“How much should we spend on ads?” has no benchmark answer, because the correct number depends entirely on your click costs and what a customer is worth to you. But it does have a calculation, and it takes about ten minutes.
Work backwards from a customer
Start with average customer value, ideally lifetime rather than first purchase. Say £2,000. Decide what you are willing to pay to acquire one; for many service businesses, ten to twenty percent of value is comfortable, so £200 to £400.
Now estimate the funnel. If your landing page converts five percent of visitors into enquiries, and you close one in four enquiries, then you need eighty visitors per customer. At a £3 cost per click, that is £240 to acquire a £2,000 customer. Viable.
Run the same arithmetic with a £12 cost per click, which is normal in competitive B2B, and the number becomes £960. Now the channel only works if you improve conversion rate, close rate, or targeting precision. That is a useful thing to know before spending rather than after.
The minimum viable budget
Separate from affordability, there is a learning threshold. You need enough clicks per month to gather data: roughly a hundred, which at a £3 cost per click means £300 and at £12 means £1,200. Below that, campaigns cannot be optimised, they can only be observed.
For most UK service businesses this lands between £750 and £1,500 per month in ad spend, plus management. Below it, results still happen, they just arrive slowly enough to be frustrating for everyone.
The mistake that beats both
Splitting a small budget across Google and Meta “to test” and starving both. Pick the platform your demand type suits, win it, then expand from a position of knowledge. That decision is covered in more detail in the comparison of Google Ads and Meta Ads.
What to expect early
The first month buys data, not efficiency. Month two applies what month one taught. Month three is usually where the numbers start looking like the plan. Budget for three months minimum, or do not start.