Andrew Purdum

Offers

Why a proven offer stops working after you improve it

Flat editorial illustration: an arch of black blocks cracking at the base after a red-orange block is added on top.

Short answerWhen a proven offer stopped working right after you improved it, look at what the “improvement” added: more content, more process, more explanation, a new price, a reworked page. Any of those can raise how much the buyer has to believe before saying yes, break the order the market needed to hear things in, or spoil the data your ads built on the old version. Change one thing at a time, and protect what was already working.

It’s one of the most frustrating things in this business. You have an offer that sells. You know it could be better, so you make it better. More value, a cleaner page, a sharper price. And then it stops selling.

When the offer stopped working, the natural conclusion is that the market changed, or the ads died, or it was luck all along. Sometimes that’s true. But a lot of the time, the improvement is what did it. Not because better is bad, but because the “better” version quietly broke something the old version was doing that you couldn’t see.

Here are the ones I run into most.

You raised the belief threshold

The most common improvement is adding. More modules, more steps, a fuller explanation of the method, the whole process laid out so people can see how much they’re getting.

On a front end, that often backfires. Somebody coming in cold isn’t saying yes. They’re saying maybe. The moment they read something they’re not sure about, that “I’m not sure” is the enemy of the sale.

I had a client who wanted to put his multi-part method, several steps and a stack of math behind it, right on his cold-traffic front end. It’s a great argument. It’s just a backend argument. I told him that on the front end it would destroy the offer, because it assumed his buyers already held a whole set of beliefs they didn’t have yet. I call that a high belief threshold offer: an offer that needs a lot of very specific beliefs in place before anyone can say yes. That kind of thing works in an upsell or a backend workshop, after they’re already in the door.

Same goes for talking more about what’s inside. The less you talk about the product itself, typically the higher it converts, especially with a sophisticated audience. People buy the outcome and the dream. Every paragraph you add describing modules is a paragraph they’re not spending imagining the result. The same logic is behind why bonuses can make an offer weaker.

You broke the order it was selling in

Every market has a sequence it needs to hear things in before it buys. For a jaded crowd, that usually means proof early, then why it works, then the offer.

When it comes to copy, sequencing matters more than quality. As long as the sequence is right, the copy can be pretty mediocre and still sell. So a rewrite that makes every sentence nicer but moves the proof to the bottom, or opens with the method instead of the promise, can lose sales even though it reads better.

This is the sneaky one, because the new version genuinely looks like an upgrade. Nobody looks at a more polished page and thinks, that’s why the offer stopped working. If you rewrote your page recently, lay the old and new versions side by side and compare the order of the argument, not the quality of the writing. I wrote more about that ordering in sequencing a page so belief builds before the price.

You spoiled the data your ads built

This one lives in the ad account, so it catches people off guard.

A Post ID is basically the permanent link for one ad (same copy, same image or video, same headline), and it carries a lot of data. When a Post ID gets a lot of results in a short time, those results tend to continue, because Facebook hones in on who took action from that post. But if results suddenly drop off after you change something in the funnel, the sales page or the price, that Post ID can build bad data. From what I’ve seen, Facebook starts treating it as an ad that isn’t getting results, and costs go up.

Someone who’d bought one of my trainings ran into exactly this. He raised the price on his offer (which can be fine), but kept the same ads running with the same Post IDs, and results fell. He changed his ads enough to get new Post IDs, and his sales went up that Friday and more than doubled over the weekend.

So if you improve the offer, consider launching fresh ads for it rather than asking old Post IDs to carry a changed offer.

You moved it out of its price pocket or past the first obstacle

Two more quick ones.

Price. Every market has groups that will only pay certain prices, and changing the price moves you into a different group of buyers. Raising it can be exactly right. It can also take you out of the pocket that was buying. The reverse is true too, and there’s a piece on why lowering the price often lowers conversions.

Scope. A good front end gets someone past the first obstacle, not the whole way. When you improve it by making it cover everything, the promise gets bigger, and a bigger promise at a small price is harder to believe. It can also start eating the backend you were planning to sell them.

When the improvement isn’t why your offer stopped working

I want to be fair here, because sometimes you did nothing wrong.

Offers wear out on their own. If the market’s been hearing the same message for months, the pocket of buyers it reached may be exhausted, and your improvement just happened to land the same week. The tell is timing. If results had already been sliding for weeks before you changed anything, the change probably isn’t the cause. If they fell right after it, it probably is.

And improvements do work, a lot of the time. I’ve seen great results just from refreshing a sales page. The point isn’t to never touch a winning offer. It’s to know which parts were doing the work before you touch them.

When I audit an offer, one section of the output is always “what’s already working, don’t touch.” It’s usually the shortest section, and it’s the one people ignore.

What to do this week

If a proven offer stopped working after you changed it, write down every change you made in the last 60 days, with dates, next to your daily sales. Find the change that sits right before the drop. Put that one piece back the way it was, launch fresh ads so you’re not relying on old Post IDs, and give it a week. Only then decide whether the rest of your improvements stay.

Why did my sales drop after I updated my sales page?

Usually because the new version changed the order of the argument, added detail that raises doubt, or made the promise bigger and harder to believe. It can also spoil the data on ads that were built around the old page. Compare the old and new pages side by side for sequence, not writing quality.

Should I add more content to my low ticket offer?

Usually not. A front-end offer only needs to get someone past their first obstacle. More content and more process on the page tend to create doubt in cold buyers. Save the fuller method for an upsell or backend, where people already trust you enough to take it in.

How do I improve an offer without breaking it?

Change one thing at a time and note the date. Know what’s already working and leave it alone. When you change the offer, launch fresh ads for it instead of relying on old ones. Give each change about a week before judging, and check whether results were already sliding before you touched anything.

P.S. Keep a simple change log for your funnel, one line per change with the date. The day an offer drops, it’ll save you hours of guessing.

Andrew Purdum

Buying traffic since 2014, north of $10M in managed ad spend. Writes about what actually moved the numbers, including the expensive mistakes. Work with me →

More on offers

All 13 →

Work with Andrew

Most accounts don't need more traffic. They need someone to find the leak.

If you're spending real money on ads and can't tell which part is failing — the creative, the offer, or the page — that's a diagnosable problem, and it's the one I'm best at.