When someone new comes onto an account, I want them to ask for the history. What have we run, and what's worth knowing about it?
The people who've been on an account a while tend to stop asking it, because they think they know the answer.
Answering it means walking somebody through every funnel from scratch. That's where you find the ones nobody has touched in a long time. You can't see them from the inside, because you walk past them every week.
It happens to funnels that work.
I've had one that was up for a long stretch, rotating through the same few ads. From what I could see, it had worked. Its numbers didn't give anybody a reason to look, so nobody looked. I thought we could get a lot more out of it. To me, it was low-hanging fruit.
When something stays live for months, it stops feeling like a decision somebody made. It turns into part of the account. Somebody asks how the funnel's doing, the answer is "it's running," everybody nods, and the meeting moves on. After a while, "it's running" starts to sound a lot like "it's working."
Those mean two different things. "It's running" means the ads are delivering and the page loads. "It's working" means it brings back more than it costs.
A funnel that's been up that long can feel like it passed a test. Nothing new ever went up against it.
Marketers talk about ad fatigue like it's an audience problem. People see the same image until they scroll past it, so you give them something new to look at.
That's real. There's another kind, and it's on our side of the screen.
Teams are great at getting assets for a launch. There are new images and new videos, and everybody's leaning in. Once a campaign has been running a while, the new images and the split tests dwindle. Nobody decides that. The thing isn't broken, so it doesn't make anybody's list.
None of that is about anybody being lazy. My guess is that launches get the energy because launches have a deadline. A page that's been live for a while doesn't have one.
The working ones are the ones that bug me. A funnel that's losing money tends to come up in a meeting at some point. A funnel that's doing fine can run for a long time with a lot more in it, and nobody goes looking.
Here's the rhythm I've run on most of my accounts: new images or new videos every couple of weeks, and new images every other day for the ones spending a lot.
The point of that rhythm is that somebody keeps asking whether the thing on the screen is the best thing we've got.
When a new version loses, that's fine. You keep the one that was winning, and you know a little more than you did.
A new version should make a different case. Change the promise in the headline, or the order the page tells the story in. A new picture is fine for an ad. On a landing page, a new picture on the same argument won't teach you much.
For anything that's been up a while, ask two questions.
When did we make the last change to this?
What did that change teach us?
If nobody can answer the second one, I'd start with that page.
It doesn't have to be a new hire. Anybody who'll ask why a page looks the way it does will work.
If you've got a page or an ad you can't remember changing, pull it up and walk somebody through it. You'll hear which ones haven't moved.
After that, pick one and build it a real second version.
Source note: Drawn from a client call, with identifying details generalized. No client numbers are used.
I turn this kind of reading into a working document for one business at a time: who your buyers imitate, what they want beneath the surface, and the words they use when nobody is selling to them. The first bounded read is free. Mimetic Intelligence, built by hand with source-linked evidence and stated confidence.