How Much Should You Spend on Meta Ads to Get Results?

“How much should I spend on Meta Ads?” usually gets answered with a number pulled from nowhere. Five pounds a day to start. Five hundred a month. A thousand to test properly. None of those mean anything without knowing what a result is worth to you and what it is likely to cost.

There is a way to calculate a sensible starting budget, and it takes about ten minutes. This explains it, along with why budgets below a certain level tend to produce nothing regardless of how well the campaign is built.

Start from what a customer is worth

Work out two figures before looking at the ads platform at all.

The first is what a customer is worth to you, ideally over their lifetime rather than on a single transaction. A gym member paying monthly for an average of fourteen months is worth considerably more than one month’s fee. A roofing company doing one job per customer has a different figure entirely.

The second is what proportion of enquiries become customers. If one in four of the enquiries you receive turns into work, then four enquiries are needed per customer, and the amount you can afford per enquiry is a quarter of what you can afford per customer.

From those two numbers you get the figure that matters: the maximum you can pay for an enquiry and still be ahead. Everything else follows from it.

Then work out what the system needs to function

Meta’s delivery system learns from conversion events. While it has too few, the ad set sits in the learning phase, where delivery is unstable and the cost per result is not representative of anything. Meta documents the threshold at roughly fifty optimisation events in a week for an ad set to exit learning.

That number creates a practical floor. If your cost per enquiry is likely to be around twenty pounds, fifty events a week implies about a thousand pounds a week to get the system properly out of learning, which is beyond most small business budgets.

This is where realism matters. Few small businesses can fund the ideal. What they can do is avoid the worst version, which is spreading a small budget across several ad sets so that none of them accumulates any data at all. Concentrating a modest budget into one ad set, optimising for a cheaper event higher up the funnel, and accepting that delivery will be less stable is a far better use of limited money.

A workable starting budget

For most small businesses, the practical approach is to fund a test that can produce a statistically meaningful number of results within a defined period, rather than a trickle that runs indefinitely.

Estimate your likely cost per result, which for lead generation in many service markets sits somewhere between ten and fifty pounds depending on sector and competition. Decide how many results you would need to see before you could judge the outcome, which is rarely fewer than twenty to thirty. Multiply the two, and that is the size of your test.

Running that over four to six weeks gives a daily budget. If the resulting number is uncomfortable, the honest options are to run a shorter, more concentrated test, to optimise for a cheaper event, or to accept that this channel may not be affordable at your current economics yet.

Why very small budgets usually disappoint

At five or ten pounds a day, three things happen. The ad set never exits the learning phase, so costs stay volatile. The sample is too small to tell a bad campaign from an unlucky fortnight. And the advertiser, seeing nothing after three weeks, changes something, which resets learning and restarts the cycle.

That pattern is responsible for most of the “Meta Ads do not work for my business” conclusions. The campaign was never given the conditions it needed to produce a readable result.

If a very small budget is all that is available, concentrate it: one campaign, one ad set, a handful of creatives, broad targeting, and no edits for the duration. That at least produces a clean answer.

Splitting budget between cold and warm audiences

Retargeting consistently costs less per result than cold prospecting, because the audience already knows you. It is also limited by how many people visit your site, so it cannot carry a campaign on its own.

A common and sensible split is the majority of the budget on cold acquisition, with a smaller portion on retargeting site visitors and engaged users. If your site traffic is low, retargeting will barely spend, which is itself a signal that the first job is getting more people to the site rather than advertising to the few who arrive.

What to spend money on besides the ads

Two investments routinely return more than an equivalent increase in ad budget.

The first is creative. Performance is now driven far more by what the ad says and shows than by audience settings, and a steady supply of new creative is the main defence against the fatigue that kills working campaigns. Budgeting for ongoing creative rather than one batch is the difference between a campaign that lasts three months and one that lasts three weeks.

The second is the landing page. Doubling the conversion rate of the page halves the cost per enquiry across every pound spent afterwards, which is a better return than any bidding adjustment. The common failures are covered in why landing pages do not convert.

When to increase spend

Raise the budget when the cost per result is comfortably below what a result is worth, and when that has held steady over several weeks rather than a few good days.

Increase gradually. Large sudden budget changes push an ad set back into learning and frequently make performance worse before it recovers. Increments of twenty to thirty percent, with time between them to stabilise, keep the campaign’s performance intact.

Also watch frequency. When the same people are seeing the ad repeatedly and results decline, the constraint is creative rather than budget, and more money simply buys more repetition.

When to stop

If the cost per qualified enquiry is consistently above what an enquiry is worth, after the tracking has been verified, the objective checked and the landing page addressed, the sensible answer is to stop rather than to spend more. Some businesses are better served by search advertising, where the customer arrives already looking, as covered in Meta Ads versus Google Ads.

Deciding that in advance, with a number attached, is what separates a test from an open-ended expense. The timing side is covered in how long to run an ad before judging it, and the wider mechanics in our complete guide to Meta Ads.

Working out your number

If you know what a customer is worth and roughly what proportion of enquiries convert, the rest is arithmetic. If you do not know those figures, that is the first thing worth fixing, because without them no advertising decision can be judged.

Tell us what you sell, what a customer is worth and what you are spending now. We will tell you whether the budget is capable of producing a readable result, and where it would be better placed. Our Meta Ads management starts with that arithmetic rather than with a campaign build.