the benchmark set for the private brands we advise because the FY25 / FY26 10-K disclosures are now usable for cross-comparison and because the cohort has cleanly absorbed the same post-iOS, post-CPM-inflation, post-sentiment-crash environment the private mid-market is underwriting against. The Q2 2026 read on the cohort tells a clear, structural story: customer acquisition cost is up roughly 8-12% across the 24-month window, gross margin has held within a couple of points of trend, and contribution has compressed where the brand has not actively rebuilt the channel mix or the cohort discipline. The brands clearing the new bar are the ones with a coherent omnichannel composition and reported cohort retention. Here is the working read against five named public benchmarks.
01 The cohort frame and what the filings actually disclose
The public DTC and CPG cohort I track for benchmarking purposes is built from SEC EDGAR 10-K and 20-F filings, with five anchor names that span the operating shapes that show up in private mid-market processes: e.l.f. Beauty (ELF) for mass-retail-anchored beauty, Olaplex (OLPX) for premium beauty in compression, YETI Holdings (YETI) for mature DTC and omnichannel hardgoods, Warby Parker (WRBY) for retail-led DTC, and Hims & Hers Health (HIMS) for subscription DTC. The cohort intentionally spans the gross-margin range that mid-market sellers will recognise — from low-50s to high-70s — and the channel-mix range from near-pure DTC through wholesale-anchored CPG.
The disclosure quality is imperfect but usable. Marketing as a percent of revenue is reported as a separate line for some names (WRBY at 15.8% of FY24 revenue, HIMS at 39.6%, ELF at 26.4% within the SG&A footnote) and embedded inside SG&A for others (YETI, OLPX). Customer-level CAC and LTV:CAC are disclosed by only two names in the cohort — HIMS reports payback under 12 months and LTV:CAC above 4x on its 2024 Investor Day deck; FIGS reports cohort payback under 6 months and LTV:CAC above 3x in its 2024 investor presentation. The rest of the cohort gives you marketing-as-percent-of-revenue and customer-count trajectories from which CAC can be backed out, which is what we do on the underwriting side.
The methodology I run for the benchmark is straightforward: pull the trailing twelve months of revenue, COGS, marketing spend (or marketing-equivalent SG&A line), and operating cash flow from the latest 10-K; compute gross margin, marketing as percent of revenue, the contribution-margin proxy (gross margin minus marketing percent), and operating-cash-flow margin; tag the channel mix (DTC, retail/wholesale, professional, subscription) from the disclosed segment data. The result is a five-name cohort comparable that the private mid-market brand can benchmark each line against, with a clear awareness of which numbers are apples-to-apples and which require an adjustment for disclosure mechanics.
The public cohort is not the standard private brands should clear — it is the standard private buyers underwrite against. Knowing the gap is the difference between leaving a turn of multiple on the table and not.
02 CAC, CPM inflation, and the post-ATT context
You cannot read 2025-2026 DTC unit economics without naming the CPM and signal-loss environment that defined them. The McCarthy et al. SSRN 2024 study on the ATT impact remains the cleanest published causal estimate — Meta conversion-optimised click-through rate compressed 37% post-ATT, with Meta-exposed e-commerce firms seeing revenue impacts ranging from 8% to 40% depending on Meta-spend dependence, and small Meta-heavy DTC brands hit with revenue declines around 60% relative to less-exposed peers. The Meta ad-spend pool shrank 6.8% overall as budgets reallocated to Google search, where conversion intent is visible and ATT is less binding.
The platform CPM trajectory tells the second half of the story. Tinuiti's Q1 2026 Digital Ads Benchmark Report has Meta CPM down 3% YoY in Q1 2026 — the second consecutive quarter of CPM decline — even as Meta impressions grew 17% and spend grew 13%. Get-Ryze 2026 benchmarks have median Meta CPM at $14.19, up 20% YoY from $11.82, and median cross-industry CPA at $38.19, up 8.5%. Reddit CPM ran +71% YoY in Q1 2026, with CPM growth over 30% YoY for three consecutive quarters as DTC budgets diversified into the smaller walled gardens. AppLovin AXON 2.0 is delivering Net Revenue Per Installation roughly 75% above the prior year and Triple Whale-tracked DTC operators now treat AppLovin as a top-three paid channel after Meta and Google.
For the public cohort specifically, the practical effect is that marketing-as-percent-of-revenue has either crept up modestly (ELF moved from low-20s to mid-20s over the FY22-FY26 period; HIMS holds around 38-40%) or held flat while underlying CPMs inflated and signal quality deteriorated (WRBY at 15-16%, FIGS at 20%, BIRD at 21-22%). The brands holding marketing percent flat in the inflating-CPM environment are doing it through some combination of channel diversification away from Meta, retention-revenue contribution, and the AI-driven Meta efficiency recovery that Tinuiti has documented. Private mid-market brands without those levers have absorbed CPM inflation as a contribution-margin hit.
03 Gross margin held — but the dispersion widened
The cohort gross margin range in FY25 / FY26 disclosures runs from 54.0% (WRBY) to 73.8% (HIMS), with ELF at 70.7%, OLPX at 69.4%, and YETI at 57.4%. The headline read is that gross margin held — most names sit within two points of their trailing five-year average — but the dispersion underneath that headline widened materially over 2023-2026. Three structural drivers explain the dispersion and they are the same drivers that show up underneath private-brand gross-margin compression.
Category and product-architecture floor
Subscription telehealth and premium professional beauty sit at the top of the band structurally — HIMS at 73.8%, OLPX at 69.4%, ELF at 70.7% — because the COGS-to-price ratio of those product categories is mechanically attractive at scale. Premium drinkware-and-coolers (YETI) and optical retail (WRBY) sit in the mid-50s because the product COGS is materially higher per unit. The category and product-architecture floor is the single most consequential variable for the gross-margin ceiling, and it does not move with operating improvements. Private mid-market sellers underwriting against the wrong sub-cohort consistently overestimate their achievable gross-margin band.
Channel mix and trade-spend mechanics
YETI's FY25 gross margin of 57.4% is up roughly 9.5 percentage points from the FY22 trough of 47.9%, and the CFO has explicitly attributed a meaningful share of the recovery to the shift in DTC share from the high-40s to roughly 59% of revenue. ELF's 70.7% gross margin on a wholesale-anchored model (DTC is a low-teens-percent insight layer, with 85-90% of volume flowing through Walmart, Target, Ulta, CVS, and Boots) is structurally higher than what a pure-DTC beauty operator can sustain, because trade spend is more predictable and lower-volatility than the equivalent paid social CAC. The mechanics matter for private brand benchmarking: a beauty brand at 65% gross margin pure-DTC is operating at the same level as ELF clearing 71% mostly-wholesale; the underwriter knows the difference, and the multiple reflects it.
Input-cost discipline and SKU-level profitability work
OLPX held gross margin at 69-70% from FY21 through FY25 despite a $281M revenue decline from the FY22 peak of $704M to the FY25 print of $423M. The product COGS architecture absorbed the volume deleveraging without showing up in gross margin — the entire compression flowed through operating expense lines, not COGS. That is what disciplined input-cost work and SKU-level profitability tracking look like on a public filing: gross margin holds in a falling-revenue environment because the COGS line is actively managed. The brands in the private mid-market that have done the SKU profitability and landed-cost work tend to clear at gross-margin levels approaching their public-cohort sub-band; those that have not show 200-400 basis points of compression in trailing twelve.
04 Contribution margin — where the cohort actually separates
The contribution-margin proxy I use across the public cohort is gross margin minus marketing as percent of revenue, then minus an estimated variable-fulfillment line (typically 2-4 points for the cohort) where the filings let me see it. The result is a working contribution-margin band that maps cleanly to the ATTN Agency 2026 DTC Exit Valuation guide framework — below 30% is a red flag, 35-50% is healthy, and 55%+ is premium territory. Across the cohort the FY25 / FY26 contribution-margin proxies are: HIMS approximately 34% (73.8% GM less ~40% marketing), ELF approximately 43% (70.7% less ~28%), WRBY approximately 40% (54.0% less ~14%), YETI approximately 47% (57.4% less ~10% on the SG&A-embedded marketing estimate), and OLPX approximately 59% (69.4% less ~10% marketing on the SG&A breakdown).
Two things are worth noting about that contribution-margin distribution. First, every name in the cohort clears the 35% "healthy" floor at the contribution-margin proxy level — including the brands with the highest marketing spend (HIMS) and the lowest gross margin (WRBY). The "DTC is broken" narrative that drove the 2022-2023 multiple compression in the cohort was always more about operating margin and discounted-cash-flow durability than about contribution margin. Second, the contribution-margin dispersion (34% to 59%) is approximately twice as wide as the gross-margin dispersion (54% to 74%) and roughly four times as wide as the operating-cash-flow-margin dispersion (the band is 12.7% to 13.9% across all five names). Contribution-margin is where the cohort actually separates; OCF margin is where it converges.
For private mid-market brands, the implication is operational: if the only number you can show the buyer is gross margin and "marketing efficiency," you are reporting on the wrong layer of the unit-economics stack. The buyer will compute contribution against the public-cohort comparables and underwrite from there. Reporting a clean, sub-cohort-relevant contribution-margin number — gross margin minus paid-channel marketing minus shipping and processing — and showing it stable or expanding over trailing twelve gets you to the table at the right multiple.
05 Compression versus recovery — the cohort tale of two trajectories
The single most useful pair of public-cohort case studies for a private mid-market seller right now is the OLPX vs WRBY comparison. Both names sit in the same operating-margin neighbourhood at FY25 reporting — OLPX at +1.6% operating margin, WRBY at -0.6%. Both are below the "premium" public-cohort benchmark of 10-15% operating margin. But the trajectories are opposite, and the trajectory is what underwriters pay for.
Olaplex — what compression looks like on the page
OLPX FY22: revenue $704.3M, gross margin 73.8%, operating income $364.4M, operating margin 51.7%, operating cash flow $255.3M. OLPX FY25: revenue $423.0M, gross margin 69.4%, operating income $7.0M, operating margin 1.6%, operating cash flow $58.7M. Gross margin compressed 4.4 percentage points; operating margin compressed 50.1 percentage points. Revenue dropped 40% over three years. The brand still holds $650.5M of LBO-era long-term debt against an FY25 OCF of $58.7M — a debt-to-OCF ratio that frames the equity story entirely. This is what happens when a category peak (the 2020-2022 Olaplex moment) reverses, brand health softens, channel conflict between professional and specialty retail erodes pricing power, and the company has no operating-margin slack to absorb the deleveraging.
Warby Parker — what recovery looks like on the page
WRBY FY21: revenue $540.8M, gross margin 58.7%, operating loss -$143.7M (-26.6% margin), operating cash flow -$32.0M. WRBY FY25: revenue $871.9M, gross margin 54.0%, operating loss -$5.3M (-0.6% margin), operating cash flow +$110.8M (12.7% of revenue). Gross margin compressed 4.7 percentage points over the four-year window — comparable to OLPX — but operating margin improved 26 percentage points and operating cash flow flipped from -$32M to +$111M. Revenue grew 61% over the same period. Capex of $67M (7.7% of revenue) continued to fund retail-store expansion. The trajectory is what the public market pays the revenue-multiple premium for: the contribution-margin engine has reached the inflection where each incremental dollar of revenue clears to cash flow, and the operating-margin recovery is in front of you, not behind.
Olaplex shows you what unit-economics compression looks like at scale. Warby Parker shows you what recovery looks like. Same gross-margin trajectory, opposite operating-margin trajectory — and the public market correctly prices the gap.
For private mid-market sellers, the OLPX/WRBY pair is the cleanest available frame for the timing question. If the trajectory looks more like OLPX — gross margin holding but operating margin and cash flow compressing as a category peak normalises — the right answer is almost never to open a process into the compression; the right answer is to rebuild the contribution-margin layer first, then time the marketing into the recovery. If the trajectory looks more like WRBY — operating cash flow expanding, operating margin recovering toward break-even and above, with a credible path to mid-teens — opening into the next 18 months on a clean sell-side package is the strongest available trade.
06 The channel-diversification premium and what it actually requires
The "omnichannel premium" is real in the public cohort but it is conditional, and the conditions are operational. The two cleanest examples in the cohort — YETI and ELF — clear the premium because the channels are coherent (each does a defined job in the model) and because the contribution-margin layer holds up across the mix shift. YETI moved from roughly 56% DTC / 44% wholesale in FY21 to roughly 59% / 41% in FY24, with the gross-margin recovery (47.9% in FY22 to 57.4% in FY25) directly attributable to the DTC mix shift. ELF runs roughly 85-90% wholesale through Walmart, Target, Ulta, CVS, and Boots, with DTC as a low-teens-percent insight engine — and the wholesale-anchored 70.7% gross margin clears the contribution-margin floor that pure-DTC beauty operators struggle to match.
WRBY is the third coherent omnichannel case in the cohort — roughly 63% own-retail and 37% e-commerce, with no meaningful wholesale — but the gross-margin floor is lower (mid-50s versus 70%+ for ELF and OLPX) because the unit-COGS architecture of optical retail is structurally heavier than packaged beauty. The market correctly prices WRBY closer to a specialty retailer than a high-multiple DTC platform. HIMS is the counter-case: 90%+ pure DTC subscription, with retail partnerships (Target, Walmart) functioning as awareness and OTC distribution rather than a meaningful share of revenue. The HIMS premium in the market is paid for subscription-cohort economics, not for channel diversification.
The lesson for private mid-market brands is that "diversifying channels" is not a free lever. Adding wholesale to a pure-DTC operator without the operating discipline to manage trade spend, retailer relationships, and SKU mix produces gross-margin compression with no offsetting CAC reduction — the worst of both worlds. The brands clearing the channel-diversification premium in the cohort have done one of two things: built the wholesale relationships from the operating ground up over multiple years (ELF starting in mass beauty, YETI starting in outdoor specialty) or built the retail footprint as a deliberate own-asset (WRBY). The contribution-margin work — landed cost, SKU profitability, fulfillment-cost discipline — is what makes the channel mix profitable, not the mix itself.
07 The public-to-private gap and what private mid-market should benchmark
When the private mid-market brands we advise benchmark themselves against the public cohort, three lines do most of the diagnostic work. Northbeam's 2025 cohort study makes the underlying private-market dynamic clear: median first-time CAC up 9% YoY, median MER down roughly 2 percentage points YoY, conversion rates declining, and first-time MER deteriorating faster than blended MER as growth gets subsidised by returning customers and brand demand rather than new-customer efficiency. The public cohort can survive on growth optics for longer than the private buyer will tolerate. Private buyers underwrite cash conversion.
First-time CAC, not blended CAC
The public-cohort gap is widest at the first-time CAC level. Brands in our practice who report blended CAC consistently understate the new-customer acquisition reality by 30-50% because subscription, returning customers, and email-driven orders pull the blended number down. Buyers reading the QoE will rebuild the first-time CAC from the order-tag data and underwrite from there. Surfacing the first-time CAC trajectory cleanly — and showing it stable or improving over the last 8 quarters — is the single highest-leverage piece of the contribution-margin narrative.
Contribution margin, decomposed and channel-specific
The cohort comparison above puts the floor at roughly 35% blended contribution margin for the private mid-market brand expecting to clear at the right multiple in a 2026 process. The decomposition matters more than the headline — buyers want gross margin minus paid-channel marketing minus shipping-and-processing, ideally split by channel (DTC, wholesale, Amazon, retail). The brands that arrive at the process with a channel-by-channel contribution waterfall save the diligence team weeks of rebuild work and preserve the multiple in negotiation.
OCF as the cross-cohort comparable
The 12.7-13.9% operating-cash-flow margin band across the public cohort is the cleanest single benchmark we know of. Private mid-market brands clearing 10%+ OCF margin are in the public-cohort neighbourhood; brands below 5% are underwritten against the OLPX compression case. The OCF discipline matters more than the GAAP operating margin number in 2026 buyer underwriting — the gap between FY25 operating margin and FY25 OCF margin across the cohort is wide (e.g. WRBY -0.6% operating, +12.7% OCF), and buyers have learned to look through to the cash number.
08 What this means for sellers, buyers, and operators in 2026
Three operational implications come out of the Q2 2026 cohort read, sequenced by who is asking the question.
- 01 For sellers planning a 2026 or 2027 process: Rebuild the contribution-margin layer first; benchmark against the relevant sub-cohort, not the headline cohort. If your category sub-cohort is beauty (ELF, OLPX) the contribution-margin floor is approximately 45%; if it is hardgoods (YETI), approximately 40%; if it is subscription DTC (HIMS), approximately 30% but with a 4x+ LTV:CAC defending the band. The companion 18-month exit-prep work — landed cost on the SKU profitability layer, returns reserve adequacy, channel-by-channel contribution waterfall — is what holds the contribution-margin number through diligence. Cross-link to the SKU profitability and landed-cost playbook (/blog/sku-profitability-dtc-landed-cost/) and the returns-reserve methodology (/blog/returns-reserve-most-mismodeled-dtc-finance/) for the operating layers.
- 02 For sellers already in a process: The QoE will rebuild the first-time CAC, the channel-mix contribution, and the cohort retention curves from the underlying data. Surfacing these cleanly in the data room — with sub-cohort public-comparable benchmarks attached — preserves approximately 0.5-1.0 turns of multiple on its own. The DTC and CPG mid-market M&A multiples Q2 2026 read (/blog/dtc-cpg-mid-market-ma-multiples-q2-2026/) is the companion benchmark for what the headline multiple should be by sub-sector and operating shape.
- 03 For buyers underwriting against the cohort: The public-cohort contribution-margin floor is approximately 35% across the five names. Brands in the private mid-market reporting below the floor are either undermanaged on the marketing layer (CAC compression is the highest-ROI lever) or facing a structural channel-mix challenge (pure-DTC into a high-CPM environment with weak retention). The diligence question is whether the gap to the cohort floor is operationally closeable in the buyer's hold period — and the answer is almost always yes for the marketing-layer brands and almost always no for the structural-mix brands.
We publish this cohort read quarterly. The Q3 2026 update is scheduled for August 2026 and will incorporate the next round of 10-K and 10-Q updates from the cohort, the FRED ECOMSA and UMCSENT trajectories, and the next-quarter Tinuiti and Get-Ryze CPM benchmarks. The working-capital companion read for DTC operators sits at /blog/dtc-working-capital-q2-2026/ for the second-line operating benchmark.
Frequently asked questions
What is the median public DTC and CPG gross margin in FY25 / FY26?
How much has DTC CAC increased post-iOS-privacy changes and CPM inflation?
What is a healthy contribution margin for a private mid-market DTC brand in 2026?
Why is Olaplex an important cautionary case for DTC and CPG sellers?
How does Warby Parker compare to Olaplex as a public benchmark for private DTC?
What is the channel-diversification premium and how does a private brand qualify?
What should a private mid-market DTC brand benchmark against the public cohort?
Public-company financials: SEC EDGAR XBRL filings for e.l.f. Beauty (ELF, FY26 10-K filed 2026-05-21), Olaplex Holdings (OLPX, FY25 10-K filed 2026-03-05), YETI Holdings (YETI, FY25 10-K filed 2026-02-27), Warby Parker (WRBY, FY25 10-K filed 2026-02-26), Hims & Hers Health (HIMS, FY25 10-K filed 2026-02-23).
Macro and consumer context: Federal Reserve Bank of St. Louis FRED series ECOMSA (E-commerce Retail Sales, quarterly SA, $M) and UMCSENT (University of Michigan Consumer Sentiment, monthly), observations through April 2026.
CPM, CPA, and signal-loss benchmark data: Tinuiti Q1 2026 Digital Ads Benchmark Report; Get-Ryze 2026 Meta benchmarks (CPM, CPC, CPA by industry); AdAmigo 2026 Meta CPL benchmarks; McCarthy et al. (SSRN 2024) "Evaluating the Impact of Privacy Regulation on E-Commerce Firms"; AppLovin AXON 2.0 operator commentary via Triple Whale and DTC case studies.
Private mid-market unit-economics framework: ATTN Agency 2026 DTC Exit Valuation Guide (contribution-margin band methodology); Northbeam 2025 cost-of-growth cohort study (first-time CAC and MER trajectories); Finaloop DTC contribution-margin construction playbook; Klaviyo cohort-analysis methodology for retention curves.
Full source list and per-company financial extracts at content-pipeline/research/public-dtc-unit-economics-2026-cac-gm-contribution/sources.md in the Putra & Co content pipeline.