When does each device cohort pay back its acquisition cost through service and accessory revenue, and how does it compound?
№ 133 · Device ROAS curve
Definition
MetricCumulative revenue per acquired device cohort divided by acquisition cost, by months since acquisition
Unitx (cumulative contribution over fully loaded acquisition cost), plotted by month since acquisition
Cumulative device plus service margin per acquired customer divided by acquisition cost, plotted across months. Multiple cohort lines. Build by tracking per-cohort device profitability cumulatively. Example: Top cohort 1.0x at M18; bottom cohort never crosses (kill that channel).
- ROAS
- Return on ad spend. Revenue attributed to a campaign divided by the spend on that campaign.
Benchmarks
| Bottom 30% | Median | Top 30% |
|---|---|---|
| Still below 1.0x on contribution at month 18 | Break-even on contribution at 9 to 12 months, around 2.0x by month 24 | Above 1.0x on contribution by month 6 and above 3.0x by month 24 |
Direct-to-consumer consumer hardware with attached accessory and subscription revenue, 2025 to 2026. Contribution basis, meaning gross profit after returns, warranty accrual and fulfilment, over fully loaded acquisition cost including creative, agency and tooling. · Definitions collide badly on this card and the numbers are not interchangeable. Platform ROAS is modelled revenue over channel spend and one 2025 to 2026 panel found per-channel ROAS overstates true return by around 2.3x, so a reported 2.87x average ecommerce ROAS is not comparable to the cohort contribution curve described here. Published payback figures for DTC electronics also conflict: one source sets a healthy target under four months while the same publisher's vertical table observes electronics at six to twelve months or more and names it the category that extended most in 2025 to 2026. Treat the sub-four-month figure as a target for a high-frequency DTC model and the six to twelve month range as what one-time-purchase hardware actually does. Rule of thumb: break-even MER equals one divided by contribution margin, so a 40% contribution brand needs 2.5x before anything is earned.
Category split omitted: The available payback splits compare hardware against other DTC verticals such as beauty and food, which is a cross-model comparison rather than a split within connected hardware, and no located source breaks the curve down by device category.
When it looks bad
The curve jumps almost vertically in month zero from the device sale and then runs nearly flat, so each cohort is a step rather than a curve and the entire return is decided at the moment of purchase.
A cohort reaches 0.9x contribution at month one from device margin alone, then adds about 0.05x a month and is still at 1.3x by month 18 against a $92 fully loaded CAC.
What to do about it
- Plot the curve on contribution after returns, warranty accrual and fulfilment rather than on gross revenue. At a 35% device margin a 10% return rate removes more contribution than most plans model, and electronics return rates run around 10% with smart home around 13%.
- Attack the month two to month six slope with accessory and consumable offers before adding prospecting spend. A second-purchase rate moving from roughly 22% to 35% within 90 days shortens payback faster than any CAC optimisation available on the media side.
- Convert device buyers to a paid subscription at activation, not later. Roughly $45 of annual service revenue per subscriber turns a flat cohort line into a compounding one inside two years and is the only reliable frequency lever a one-time-purchase hardware model has.
- Steer on blended MER against a break-even line computed from your own contribution margin, since break-even MER equals one divided by contribution margin, and stop comparing to category-average ROAS figures that one panel found overstate true return by around 2.3x.
Sources
- Eightx Electronics CAC payback runs 6 to 12 months or more, against 1 to 3 months for food and beverage and 2 to 4 for beauty and pet. Electronics extended the most across 2025 to 2026 as one-time purchase economics met rising CAC with no frequency lever. eightx.co ↗
- Eightx Average CAC is around $76 per customer (attributed to FirstPageSage 2026), one of the highest of any ecommerce vertical and reportedly up around 40% year on year. Online electronics return rate is around 10%, smart-home around 13%, against about 19.5% for all ecommerce. eightx.co ↗
- Eightx States a healthy DTC electronics CAC payback under four months against a broad-DTC median near 3.4 months and a 3:1 LTV to CAC floor, implying roughly $228 of lifetime contribution at a $76 CAC, about 3.4x a single order. eightx.co ↗
- Eightx The Triple Whale panel shows a median MER around 2.4x, about 41% ad spend share, below healthy floors, because the panel skews toward growth-stage brands running aggressive acquisition. Consensus CAC payback median is around 3.4 months across ecommerce generally. eightx.co ↗
- Daymark Average ecommerce ROAS was about 2.87x in 2025 with roughly half of stores below 2.0x. Break-even ROAS and break-even MER are both one divided by contribution margin, so a 20% margin brand needs 5.0x just to break even and a 50% margin brand breaks even at 2.0x. usedaymark.io ↗
- LayerFive Across 200-plus brands, platforms overstated true ROAS by 2.3x and brands measuring only ROAS over-spend on paid by 20% to 40%. Median DTC contribution margin fell from 35% in 2021 to 22% in 2025. Targets given are LTV to CAC of 3:1 to 5:1, CAC payback under 90 days and MER of 2.5x to 4x. layerfive.com ↗
- MetricUno A blended ROAS of 3.0x to 4.0x is healthy for most Shopify stores and a target MER of 3.5x to 4.5x is typical for healthy DTC. Branded search typically reports 6x to 10x and inflates the blended figure by 0.5 to 1.5 points. metricuno.com ↗
- US SEC (GoPro, Inc.) 2.36 million subscribers generated $106m of subscription and service revenue in 2025, roughly $45 of annual service revenue per subscriber, the compounding component that separates a flat cohort line from a rising one. sec.gov ↗