How to use CAC by Channel
Blended CAC hides where your acquisition spend is actually working. Here's how to calculate, benchmark, and act on channel-level CAC across Meta, Google, TikTok, email, and organic.
CAC by Channel
Customer acquisition cost calculated separately for each acquisition channel — Meta, Google, TikTok, email, organic, referral.
CAC by channel splits your blended acquisition cost into per-channel numbers so you can see which sources are actually pulling their weight. Instead of one company-wide CAC of, say, €38, you get Meta at €31, Google Brand at €9, Google Non-brand at €54, TikTok at €72, and email at €4 — and suddenly the reallocation decision writes itself.
It sits one level below blended CAC measurement and one level above channel-specific CAC reduction tactics. It's the diagnostic that turns the vague complaint 'CAC is too high' into the specific instruction 'TikTok CAC is 2x Meta CAC at the same scale — cap TikTok and push the delta into Meta retargeting.'
Most stores in the €1M–€15M range track blended CAC religiously and then can't explain why it crept from €34 to €41 over a quarter. The answer is almost never 'all channels got worse equally.' It's that one channel scaled into diminishing returns while another stayed efficient, and the average hid it.
Channel-level CAC is the resolution upgrade. Same denominator (new customers), same numerator philosophy (fully-loaded spend), but cut by where the customer first came from. Once you can see Meta CAC sitting at €28 while TikTok CAC sits at €71, the conversation moves from 'we need to lower CAC' to 'we need to cap TikTok at last month's spend and push the delta into Meta lookalikes.'
Why blended CAC hides your real decisions
Blended CAC is one number describing five-to-eight very different motions. Meta prospecting, Meta retargeting, Google brand, Google non-brand, TikTok, organic search, email, and referral all have wildly different cost structures, intent signals, and scaling curves. Averaging them is like averaging your best and worst SKU margin and calling it 'product profitability.'
The specific failure mode: a channel that scales linearly with spend (Meta prospecting) gets averaged with one that's effectively free at the margin (organic, returning email). Your blended number looks fine until you try to grow — then it spikes, because the cheap channels are capacity-bound and every incremental euro is going to the expensive channel.
The fix is structural, not tactical. You need a per-channel CAC view refreshed weekly, sitting next to a per-channel new-customer count, so you can see both the cost and the volume each source is delivering. Without it, the finance team and the performance team will argue past each other for months.
The trap: averaging brand and non-brand Google
If you report 'Google CAC' as a single number, you're hiding the most important split in your paid stack. Brand search is usually €5–€15 per customer; non-brand can be €40–€90. Lump them together and you'll either over-invest in non-brand (because brand subsidises the average) or starve brand (because non-brand drags it). Always split them.
How to calculate CAC per channel
The formula is the same as blended CAC — total fully-loaded spend divided by new customers acquired — but applied within each channel boundary. Fully-loaded means platform spend plus creative production plus the share of agency or in-house salary attributable to that channel. Stores that only count platform spend systematically understate paid CAC by 15–25%.
Attribution model matters more than people admit. Last-click will overcredit brand search and underweight TikTok and Meta upper-funnel. Data-driven attribution in GA4 or a position-based model gives a more honest split. Pick one model, document it, and stick with it for at least a quarter — comparing CAC across channels under different models is meaningless.
Typical channel-level CAC for an apparel store on Shopify (€60 AOV)
The shape above is more important than the absolute numbers. Owned channels (email, organic, brand search, referral) cluster at the bottom because they're harvesting demand you already created. Prospecting channels sit at the top because they're creating demand from cold audiences. A healthy channel mix has both — pure harvesting can't grow, pure creation burns cash.
Reading the numbers: what good looks like
Channel CAC only means something next to two other numbers: the channel's new-customer volume and its 90-day repeat rate. A channel at €70 CAC that drives 40% of new customers and has a 35% repeat rate is structurally different from a channel at €70 CAC delivering 5% of volume with a 12% repeat rate. The first is a growth engine, the second is a leak.
Below are ballpark ranges for a beauty or apparel store doing €3M–€8M annually. Treat them as orientation, not targets — your specific AOV, margin, and repeat behaviour move these by 30–50% in either direction.
Channel-level CAC ranges for DTC apparel & beauty (€40–€80 AOV)
| Channel | Typical CAC (€) | Share of new customers | 90-day repeat rate |
|---|---|---|---|
| Email / SMS | 3–8 | 5–10% | 40–55% |
| Organic search | 5–12 | 8–15% | 30–40% |
| Google Brand | 8–15 | 10–20% | 35–45% |
| Referral / word-of-mouth | 12–25 | 5–12% | 35–50% |
| Meta retargeting | 18–32 | 10–18% | 28–38% |
| Meta prospecting | 30–50 | 20–35% | 20–28% |
| Google Non-brand | 40–70 | 8–15% | 22–30% |
| TikTok ads | 50–90 | 5–15% | 15–22% |
Notice the inverse relationship between CAC and repeat rate. Customers acquired through email and organic are usually already in your orbit — they convert cheaply and stick around. TikTok-acquired customers often come for a single viral product and never come back, which is why a €70 TikTok CAC is much worse than a €70 Meta prospecting CAC in LTV terms.
Acting on the data: the reallocation playbook
Once you have weekly channel CAC, the first move is almost never 'cut the expensive channel.' It's 'find the channel where CAC is rising fastest at constant spend' — that's the one hitting diminishing returns. Cap it at last week's spend, then test whether the next-cheapest scalable channel can absorb the delta without its own CAC blowing up.
Pair this with the broader CAC reduction levers — creative refresh cadence, landing page conversion rate, offer testing — applied to the specific channel that's struggling. A 15% lift in landing page CVR on your Meta prospecting traffic drops Meta CAC by roughly 15%, which is often more impactful than moving budget around.
The weekly 15-minute review
Every Monday, pull a five-column table: channel, spend last week, new customers, CAC, week-over-week CAC delta. Flag any channel where CAC moved more than 15% in either direction. That's your entire performance meeting agenda — the channels that didn't move don't need discussing.
Frequently asked questions about CAC by channel
Blended CAC divides all acquisition spend by all new customers — one number for the whole business. Channel-level CAC does the same calculation within each channel separately, so you can see that Meta CAC is €31 while TikTok CAC is €72. Blended tells you the result; channel-level tells you the cause.
Pick one attribution model — GA4 data-driven or a position-based model are reasonable defaults — and apply it consistently. Don't mix last-click for one channel and view-through for another. The absolute numbers will be imperfect, but the relative comparison between channels stays directionally honest.
Yes, with attributed costs. Organic isn't free: SEO content, agency retainers, and the in-house time spent on it all count. Allocating those costs gives organic a realistic CAC (usually €5–€15) instead of zero, which prevents teams from over-celebrating 'free' acquisition that actually costs six figures a year.
Three usual reasons: TikTok's auction is still less efficient for direct-response than Meta's; TikTok-acquired customers convert at lower rates on traditional product pages; and TikTok attribution windows are shorter, undercounting conversions. The first two are real CAC problems; the third is a measurement artefact you can partially fix with post-purchase surveys.
Weekly at minimum, with a rolling 4-week view to smooth noise. Daily is too noisy for most stores under €10M — you'll chase signal that isn't there. Monthly is too slow to catch a channel sliding into diminishing returns before you've wasted significant spend.
For most online stores, aim for payback inside 3 months on owned channels (email, brand search, organic), 4–6 months on Meta and Google non-brand, and under 9 months on TikTok and upper-funnel YouTube. Anything beyond 12 months is a cashflow problem regardless of LTV, unless you have very patient funding.
Yes. If you spend €4K/month on UGC creators for TikTok, that's part of TikTok CAC. Excluding it makes TikTok look artificially cheap and Meta look artificially expensive (if your Meta creative comes from the same studio). Allocate creative costs to the channel they're produced for, or split proportionally if shared.
CAC alone is incomplete — a €70 CAC channel with €280 LTV is healthier than a €30 CAC channel with €60 LTV. Always pair channel CAC with channel-level LTV (or at least 90-day repeat rate as a proxy) before making reallocation decisions. Some channels acquire fundamentally different customers.
There's no universal target, but stores that scale sustainably usually have 30–50% of new customers from owned/earned channels (email, organic, brand search, referral) and 50–70% from paid. Skewing too far to paid means your CAC will rise the moment you turn up spend; too far to owned means you can't grow above your organic ceiling.
Yes — they're effectively different channels. Meta retargeting harvests warm intent at €18–€32 CAC; Meta prospecting creates new demand at €30–€50. Lumping them gives you a meaningless midpoint, and it hides whether your retargeting pool is shrinking (a leading indicator that prospecting needs more budget, not less).
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.