LTV:CAC by Acquisition Channel: When Blended Lies
A blended 3:1 LTV:CAC routinely masks paid-social cohorts burning cash under healthy organic and email numbers. Here's how to split it by channel — and what to do once you see the truth.
Quick answer
Blended LTV:CAC averages your worst channel with your best, so a 3:1 headline can hide a Meta prospecting cohort at 1.2:1 propped up by organic and email at 8:1+. Recalculate LTV:CAC per acquisition channel using first-touch cohorts and 12-month repeat behaviour — that's the only view that tells you where to cut and where to scale.
LTV:CAC by Acquisition Channel
Channel-level LTV:CAC measures the lifetime-value-to-acquisition-cost ratio separately for each traffic source instead of as one blended average.
Channel-level LTV:CAC splits the standard ratio by the source that originally acquired the customer — paid social, paid search, organic, email, referral, affiliate — so you can see which channels actually fund the business and which ones quietly drain it.
The mechanic is simple: tag each new customer with their first-touch acquisition channel, then track the 12- or 24-month revenue of that cohort against the fully-loaded CAC for that channel. A blended company-wide LTV:CAC of 3:1 is a portfolio average; the channel view almost always shows a 5-10x spread between the strongest and weakest sources.
Most Shopify brands in the €1M–€15M range track one company-wide LTV:CAC number and stop there. It looks healthy, the board accepts it, and Meta budgets keep growing month over month.
The problem shows up six months later, when blended ratio drifts from 3.1 to 2.4 and nobody can explain why. The reason is almost always the same: paid social was always underwater, and organic was carrying it.
Why blended LTV:CAC misleads
Blended LTV:CAC is a weighted average. A channel acquiring 60% of your customers at 1.2:1 gets diluted by a channel acquiring 15% at 8:1, and the math lands somewhere comforting in the middle. The headline number stays green while the marginal euro you're spending stays red.
It also hides cohort drift. When you scale Meta prospecting from €40k to €120k/month, the new cohorts have lower repeat rates and weaker AOV than the early adopters who came in via organic search. Blended LTV:CAC won't catch that until the bad cohorts age into your 12-month window — usually a quarter too late to react.
The 3:1 trap
If your blended LTV:CAC is 3:1, your worst paid channel is almost certainly under 1.5:1. The rule of thumb: the spread between your best and worst channel is usually 4-7x. A 3:1 average with that spread means roughly half your paid spend is unprofitable on a 12-month basis.
How to split the ratio by channel
Start with first-touch attribution, not last-click. A customer who first discovered your brand on Instagram, then converted three weeks later via a branded Google search, belongs to paid social — that's the channel that paid to put them in your funnel.
Then build cohort LTV curves per channel. For each first-touch source, take all customers acquired in a single month and track their cumulative revenue at month 1, 3, 6, 12. You need at least 6 months of history per cohort for the ratio to mean anything; importing historical GA4 data on day one removes the cold-start problem.
Load CAC fully. That means ad spend plus the platform fee, plus any creative-production cost attributable to that channel, divided by net new customers (not orders). The most common error here is using gross orders, which double-counts returning customers and inflates every channel by 20-40%.
What a real channel split looks like
Typical 12-month LTV:CAC by channel for a Shopify apparel brand at €5M GMV
| Channel | CAC (€) | 12-mo LTV (€) | LTV:CAC | Share of new customers |
|---|---|---|---|---|
| Organic search | 8 | 165 | 20.6 : 1 | 18% |
| Email / SMS (Klaviyo) | 5 | 210 | 42.0 : 1 | 9% |
| Referral | 12 | 180 | 15.0 : 1 | 6% |
| Paid search (brand) | 14 | 155 | 11.1 : 1 | 11% |
| Paid search (non-brand) | 62 | 140 | 2.3 : 1 | 12% |
| Meta retargeting | 28 | 130 | 4.6 : 1 | 14% |
| Meta prospecting | 95 | 115 | 1.2 : 1 | 27% |
| TikTok prospecting | 78 | 105 | 1.3 : 1 | 3% |
| Blended | 48 | 138 | 2.9 : 1 | 100% |
Notice what the blended row hides: paid social and TikTok prospecting — together 30% of acquired customers — are under 1.5:1. The blended 2.9:1 only looks acceptable because organic, email, and brand search are doing the heavy lifting at 10:1+. Cutting Meta prospecting by 40% and reallocating to retargeting and non-brand search typically moves blended LTV:CAC from 2.9 to 3.6 within two quarters.
Where to cut, where to scale
Channels below 2:1 are losing money on a 12-month horizon — cut budget by 30-50% and watch what happens to blended LTV. If blended LTV holds, you found subsidy spend. If it drops proportionally, those customers were actually contributing and the issue is creative or audience targeting, not the channel itself.
Channels above 5:1 are usually under-invested. The instinct is to protect the ratio, but a 10:1 channel with 6% volume share is leaving money on the table. Test scaling spend until the ratio compresses to 4:1 — that's where marginal CAC equals marginal LTV and incrementality is maxed.
Experiments to run once you can see the split
Run a 4-week Meta prospecting holdout in one geo. Pause prospecting in, say, the Netherlands while keeping retargeting live, and measure incremental organic + brand-search lift against a matched control geo. Most brands find Meta prospecting incrementality is 40-60% of what last-click reports — which makes the real LTV:CAC closer to 0.7:1, not 1.2:1.
Then run channel-specific welcome flows. Customers acquired via paid social have different intent signals than email subscribers — a Klaviyo flow tailored to first-touch channel typically lifts second-order rate by 8-15%, directly improving the LTV side of the ratio for your weakest channels.
Frequently asked questions
Blended LTV:CAC averages all customers regardless of source, producing one company-wide ratio. Channel-level splits that ratio by first-touch acquisition source, exposing the per-channel economics that the blended average smooths over. Blended is fine for board reporting; channel-level is what you use to make budget decisions.
First-touch. Last-click over-credits channels at the bottom of the funnel — brand search, email, direct — and under-credits the prospecting channels that actually paid to acquire the customer. If you measure channel LTV:CAC on last-click, you'll cut the channels filling your top of funnel and wonder why volume collapsed.
Minimum 6 months, ideally 12. Repeat behaviour stabilises around month 4-6 for most apparel and beauty brands, longer for considered purchases like furniture or electronics. Anything under 90 days is essentially first-order CAC payback, not LTV:CAC.
Retargeting captures customers already in-market who would likely have converted anyway, so its CAC is low and its LTV is similar to organic. Prospecting pays to interrupt cold audiences, so CAC is 3-4x higher and the resulting cohort has weaker repeat rates. The retargeting ratio is partly an illusion — much of it is cannibalised organic conversions.
Above 3:1 is healthy, 2-3:1 is acceptable for prospecting channels driving top-of-funnel volume, under 2:1 is a problem. Owned channels like email and referral should run 10:1+ — if they don't, the issue is usually under-investment in retention flows, not the channel itself.
Stamp them with their first-touch channel and don't change it. Multi-touch models that distribute fractional credit across channels make the math more 'fair' but unusable for budget decisions — you can't allocate 0.3 of a euro to Meta. Pick first-touch, document the choice, stay consistent quarter to quarter.
It changes the CAC side. Meta-reported CAC is now systematically understated because conversions get attributed late or lost entirely, so your Meta LTV:CAC looks better in Ads Manager than it actually is. Use server-side attribution or GA4 first-touch as the source of truth, not the Meta dashboard.
Yes — use net revenue, not gross. Returns hit some channels harder than others; paid-social impulse buyers return at 2-3x the rate of organic-search buyers in apparel. If you measure LTV on gross revenue, you'll systematically overstate your worst channels and make exactly the wrong cuts.
Monthly for trending, quarterly for budget decisions. Monthly cadence catches cohort drift early — when this month's Meta prospecting cohort is suddenly weaker than last month's, you want to know in 30 days, not when the 6-month cohort closes. Quarterly is when you actually reallocate spend.
Both, in that order. First cut spend by 30-40% to remove the most marginal customers, which usually lifts the ratio by 0.3-0.5 on its own. Then work the LTV side — channel-specific welcome flows, post-purchase upsells, and audience-tuned creative typically add another 15-25% before you've changed budget allocation.
Track CAC, channels, and funnel conversion in one place
Metricuno connects ad spend, funnel events, and revenue so you can see CAC by channel, cohort, and campaign — without stitching together five tools.