How to lose money with a 2.5x ROAS

By Eddie Cheng, founder of performance agency Penang Media and co-owner of global footwear brand VIBAe.

Pulled this from my best Twitter post last week. It's the math nobody running paid actually wants to sit with for more than a few seconds.

$80,000 in blended ad spend. $200,000 in revenue.

The kind of month a lot of brands would celebrate in the team Slack.

Now run it for real.

57% of that revenue goes to COGS, fulfillment, and returns. That's $114,000 gone before you touch a single other line item.

Subtract the $80k you just handed to Meta.

You're left with about $4,600.

Not $46,000. Not $460,000.

Four thousand six hundred dollars.

Now layer an agency fee on top of that. Or payment processing. Or the software stack. Or the salary of the person actually running the account.

You're underwater. Or close enough that one bad week drowns you.

This is the part of DTC nobody really wants to talk about at the conference.

Because the whole paid social game everybody's playing quietly assumes your unit economics can absorb aggressive customer acquisition. If your baseline costs are already sitting above 50% before Ads Manager is even open, you're not playing the same game as the brand next to you with a 70% gross margin. You're playing a harder one. With less oxygen.

And it compounds. The brand with better margins gets to test more creatives, lose on more of them, and bid higher in the auction. You don't. Every dollar you spend has to work harder than theirs, because you have less runway on the other side of the sale.

So what do you actually do if you're stuck there?

You stop trying to win the unit-by-unit game on paid social. You start thinking in bulk. B2B distribution. Wholesale accounts. Subscription that stretches LTV over 12 months instead of one. Bundles that push AOV past the breakeven line. Anything that breaks you out of the "one customer, one transaction, one margin calculation" loop.

The other option is the hard one. Fix the unit economics themselves. Raise price. Cut a cost. Renegotiate with the factory. Ditch the free shipping promise that's quietly eating you alive.

I'll tell you what it looks like on our side at VIBAe.

We don't compromise on the shoe. Real Portuguese leather, a proper factory, the kind of construction that actually holds up after a year of wear. Nothing about the product got cheapened to make the margin work.

But we priced it sharply in the market. Sharp enough that customers feel it's fair. Sharp enough that the numbers still breathe when we turn the paid side on. The feedback loops back in, the take rate holds, and the economics let us keep pushing instead of constantly firefighting.

That breathing room is the whole game. It's the difference between a campaign that scales and one that just survives a good week.

Here's what I think. Most brands losing money on paid don't have a creative problem or an agency problem. They have a unit economics problem dressed up as a performance problem. And no amount of better hooks or tighter audiences is going to save a contribution margin that was broken before the ad ever ran.

I'll hedge that, because I only see so much of the picture. Some brands pull things off I wouldn't have bet on. Some are scaling at margins I'd consider unworkable. There's always someone figuring out a version of the game I haven't thought of yet.

Still. If you're staring at reports wondering why a 2.5x ROAS month doesn't translate into anything you can actually spend, the answer is probably sitting one tab over in the P&L.

Not in the ad account.

— Eddie