What a healthy cost per result actually looks like for a low ticket offer

“What’s a good CPA on Facebook ads?” is one of the questions I get asked a lot, and people always want a number.

I do have one. I’ve posted it publicly: if it costs more than $25 to get a cold traffic sale, and you’re selling low ticket, either the offer isn’t good enough or the market is too small.

But that’s a gut check, not a target. I’ve happily run my own front end above it. So here’s how I actually work out what’s healthy for a specific offer. It’s mostly spreadsheet work, and it’ll save you from killing things that are working (and keeping things that aren’t).

Step one: find what a buyer is really worth on day one

The front-end price is the wrong number to judge against. Almost nobody running a low-ticket funnel makes their money on the front end alone. Order bumps and one-time offers are where the economics come from.

For a rough sense of scale, these are the numbers I’ve taught people to aim for on a low-ticket funnel. The front end converting somewhere around 5 to 8 percent of visitors, up to 15. The first upsell taking 15 to 20 percent of buyers. A second upsell or a downsell around 10.

Now some example math, so you can see how much this moves. Say your front end is $27. You add a $37 bump that 20 percent of buyers take, and a $97 upsell that 15 percent take. Your average buyer is now worth about $49 on day one, not $27. That’s nearly double the money you can spend to acquire them.

Use your real numbers, though, not mine. Pull the last thirty days of orders from your payment processor (not Ads Manager) and divide total revenue by the number of front-end buyers.

While you’re there, make sure the purchase value your pixel reports matches what people actually paid. If it’s wrong, every ROAS number in your account is wrong with it.

Step two: add only the backend you can count on

If front-end buyers reliably go on to buy something bigger, some of that can go into the math. The word doing the work there is reliably.

Breakeven, roughly, is that average order value plus the backend you can actually count on, minus product costs and fees. That’s the line between making money and losing it.

Be conservative. A backend you hope for is not a backend. If you’ve never tracked how many front-end buyers ascend, leave it out for now and treat whatever shows up later as a bonus.

Step three: write down two numbers, not one

To judge a good CPA for Facebook ads, I always want a target CPA and a breakeven CPA. Target is the ambition. Breakeven is survival.

That gives you three zones. Under target is winning. Between target and breakeven isn’t winning yet, but you’re not losing either, and that’s where a lot of good accounts live while you work on them. Over breakeven is the only zone that’s actually costing you money.

For a lot of the offers I run, for myself and for clients, we’ll spend hundreds or even thousands a day and either break even or make a little profit on the front end alone. Then by the time buyers hit the upsells, we’re at a four to five X ROAS within a seven-day window. A front end that “only” breaks even can be a very healthy front end.

Here’s my own example, above my own gut-check number. When I launched a low-ticket Facebook ads training of mine on cold traffic, I posted that $43 was a great CPA in the business and marketing niche. And a bit later, with just the front end plus order bumps, it was sitting at 2X ROAS on cold traffic ads. That’s one of the toughest niches out there, and the bumps carried it.

Step four: judge the blended number over enough time

In Ads Manager, cost per result is your amount spent divided by the number of results, and a “result” is whatever you told Meta to optimize for. So on a low-ticket funnel, optimize for purchases. If you optimize for leads, your cost per result is a cost per lead, and it tells you very little about whether the funnel makes money.

Then compare your blended CPA against those two numbers. That’s total spend divided by total purchases across the account, not the CPA on one ad. The way I see it, Meta sequences your ads (people often see several before they buy), so one ad’s number on its own will mislead you.

And give it time. Under two full days is too early to judge anything. I’ve written more about how long to leave an ad set before you judge it, and it’s worth the read before you pull anything.

Step five: check the low end too

This one surprises people. There is such a thing as too low of a CPA.

If I see a CPA on my Facebook ads that looks too good, with a really low CPM and a really high click-through rate, I’m going to go check the backend. Are these real buyers? Do they refund? Do they ever buy again? If you get the wrong kind of people in your business, it’s a curse, not a blessing. CPA, AOV and quality of customer come before everything else.

Same goes for conversion rates people quote online. There are folks who think you should get a 30 to 40 percent conversion rate on ice-cold ads. Whenever I hear that, I figure they don’t know what they’re talking about.

When the $25 rule is right, and when it isn’t

The gut check is most useful when you have no bumps, no upsells and no backend yet, and the front end has to pay for itself. In that situation, a cold sale costing well over $25 usually means the offer isn’t strong enough, or you’re fishing in a pond. (A quick way to check: look at the biggest Facebook groups in your space. If they top out around 15 or 20 thousand people, that’s a pond, not a lake.)

It’s least useful in tough markets like business and marketing, and on funnels with strong bumps and a proven backend. There, a higher CPA can be perfectly healthy. If you’re building that backend now, make sure the front end doesn’t cannibalize what comes after it.

What to do this week

Pull thirty days of real orders from your processor. Work out your average order value including bumps and upsells, add only the backend you can prove, subtract costs, and write your breakeven CPA and your target CPA on a sticky note next to your monitor. Then compare your blended account CPA against both, and make your next decision from that, not from one ad’s number.

What is a good cost per purchase for a low-ticket offer on Facebook?

How do I calculate breakeven CPA?

Can a CPA be too low?