Blended CAC vs Channel CAC: Which One a CR Lift Actually Moves
After a winning conversion test ships, blended CAC drops the same week but Meta and Google CAC lag for 4-6 weeks while the platforms relearn. Here's how to read both during the gap.
Blended CAC vs Channel CAC after a CR lift
A site-side conversion lift cuts blended CAC the same week, but channel-reported CAC takes 4-6 weeks to reflect it while ad platforms relearn.
Blended CAC is total marketing spend divided by total new customers, calculated outside the ad platforms — usually in your finance sheet or analytics tool. Channel CAC is what Meta Ads Manager or Google Ads reports for that channel specifically, based on platform-attributed conversions.
These two numbers respond to a conversion-rate win on completely different timelines. The moment a winning variant ships and more sessions turn into orders at the same ad spend, blended CAC falls. Channel CAC barely moves for the first two weeks because the algorithm is still bidding off its prior conversion-rate assumption, and only fully catches up after roughly 4-6 weeks of learning.
You ship a winning checkout test on a Tuesday. By Friday, blended CAC is down 12% in your dashboard. You open Meta Ads Manager expecting the same story — and CAC there looks identical to last week. Nothing broken, just two metrics that update on different clocks.
The gap matters because performance teams use channel CAC to make daily budget calls. If you wait for Meta's reported CAC to confirm the win, you'll under-spend a profitable channel for a month. If you over-trust blended CAC, you'll mis-attribute a seasonal tailwind to your test.
How each CAC metric moves in the weeks after a +10% CVR lift ships
| Weeks after launch | Blended CAC change | Meta-reported CAC change | Google-reported CAC change |
|---|---|---|---|
| Week 1 | -9% to -10% | -1% to -2% | -2% to -3% |
| Week 2 | -9% to -10% | -3% to -4% | -4% to -5% |
| Week 3-4 | -9% to -10% | -5% to -7% | -6% to -8% |
| Week 5-6 | -9% to -10% | -8% to -10% | -8% to -10% |
| Week 7+ | -9% to -10% | -9% to -10% | -9% to -10% |
The table assumes traffic mix and AOV stay flat — in reality both shift slightly as the platforms re-optimise, which is why channel CAC sometimes overshoots blended CAC briefly in week 5-6 before settling. The shape of the curve, not the exact numbers, is what to internalise.
What blended CAC actually shows after a CR lift
Blended CAC moves immediately because the math is mechanical: same spend, more orders, lower cost per order. There's no learning phase, no attribution model, no platform deciding which click gets credit. If your CVR went from 2.4% to 2.7% on flat traffic, blended CAC drops about 11% the same week.
That's also why blended CAC is the right number to report to finance and to use in your CRO impact on CAC story to leadership. It reflects the unit economics the business actually lives with, not what any one platform claims credit for. The companion metric — blended vs channel ROAS — behaves the same way for the same reasons.
The seasonality trap
Blended CAC's instant response is also its weakness. If your test shipped the week Black Friday teasers started, some of that drop is the season, not your variant. Always read post-launch blended CAC against a holdout or a year-over-year same-week baseline — not against the four weeks before launch.
What channel CAC shows during the platform relearn
Meta's bidding system optimises against an expected conversion rate it learned from your historical traffic. When your site suddenly converts 10% better, the algorithm doesn't know yet — it keeps bidding the old amount per click, which is now too low for the new conversion economics. Reported CAC stays flat because spend per conversion looks the same to the platform.
Over 4-6 weeks the algorithm relearns, bids up on previously-marginal audiences that now convert profitably, and reported CAC drifts down to meet blended CAC. Google Ads typically catches up slightly faster than Meta because Smart Bidding has shorter lookback windows on most shopping campaigns.
CAC index after a +10% CVR win ships (week 0 = 100)
Blended CAC
Meta-reported CAC
Google-reported CAC
Blended CAC vs channel CAC after a CR lift — FAQ
Report blended CAC as the headline, with the year-over-year same-week comparison as the baseline. It reflects the actual unit economics and isn't distorted by the platform relearn lag. Channel CAC is useful as a secondary line to show where the platforms haven't caught up yet.
Increase gradually. Blended CAC has already dropped, so the headroom is real — but scaling Meta spend before the algorithm finishes relearning can trigger another learning phase and delay the catch-up. A 10-15% weekly budget step is usually safe.
Meta's bidding optimises against a learned conversion rate. Until enough new conversion events flow in to update that model — usually 50+ conversions per ad set over 1-2 weeks — bids stay calibrated to your old CVR and reported CAC barely moves.
Same dynamic, inverted denominator. Blended ROAS rises immediately after a CVR or AOV win; channel ROAS lags 4-6 weeks while platforms relearn. If you're comfortable reading one, you can read the other.
Yes. With weaker signal, platforms need more conversion volume to update their model, which stretches the relearn window. Stores running aggregated event measurement on Meta often see closer to 5-7 weeks before reported CAC fully catches up.
Usually means your CRO win shifted the traffic mix — for example, more conversions from organic and email, so paid is taking credit for a smaller share of total orders. Check new-customer share by channel before concluding the channel got worse.
Blended CAC: 2-3 weeks post-launch is enough to rule out a one-week fluke. Channel CAC: wait the full 4-6 weeks. If you need to defend the win to finance sooner, lead with blended and footnote the channel lag.
Algorithmic channels (Meta, Google, TikTok) all have a relearn lag. Manual-bid channels (some affiliate, some influencer flat-fees) update as fast as blended CAC because there's no algorithm to retrain. Search brand campaigns also catch up faster because intent is already high.
Somewhat. Make sure your conversion events are firing cleanly post-test, avoid simultaneous big changes to creative or audiences, and keep budgets stable for the first two weeks so the algorithm has consistent signal. Don't pause and restart ad sets — that resets learning.
For stores above ~€5M revenue, yes — a lightweight media mix model resolves blended-vs-channel gaps better than either dashboard alone. Below that, the cost of running MMM outweighs the precision gain; blended CAC plus a YoY baseline is enough.
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.