If you’d asked me a year ago what to do when frequency hits 5+, I wouldn’t have hesitated.
Five?
Alert mode.
Seven or eight?
Danger zone. Shut that shit off immediately.
That was the rule.
And honestly, it came from a good place. You’ve probably felt it too. That creeping irritation when you see the same ad for the tenth time. You start thinking, “Surely they can’t think this is helping.”
Now Meta has rolled out Frequency Control in some ad accounts, and I’ve finally seen it live.
You can now literally tick a box and tell the algorithm to calm down.
The question is: should you?
Let’s break it down properly.
The short version
Frequency isn’t the villain. Revenue is the referee.
- Meta’s frequency control caps impressions per person. Powerful, and easy to misuse.
- High frequency plus falling ROAS is fatigue. High frequency plus 21X ROAS is profitable repetition.
- Context decides everything. Cold and broad burns; warm retargeting with proven creative compounds.
- Cap too hard and you choke delivery: higher CPMs, slower learning, fewer conversions.
- For most performance campaigns, fix frequency with better creative and audiences, not a hard cap.
What Meta’s frequency control actually does
You tick “Set a frequency for your ad delivery.”
Then you choose Target or Cap.
Example: Cap = 2 impressions every 7 days.
Meta then attempts to keep each individual under that impression limit within the selected window.
For the first time in performance campaigns, you have a hard lever to limit repetition. Not just monitor it. Control it.
That sounds powerful.
It can also be dangerous.
If you test it, start loose. Targets before hard caps. And never set it without watching delivery metrics closely.
Because this is where things get interesting.
When frequency control actually makes sense
This tool shines in small, tight audiences where repetition becomes excessive fast.
If you’re running ads to a niche pool, Meta will recycle impressions aggressively. Without guardrails, people can see the same ad 15 to 20 times.
That’s not brand building. That’s harassment.
How to use it
- Small or niche audiences
- Awareness or consideration campaigns where the goal is “stay present, don’t annoy”
- Situations where you know from data that performance drops after X impressions
This is where restraint protects perception.
And perception matters.
The St. George dinosaur lesson
A perfect example of ad fatigue?
St. George Bank and that dinosaur YouTube ad.
I saw that thing every single day. Multiple times a day. For what felt like months.
At first it was fine.
Then repetitive.
Then irritating.
I genuinely swore I would never bank with them purely because of how overexposed I felt.
That’s what happens when frequency becomes saturation. Consumer perception suffers.
So yes, high frequency can absolutely backfire.
But here’s where my opinion changed.
My unpopular opinion on frequency
Frequency alone is not the villain.
Revenue is the referee.
A year ago, I would have shut anything down once it crossed 5 to 7.
Today? Completely different story.
Example one: wholesale supplier
Since just before the Meta Andromeda update, this well-known wholesale supplier has consistently sat between 5 and 7 frequency.
Old me would have panicked.
Instead, I looked at performance.
- Average ROAS over four months: 21X
- Daily purchases increasing
- Stable CPA
Why haven’t those creatives been turned off?
Because they keep generating more purchases and higher returns.
If ROAS dropped below 16X, they’d be gone instantly.
But why shut off something that’s working absurdly well?
High frequency plus declining results equals fatigue.
High frequency plus 21X ROAS equals profitable repetition.
Big difference.
Example two: the manufacturing company and the “psychedelic sheep”
This one really messes with people’s heads.
Niche manufacturing company. Smaller target market than the wholesale supplier. Limited pool. Not endless scale.
When I first took them on, their previous agency already had ads running.
Now here’s something important.
When you take over an account, you don’t shut everything off just to feel powerful. If something is working reasonably well, you don’t ego-kill it.
So I left one campaign on.
Killed the rest.
Built new ones properly.
That one campaign?
It had one creative.
One.
No variations. No testing matrix. No clever framework.
Just a single static image.
I call it the psychedelic sheep.
It’s obscure. Slightly off-brand. Visually jarring. Completely scroll-stopping.
The kind of creative that makes you pause and think, “What did I just see?”
Now here’s the part that makes frequency purists uncomfortable.
We’re 1.5 years in.
That creative is still live.
Frequency? 16+.
CTR? 0.3%.
And yet…
It consistently generates:
- The highest ROAS
- The lowest cost per purchase
Read that again.
A 16+ frequency.
A “low” CTR.
Best profitability in the account.
Why haven’t I turned it off?
Because it keeps working.
The day it stops generating profitable purchases, it’s gone. No hesitation.
But until then, I’m not shutting down a machine that keeps printing because a dashboard metric looks scary.
Important disclaimers so this doesn’t get misapplied
Before someone screenshots this and says, “Frequency doesn’t matter,” let’s slow down.
First, that psychedelic sheep ad sits in a retargeting campaign.
Retargeting is different. These people already know the brand. They’ve interacted. Repetition reinforces intent.
Second, the other campaigns in the account serve different purposes.
The top campaign? Pure creative testing.
Constant new variations. Controlled experimentation.
So what you’re seeing isn’t neglect.
It’s structure.
Cold traffic gets tested.
Warm traffic gets reinforced.
Retargeting gets reminded.
And sometimes reminded again. And again.
The shift you need to make
Here’s the mindset change.
Frequency is not a kill switch.
It’s context.
The same frequency number means opposite things depending on what sits behind it:
| High frequency burns you | High frequency compounds | |
|---|---|---|
| Audience | Cold acquisition | High-intent retargeting |
| Creative | Weak or untested | Strong, proven, scroll-stopping |
| Targeting | Broad | Tight and relevant |
| The result | Fatigue and wasted spend | Profitable repetition |
| Your move | Cap it or refresh it | Let it run |
The mistake is treating all campaigns the same.
The downsides no one warns you about
If you cap too aggressively, delivery can suffer.
Meta’s algorithm thrives on flexibility. When you restrict it too tightly, you restrict optimisation.
What you’ll see
- Higher CPMs
- Slower learning
- Fewer conversions
If you handcuff delivery, you’re fighting the system designed to find buyers.
And in performance campaigns, that’s usually a losing strategy.
Would I use strict caps for performance campaigns?
Mostly no.
Instead, I would:
- Watch frequency in reporting
- Refresh creatives before fatigue sets in
- Expand audiences
- Optimise offers
I don’t want to block Meta from showing a strong ad one more time to someone who is this close to buying.
Sometimes impression six is the one that closes.
If backend POAS is strong, I’m not pulling the plug because a metric crossed an arbitrary line.
The disclaimer you actually need
This is not permission to ignore frequency.
If:
- Purchases stall
- ROAS tanks
- Negative feedback climbs
- CPMs spike
Then yes, fatigue is likely creeping in.
Frequency should trigger investigation, not automatic shutdown.
Context is everything.
So where would I test frequency control?
- High-frequency remarketing in very small pools
- Short promotions in tiny markets
- Brand campaigns focused on reach and sentiment
And always with backend POAS in mind.
If profit drops when you remove the cap, you have your answer.
The bottom line
Frequency control is a good feature. Nice to have in the toolbox.
But for most performance setups, you’ll fix frequency issues with better creative and smarter audience strategy, not by hard-capping the system.
Control the inputs.
Let the algorithm work.
Unpopular opinion? Maybe.
Profitable? Often.
And at the end of the day, that’s the metric that actually matters.
Meta ads, managed properly
Frequency looking scary? Let’s read the real numbers.
We manage Meta ad accounts where the dashboard metric and the profit often tell different stories. Bring yours, and we’ll tell you which one to trust.
Frequently asked questions
What is Meta’s frequency control?
It’s a setting that lets you limit how often one person sees your ad over a chosen window. You pick Target (aim for a frequency) or Cap (a hard limit), for example two impressions every seven days, and Meta tries to keep each person under it.
Is a high ad frequency bad?
Not automatically. High frequency with falling ROAS and rising negative feedback is fatigue. High frequency on a warm retargeting audience with proven creative can be profitable repetition. The number only means something next to your revenue.
What frequency is too high on Meta ads?
There’s no universal cut-off. Judge it alongside ROAS, CPA, CPM and negative feedback, not on its own. A frequency of 16 can be your most profitable ad in a retargeting campaign, or a warning sign on cold traffic.
Should I use Meta’s frequency cap on performance campaigns?
Mostly no. Capping too tightly restricts the algorithm and tends to raise CPMs and slow learning. Fix frequency with better creative and audience strategy instead, and save hard caps for tiny pools, short promos or brand campaigns.
How do I fix ad fatigue?
Refresh your creative before it tires, expand or tighten your audience, and sharpen the offer. Treat a climbing frequency as a trigger to investigate, and dig in properly when your ads stop converting.