Your ads aren't the problem. The product page you send them to is.
Cold traffic lands on a PDP built for people who already know you, and bounces. We build the page that does the selling first. Live in 48 hours.
Cold traffic lands on a PDP built for people who already know you, and bounces. We build the page that does the selling first. Live in 48 hours.
The algorithm did its job. It found people who want what you sell, and it billed you for every one of them. Then it handed them to a page that opens with a bottle and a price, written for somebody who already knows your brand and is only picking a size. A stranger who clicked for a story does not need a spec sheet. They need the argument the ad started to get finished.
What are you putting into Meta a month?
That is $10,000 a month of cold traffic. Every dollar of it currently lands on a product page, because there is nothing else for it to land on.
You will spend this before tonight. Every visitor it buys opens on a bottle and a price.
You will spend this by Sunday, on people who needed one more paragraph before they were ready to buy.
You will spend all of this whether the page changes or not. Today it lands on a page built to take an order, not to win an argument.
Nothing in that list waits for you. The ad is running while you read this, and the only part of it you can still change is what the click lands on.
These are your own numbers divided up, not a projection. The ones that matter are in your ad account, and they are the first thing we ask to look at.
See if we can fix it →The first one is client work and it stays unnamed, the way yours would. The other three are our own brands. We build for ourselves as well, which is where most of this got worked out before anybody paid us for it.
Agencies that sell pages will show you a blurred screenshot of an ad account and ask you to take their word for the rest. Odd, for a business whose entire product is a page. So here is six months of real client work with the counts in it, then two of our pages and the line each one opens with.
Six months across Klaviyo, Postscript and Shopify. Almost none of it was a copy problem. Read what the counting turned up.
No slides. The page, and why every block on it is there.
Supplements, absorption. Opens on a shelf of things that did not work, then makes the form of the ingredient the whole argument.
See this one on a call ›
GLP-1, aftercare. Written for a market with money and no direct-response playbook yet. Sells the transition, never the drug.
See this one on a call ›The two pages above are concept builds. Nobody hired us to make them, and the publications and products in them are invented. We built them from scratch so you could judge the craft without us claiming a dollar of anyone's revenue. The case study above is the real thing, and every number in it is one we counted ourselves.
Most shops sell you a page and leave. The order matters more than the deliverable, and this is the order that works. Each step is its own build, delivered in 48 hours, and you stop whenever you have what you need.
Built from your best winning ad, because that ad already proved which promise your market responds to. One problem, argued properly, on a domain you own. Nothing else changes while we read the result.
Once one problem converts, the same buyer has three more. Same research, new lead, new mechanism, new page. This is where most of the cheap scale is hiding.
Ads built backwards from the winning page, so the promise in the feed is the promise on the landing. Most CPA damage is a mismatch between those two, not a bad hook.
What it costs depends on what you are already spending, because a page for a brand doing $8,000 a month is not the same job as one doing $80,000. Four questions sorts out which you are.
Send it 2,000 visitors. If it has not beaten the product page you run now, we rebuild it on a different angle at no charge. You are buying an outcome, not an afternoon of our time.
No lock-in and no minimum term. The page, the copy and the domain are yours from the day it ships, whether or not you keep us on after.
Four questions, then a fifteen minute call if it looks like a fit, and we quote on that call. Already know you want one? and skip all of it.
Most of the lift on a winning page shows up in the months after launch, and most brands never go back and take it.
Cancel any time, in one email. No minimum term and no notice period. If a month goes by where we did not earn it, cancel, and you keep the page and the domain exactly as they are.
Because there is no overhead to carry and the build is priced as a build, not as a five-figure monthly retainer. The shops charging that are selling to brands spending fifty thousand a month on Meta, and they are worth it at that size. We built this for the tier under that, which those shops turn away.
No. The page is deployed on a separate domain, which is how these are supposed to run anyway. Your theme, your apps and your checkout are never touched. The page links into your existing product or cart URL.
We do not run your account. Your media buyer keeps that seat. But we do build the creative, and the order matters more than most people expect.
The page gets built first, then the creative gets built off the page. Almost everyone does this backwards. They shoot an ad, it half works, and then they go looking for a page to put behind it, so the page spends its whole life apologising for a promise it did not make.
Build it the other way and the ad only has one job, which is to start the exact argument the page is about to finish. Same claim, same proof, same words in the first line. Nothing to re-explain after the click, and nothing for the visitor to reconcile. That is most of what people are actually buying when they buy a pre-sell page.
It is also why we ask for your best performing ad before we build. Not to copy it, but because it is the one piece of evidence you already own about which promise your market responds to. The page picks that promise up, and then the new creative gets cut to match the page.
We rebuild it on a different angle, at no charge, once. The first lead is a hypothesis and sometimes it is the wrong one, so that risk sits with us rather than with you.
The bar is defined up front so neither of us has to argue about it later. Send the page 2,000 visitors from the same ad you were already running, and we compare purchases per visitor against the product page over the same window. If the page is behind, we rebuild. Two thousand visitors is not a statistically significant test and we are not going to pretend it is, but it is enough to see which way a page is pointing, and it is a number a brand spending five figures a month can reach in a couple of weeks.
What we will not do is tell you it worked when the numbers say otherwise. If the second angle does not move it either, the honest answer is that the offer or the price is the problem and not the page, and we will tell you that instead of selling you a third one.
Yes, because you are handing us the hard part. Your best performing ad already tells us which promise your market responds to, which is the research most of a week normally goes into. From there it is copy, design and build, and there is no account manager or junior handoff to route it through.
The clock starts when we have the ad and your product details, not when you pay.
Your best performing ad, the product it points at, and access to your reviews or support inbox if you have them. That is the whole list. No calls required, no brand deck, no logins.
That ad already told us which promise your market responds to. Send it over and we will come back with the angle we would build underneath it, before you pay anything.
If you would rather talk it through, answer four questions and pick a time. If you already know you want one, you can .
Or email david@badflows.com if you would rather not fill in a form.