How Much Should You Actually Spend on Ads?
The honest answer depends on numbers most brands haven't calculated. Here's the calculation, and when the answer is 'less than you are.'
The short answer
Work backwards from contribution margin and repeat rate to find what a customer is worth, then spend against that. If conversion or retention are weak, the right budget is lower than you think — fix those first.
There is no percentage-of-revenue answer to this, despite how often one gets quoted. The right budget depends on what a customer is worth to you and how quickly you get that money back.
The calculation
Work backwards.
1. Contribution margin per order. Revenue minus cost of goods, shipping, payment processing and expected returns. Not gross margin — the number left after the order is genuinely fulfilled.
2. Expected orders per customer. Your repeat rate, measured by cohort over a window that matches your purchase cycle. This is the number most brands guess at and should not.
3. Customer value. Contribution margin × expected orders. This is what a new customer is actually worth.
4. What you are willing to pay for one. Some fraction of that value, depending on how fast you need the cash back. Paying 30% of customer value is comfortable. Paying 80% is a growth bet that requires the cash to survive it.
5. Multiply by how many customers you want. That is your budget.
Most brands do this in the opposite direction — pick a budget, spend it, then work out whether it was sensible.
The payback question people skip
Two businesses can have identical CAC and completely different outcomes.
If you recover acquisition cost on the first order, you can reinvest immediately and grow as fast as the channel allows. If you recover it over four purchases across fourteen months, you are financing growth out of working capital, and scaling spend can strain the business even while the unit economics look fine on a spreadsheet.
Know your payback period before you scale. It determines how fast you can safely grow, which is a different question from whether the ads work.
When the answer is “spend less”
This is the part that gets left out of most advice.
If your conversion rate is weak, more traffic converts weakly at greater expense. A brand converting at 1% that could convert at 2% is paying double for every customer. Fixing that is worth more than doubling the budget — and it is Retain & Optimize, not paid media.
If your repeat rate is weak, you are buying transactions rather than customers, and the whole calculation above collapses to a single order. Almost nobody can profitably acquire on one purchase in a competitive category.
If you have no capture, you are paying for visitors and keeping nothing from the ones who do not buy today — which is most of them. Email capture converts a wasted click into a future opportunity.
This is why Traffic with Intent is the fourth stage of the framework and not the first. Traffic is the loudest lever and the last one to pull.
Splitting the budget
Once the system is ready, a workable starting split:
- Roughly two-thirds to what is already working, at a scale you know is profitable
- Roughly a quarter to scaling and iterating on winning creative and audiences
- A small slice to genuine experiments — new channels, new formats, things that will mostly fail
The experimental slice matters. Channels saturate, costs rise, and the brand with no pipeline of alternatives discovers this at the worst possible moment.
The number to watch instead of ROAS
ROAS is useful but flattering, mostly because it is usually measured on last click and ignores margin.
Two better ones:
Contribution margin after ad spend. Actual money, after the actual costs.
Blended CAC — total marketing spend divided by total new customers, across all channels. It cannot be gamed by attribution windows, and it is the number that tells you whether marketing overall is working.
If platform-reported ROAS is climbing while blended CAC is also climbing, the platform is claiming credit for customers you were getting anyway.
Where this sits in the framework
Paid media belongs to Traffic with Intent — a lever, deliberately placed after the system that has to catch what it brings.
Scale a tuned system and every new visitor is worth more, so growth compounds. Scale a broken one and more traffic just exposes more leaks at a higher cost.
Related reading
Affiliate Program Development: How to Build One That Doesn't Eat Your Margin
Most affiliate programs pay commission on sales that would have happened anyway. Here's how to structure one that actually creates incremental revenue.
Getting Cited by ChatGPT: What Actually Works
AI assistants are becoming a discovery channel. The optimisation for being cited is different from ranking, and mostly comes down to being quotable.
Ambassador vs Referral Programs: Which One Actually Fits Your Brand
They get used interchangeably and they solve different problems. One turns customers into a distribution channel, the other turns fans into a content engine.