March 9, 2026

Why Your Ads Aren't Working Today

Apple Podcasts podcast player badge
Spotify podcast player badge
Audible podcast player badge
Apple Podcasts podcast player iconSpotify podcast player iconAudible podcast player icon

Ads that were working fine suddenly tank, and advertisers assume Meta changed who they're showing ads to or messed up delivery. But there's usually a simpler explanation. Jon breaks down the five real causes of sudden performance drops, including randomness, competitive variables, website issues, and event delays, and why obsessing over daily results drives you crazy when seven-day averages are what matter.

Here’s why your ads aren’t working today.

We’ve all been there. Our ads were working fine, whether it was for a few days or weeks. And then suddenly, complete garbage. Costs spike. Conversions nearly disappear.

The mistake that many advertisers make is that they assume this change is related to their settings, campaign construction, or how Meta is delivering their ads. They think that Meta suddenly changed who they’re showing their ads to, and that messed everything up.

While we can’t rule that out completely, it’s also outside of our control and impossible to prove. More often than not, there’s a much more reasonable explanation for your sudden drop in performance.

The first is randomness.

This may actually be the most common cause of sudden drops in performance. We assume sudden positive spikes in performance represent reality while drops reflect that something is wrong.

Let’s assume you’re getting 10 conversions per day. This is the type of scenario that will lead to the wildest swings in performance.

Meta will show your ads to people most likely to convert, and that will hopefully lead to consistent results. But Meta can’t force people to convert, and that’s where the randomness comes in.

Maybe the pool of people who saw or clicked your ads today just weren’t ready. You cannot ignore randomness because people aren’t robots.

Second, did you make any recent changes?

The changes you make can impact other things you have going on. It could be a new campaign, a new ad set, or a new ad.

Even if you didn’t suddenly enter learning, those changes could impact results. A new campaign or ad set may lead to auction overlap.

If you created a new ad, people are seeing that new ad for the first time, which will restart the typical customer journey. People often need to see and click on ads multiple times before converting.

So don’t ignore the impact of your changes.

Third, there are competitive variables outside of your control.

You could be running at a stable cost per conversion and then suddenly it spikes. This may not even be related to the effectiveness of your ads or conversion rate.

Don’t forget that this is an ad auction. Maybe competition suddenly increased due to seasonality, which spiked your CPM. And if CPM spikes, it spills down to other costs unless the conversion rate improves.

Fourth, it could be your website.

Maybe you’re having performance issues on your side that are impacting conversions. It could be a slow load time or links breaking.

These are easy to miss if your website performance is normally good, but you had issues recently that you may not have known about. Or maybe someone made a change to text or creative you weren’t aware of.

And fifth, there could be event delays.

Reporting isn’t immediately reflected in Ads Manager. There could be delays on Meta’s side, or maybe there’s a delay related to something you control.

It could even be a delay from a third-party tool that sends the events or from your CRM itself.

It’s not at all uncommon to have event delays, but they will clear up.

The bottom line is this: You will drive yourself crazy obsessing over day-to-day results.

Especially if your’e dealing with low volume of results, regardless of your budget, expect inconsistency. People aren’t robots. Meta can’t force people to convert.

If everything had been working fine and you’ve confirmed everything is working on your end, things should come back around.

I am not immune from this obsession with the Today filter, but I’ve also learned this valuable lesson.

I had an ad set that wasn’t performing recently, even though everything had been running fine. I stuck it out through a week of below average results, and then everything turned around.

So here’s the bottom of the glass.

Your results aren’t good today? Big deal, try not to worry about it.

You will have downturns in performance just as you’ll have positive spikes. Neither represents the average performance, which is what matters most.

As hard as it might be, ignore results from the current day. Or if you can’t ignore them, only use these results for monitoring in case something is wrong.

Have patience.

Do not shut ads or ad sets down that were previously working because of a bad day or two.

Evaluate performance based on 7-day windows because averages are more important and predictable than spikes.

A good advertiser is a patient advertiser who understands the law of averages and doesn’t overreact.