The short version: the right number of Facebook ads to run at once isn’t three, five, or seven, it’s whatever your daily budget can feed, and for most small businesses spending under £100 a day that number is one or two ad sets, not the six or eight most people launch out of nerves. Work out your target cost per result first, then let the maths tell you how many ad sets you can afford to test. Everything else is guessing dressed up as strategy.
The question you’re asking
When people ask me how many Facebook ads to run at once, what they really mean is “how do I stop wasting money.” Fair question. But it’s the wrong question, because the number of ads isn’t the lever that wastes your money, thin budgets spread across too many ads is. I’ve watched clients launch eight ad sets on a £40 daily budget because a blog post told them “test, test, test,” and every single one of those ad sets sat in Facebook’s learning phase for three weeks straight, never delivering a stable result, because none of them had enough spend behind them to learn anything.
Meta’s own delivery system needs roughly 50 optimisation events, that’s 50 conversions, add-to-carts, or leads depending on what you’re optimising for, per ad set within a rolling seven day window to exit the learning phase reliably. Below that, the algorithm keeps testing audiences and placements internally, your cost per result bounces around, and you can’t trust any of the data you’re looking at. That single fact changes how you should be answering this question.
Work backward from your budget, not forward from your ideas
Here’s the calculation I use with clients before we touch Ads Manager:
- Step 1: Set your target cost per result. Say you’re selling a £150 product and you’re comfortable paying £15 per purchase to acquire a customer.
- Step 2: Multiply that by 50 to get the weekly spend one ad set needs to hit the learning phase target. £15 x 50 = £750 a week, roughly £107 a day, for one ad set to have a fair shot at exiting learning.
- Step 3: Divide your total daily ad budget by that number. If your total budget is £100 a day, you can fund exactly zero ad sets by this strict standard, which tells you either your CPA target is unrealistic for your current budget, or you need to widen your definition of a result (using add-to-cart instead of purchase, for example, which requires far fewer events to stabilise).
- Step 4: If your budget is £300 a day, you can fund roughly two to three ad sets. Not seven. Not twelve.
This is the uncomfortable bit nobody selling you a “10 ad testing framework” wants to say out loud: most small business budgets can only support one to three well-fed ad sets at a time. Running more than that isn’t testing, it’s diluting, and it makes you feel busy while slowing down the exact learning process you’re trying to speed up.
The client account that taught me this the hard way
I worked with a coaching business a couple of years back who came to me convinced their ads weren’t working because Facebook was “punishing” their account. They had a £35 daily budget spread across nine ad sets testing different audiences, headlines and images, because a course they’d bought told them variety was king. Nine ad sets on £35 a day works out to under £4 per ad set daily. Each one was getting maybe one or two conversions a week if it was lucky, nowhere near the volume needed to stabilise, and their cost per lead swung between £8 and £64 depending on the day.
I cut it down to two ad sets. Same total £35 budget, split roughly £20 and £15. Within eleven days, both had exited the learning phase, one settled at a consistent £11 per lead, the other at £19. We killed the weaker one, moved its budget to the winner, and within a month they had a single ad set converting at £10.50 per lead, spending £35 a day, and finally producing numbers they could plan a sales forecast around. Nothing about the creative changed. Nothing about the offer changed. The only thing that changed was that we stopped asking nine underfed ad sets to do the job of one well-fed one.
How many ads inside each ad set, separately from how many ad sets
There’s a second layer to this question that gets muddled together: how many individual ads sit inside each ad set. Meta’s own recommendation is three to six ads per ad set, and I generally land on three or four in practice. Too few and you’re not giving the algorithm’s dynamic creative optimisation enough raw material to find the winning combination of image, headline and copy. Too many, again, splits the same limited budget across too many variants and none of them get enough impressions to tell you anything.
So the honest structure most accounts should run, especially anyone spending under £150 a day, looks like this:
- One campaign per objective (leads, purchases, traffic, whatever you’re optimising for)
- One to three ad sets per campaign, each targeting a distinct audience worth separating budget for
- Three to five ads per ad set, varying the hook and the visual, not just the button colour
That’s it. That’s the whole account for a business spending under £5,000 a month on ads. If your competitor’s ad account screenshot shows fifteen live ad sets, either they’re spending £2,000 a day, or their account is a mess and they haven’t noticed yet.
What changes as your budget grows
The rules loosen as spend increases, because the maths I showed you above scales with budget. A business spending £500 a day can support four or five well-fed ad sets, testing different audience segments, maybe one cold audience, one lookalike, one retargeting set, one broad targeting test with Advantage+ audience expansion switched on. At £1,000 a day and above, running separate campaigns by funnel stage, top of funnel awareness, middle of funnel retargeting, bottom of funnel cart abandoners, each with their own two or three ad sets, starts to make sense because each layer can finally get enough volume to learn independently.
One thing that trips people up here, and it’s worth checking before you decide anything about ad set count, is whether your budget is even doing what you think it’s doing across placements. If your ads are set to automatic placements, a chunk of that daily spend is being shown on Instagram whether you planned for it or not, which changes how much budget is reaching each platform’s audience and can quietly shrink the effective spend behind any single ad set. I wrote a full breakdown of exactly what happens to your budget when Facebook ads automatically run on Instagram too, and it’s worth reading before you add another ad set assuming your current ones are fully funded on Facebook alone.
Signs you’re running too many ads at once
- Cost per result swings wildly day to day with no clear pattern, sometimes £8, sometimes £40, for the same offer
- Most of your ad sets show “learning” status in Ads Manager for more than seven days straight
- You can’t say, without checking, which ad set is driving your best results this week
- Your total ad spend hasn’t grown but you keep adding new ad sets “to test” instead of scaling what’s already working
That last one is the pattern I see most often with business owners who are, understandably, a bit scared of ad spend. Adding another ad set feels like doing something. Putting more money behind the one ad set that’s already working feels like risk. It’s the opposite. The uncomfortable truth is that most people run too many ads not because their strategy calls for it, but because choosing one winner and backing it fully feels more exposed than spreading the bet across six mediocre options where no single failure feels as personal.
Signs you’re running too few
The flip side matters too. If you’ve got one ad set that’s been live for six weeks, spending the same £20 a day, with no new creative added and performance quietly declining, that’s not discipline, that’s ad fatigue, and it’s just as costly as overtesting. Facebook audiences see the same ad roughly seven to nine times before response rates start dropping noticeably, sometimes faster in a small local audience. If your frequency metric in Ads Manager is climbing past 3 or 4 within a couple of weeks, you need a second ad, or a new audience angle, not because more is always better, but because your current one has run its course.
A good middle ground I use with most accounts: build one, let it run and stabilise for at least a full learning cycle, seven to fourteen days, then decide whether to add a second ad set testing a meaningfully different audience or angle, retirees versus young professionals, cold traffic versus warm retargeting, rather than a near identical twin of the first.
Where organic content fits into this decision
Paid isn’t the only lever, and it shouldn’t carry your whole strategy on its own. If you’re running lean on ad budget, some of what you’d normally test with paid spend can be tested for free first inside organic reach, seeing which hooks and angles get engagement in a Facebook group or on your page before you put money behind them, and refining your hashtags and organic reach tactics to feed your retargeting audiences with warm traffic that costs far less to convert than cold. It won’t replace paid testing entirely, the audiences and behaviour differ, but it’s a useful, free pre-test that tells you which two or three angles are worth turning into funded ad sets rather than guessing at six.
If your account has grown past the point where you can sensibly manage this by feel, more than about £3,000 to £5,000 a month in spend, multiple objectives running at once, seasonal promotions layered on top of always-on campaigns, that’s usually the point where bringing in an AI consultant for small business to help build proper testing frameworks and automated reporting pays for itself within a month or two, because the cost of one badly structured campaign at that spend level is usually higher than the help itself.
My actual rule of thumb, summarised
- Under £50 a day: one ad set, three to four ads, full stop, don’t argue with the algorithm about it
- £50 to £150 a day: one to two ad sets maximum
- £150 to £500 a day: two to four ad sets, ideally split by different audience or funnel stage
- £500 plus a day: four to eight ad sets across two or three campaigns split by funnel stage, each still fed enough to hit roughly 50 weekly conversions
Notice the pattern isn’t about creativity or ambition, it’s arithmetic. The number of ads you can sensibly run at once is a function of the money you’re putting behind each one, nothing more romantic than that.
Frequently asked questions
How many Facebook ad sets should a small business run at once?
Most small businesses spending under £150 a day should run one to two ad sets at a time, each with three to five ads inside it, because Meta’s algorithm needs roughly 50 conversions per ad set within seven days to stabilise, and thinner budgets split across more ad sets rarely reach that number.
Does running more Facebook ads at once give you faster results?
No, it usually slows results down, because splitting a fixed budget across more ad sets means each one gets less spend, takes longer to exit the learning phase, and produces less reliable cost per result data than fewer, better funded ad sets would.
How do I know if I’m running too many Facebook ads at once?
Check Ads Manager for ad sets stuck showing “learning” status for more than seven days, cost per result that swings wildly day to day, and an inability to name your best performing ad set without checking, these all signal your budget is spread too thin across too many active ads.
Should I add more ad sets as my Facebook ad budget grows?
Yes, but proportionally, use the formula of target cost per result multiplied by 50 divided by your daily budget to work out how many ad sets your current spend can fund, then add new ones only once your budget supports them rather than on a fixed schedule.
