Open your ad platform and it will tell you how well your ads performed. Open another platform and it will say the same thing about itself. Add the two claims together and you may find they report more revenue than your store actually took.
This guide explains five measures that give you a steadier view: ROAS, CAC, blended CAC, MER and contribution margin. We use one fictional store throughout, so you can follow the arithmetic and then repeat it with your own numbers.
The key terms, with one example store
Meet a fictional store. In one month it spends $10,000 on ads: $6,000 on Meta and $4,000 on Google. The store records 400 orders and $25,000 in total revenue. Of those orders, 250 came from new customers.
- ROAS (return on ad spend) is revenue attributed to ads divided by ad spend. If Meta says its ads drove $18,000 on $6,000 spend, Meta ROAS is 3.0.
- CAC (customer acquisition cost) is what you pay to win one new customer. Spend divided by new customers.
- Blended CAC uses all marketing spend and all new customers, from every channel. Here, $10,000 divided by 250 is $40.
- MER (marketing efficiency ratio) is total revenue divided by total marketing spend. Here, $25,000 divided by $10,000 is 2.5.
- Contribution margin is revenue minus the costs that rise with each order, as a share of revenue.
Suppose product cost is $10,000, shipping and packaging $2,500, and payment fees $750. That is $13,250 of variable costs. Revenue minus these is $11,750, so contribution margin is 47% ($11,750 divided by $25,000). After $10,000 of ads, $1,750 remains to cover rent, salaries and software.
Why platform numbers disagree
Each platform counts the sales it takes credit for. Suppose Meta says $18,000 and Google says $12,000. Together they claim $30,000. Yet the store only took $25,000, and some of that came from email, organic search and direct visits.
This happens for a few plain reasons:
- Overlap. A shopper sees a Meta ad, later clicks a Google ad, then buys. Both platforms may claim the order.
- Different attribution windows. One may credit a purchase made several days after a click, or after only seeing an ad.
- Tracking gaps. Privacy settings and browser limits hide some activity, so platforms estimate.
- Credit for sales that would have happened anyway, such as people searching your brand name.
None of this means platforms are lying. They measure what they can see. Your bank balance and order system see everything.
Set a break-even ROAS from margin
Break-even ROAS tells you the lowest ROAS at which ads pay for themselves. The formula is:
Break-even ROAS = 1 divided by contribution margin
With a 47% margin, that is 1 divided by 0.47, about 2.13. Check it: $10,000 at ROAS 2.13 gives $21,300 of revenue. At 47%, that leaves $10,011, which covers the $10,000 spend.
So the store needs ROAS above roughly 2.13 just to break even on the first order. A platform ROAS of 3.0 sounds good, but if it is inflated by overlap, the true figure may sit closer to the line.
Use contribution margin, not gross margin, as margin that ignores shipping and fees will make break-even look easier than it is.
New versus returning customers
Ads are often best judged on new customers, since returning shoppers might have bought anyway. Back to the store: 250 orders came from new customers, bringing $15,000. The other 150 orders brought $10,000.
If all $10,000 of spend is aimed at new customers, new-customer ROAS is 1.5 ($15,000 divided by $10,000). That is below break-even for a first order.
Per customer, the first order brings $60 of revenue, or $28.20 of contribution at 47%. Blended CAC is $40. The gap is $11.80 per customer.
Payback period and repeat purchase
Payback period is how long it takes a customer to earn back what you paid to acquire them. Here the store needs repeat orders to close the $11.80 gap.
A repeat order averages about $66.67 ($10,000 divided by 150), which gives roughly $31.33 of contribution. Illustratively, if 40 of every 100 customers place one repeat order, that returns about $12.53 per customer (0.4 times $31.33). That just covers the gap.
This depends on real repeat behaviour. Check your own order history: of customers acquired three or six months ago, how many bought again, and when? If you sell items bought once, such as a mattress, you cannot count on repeats.
A one-page weekly dashboard
Keep it to six numbers, compared with the previous week:
- Total revenue.
- Total ad spend.
- MER.
- Blended CAC.
- New-customer share of orders.
- Contribution margin after ads, in dollars.
Add platform ROAS beside it as a reference, not as the decision-maker. Fill it in every Monday from your store system and ad accounts.
When platform and blended numbers disagree
Use simple rules.
- If platform ROAS is up but MER is flat, be sceptical. The platform may be claiming sales from other channels.
- If MER and blended CAC both improve, the business is likely getting better, whatever one platform reports.
- Change spend in steps. Raise or cut a channel by about 10 to 20%, wait one to two weeks, then check MER and new customers.
- Run a holdout if you can: pause a channel in one region and compare. Interpret with care.
Common mistakes
- Adding up platform revenue and treating it as total sales.
- Using gross margin to set targets.
- Ignoring new versus returning customers.
- Reacting to single days. Look at weekly trends.
- Chasing one ratio. A high ROAS on a tiny budget may not scale.
What to do next
- Calculate your contribution margin from last month's orders, including shipping, packaging and fees.
- Work out your break-even ROAS using 1 divided by that margin.
- Compute MER and blended CAC for the last four weeks.
- Pull repeat purchase data for customers acquired 90 days ago to estimate payback.
- Build the six-number weekly sheet and review it every Monday for a month before changing budgets.