In shortOne account that returned 8.9 times over nearly five years, one of our own that we killed after 348 clicks and zero enquiries, and the arithmetic that tells you which one you would be.
Google Ads is worth it when the profit on one job comfortably exceeds what one enquiry costs you to buy. That is the whole test. It is arithmetic rather than opinion, and it is why the honest answer to this question is "it depends", followed immediately by the numbers that make it depend.
Rather than argue the general case, here are two real accounts. One has returned 8.9 times over nearly five years. The other was ours, and we killed it.
The account that worked
Procon 24/7 is a Yorkshire concrete supplier who arrived with a Google Maps pin and no website. We have run their site, SEO and paid search since 2022.
The numbers, reconciled weekly against their own invoice ledger rather than against platform-reported conversions: £1.34M of tracked revenue on £150,443 of ad spend, a blended return of 8.9 times, 2,055 new customers and 9,550 cubic metres of concrete sold. Nearly five years, still running.
Two details matter more than the headline.
The first is the reconciliation. Platform-reported conversions flatter every account, because Google counts a conversion it thinks it caused. Procon's figure is matched to invoices they actually raised. When somebody quotes you a ROAS, that distinction is the first thing to ask about.
The second is that this return held through a collapse in demand. Category search volume for their product fell 47 per cent year on year into 2026. The account had to hold its return through that decline rather than in spite of it, which is a far better test of whether paid search works than a good quarter in a rising market.
Why does it work there? Concrete is a high-value considered purchase bought by someone actively searching for a supplier today. High job value, clear intent, local delivery radius. That is the profile paid search was built for.
The account we killed
Now the other direction, with our own money.
Lucent ran paid social through 2026. Lifetime across the whole account: £127.20 of spend, 28,356 impressions, 348 clicks and zero enquiries. Not a poor return, an absent one. We killed the channel on 12 August 2026 and moved the budget to search.
That was Meta rather than Google Ads, and the distinction is the lesson rather than a get-out. Paid social interrupts people who were not looking for you. Paid search answers people who are already looking. At a small budget that difference decides everything: 348 interruptions produced nothing, whereas the same money spent against somebody typing "concrete supplier near me" is a conversation with a buyer.
We publish that number because "is it worth it" is a question about risk, and an agency that only ever shows you the winning account is not answering it. We also kept the failed campaigns in the account rather than deleting them, renamed as killed, because the negative evidence is the reason not to restart the channel without changing something first.
The two-minute test
Do this before you spend anything.
- Find the cost per click for your main search term. Our guide to what Google Ads cost in the UK lists real current figures by trade, from £3.14 for an aesthetics clinic to £65.48 for personal injury law.
- Divide your monthly budget by that click price. That is your clicks.
- Take five per cent of the clicks. That is a realistic enquiry count for a decent landing page.
- Divide the budget by the enquiries. That is your cost per enquiry.
- Compare it to your profit on one job, and to how many of those enquiries actually become jobs.
A roofer on a £750 budget at £5.38 a click gets roughly 139 clicks, about seven enquiries, at £107 each. If one in three becomes a £4,000 roof, that is £321 of ad spend per job won. Obvious yes.
Somebody selling a £40 product on the same click price is paying £107 to get an enquiry on a £40 sale. Obvious no, and no amount of clever campaign structure rescues it. The arithmetic is the answer, and it is worth knowing before a salesperson gets to frame it for you.
When it is worth it
Paid search earns its keep when several of these are true at once.
Your job value is high. A £2,000 job absorbs a £100 enquiry cost without flinching. A £40 sale does not. Everything else follows from this.
People search for what you sell with intent to buy now. "Emergency plumber" and "concrete supplier" are somebody with a problem today. That is demand capture, and it is the thing paid search is genuinely excellent at.
Your website already converts. This is the one businesses skip. Ads land on the same pages your organic visitors land on, so if those pages leak you are paying to leak faster. Start with why your website isn't getting enquiries before you buy a single click.
You need leads this month. Ads switch on in days. SEO does not. If the pipeline is empty now, that speed is worth paying for even at a mediocre cost per enquiry, as a bridge while the slower channel builds.
You can hold a budget above the floor for three months. Roughly £750 a month minimum, and the first month is largely the bidding phase learning. A campaign judged on three weeks tells you nothing.
When it is not
Thin margins on low-value sales. Covered above. It is the single most common reason a small account loses money.
The site does not convert. Fix that first; the fix is cheaper than the wasted spend and it makes every other channel work better too. If you suspect this is you, why your website isn't converting is the diagnostic.
Nobody will watch it. Without weekly search-term review and negative keyword work, Google will spend your money on searches that were never going to buy. "Set and forget" is where most of the horror stories come from, and it is why the honest version of this service is a monthly fee for somebody actually looking at it, not a one-off setup.
Tracking is not wired up. This is the fault we find most often in accounts we inherit, and it is the worst one, because an untracked campaign cannot be judged at all. Plenty of businesses have run ads for years reporting clicks to an owner who never knew what an enquiry cost. If you do nothing else, verify the conversion fires against your real enquiry form.
You want an asset rather than a tap. Ads are rented. Stop paying and the traffic stops the same day. A page that ranks keeps arriving. That is not an argument against ads, it is an argument for knowing which one you are buying.
The honest summary
Google Ads is neither a scam nor a growth hack. It is a lever with a known price, and whether pulling it makes you money is mostly determined before the first ad goes live, by your job value, your intent match and whether your website converts.
Procon's account works because all three were true and somebody has watched it weekly for nearly five years. Our own paid social failed because the first one was fine and the intent match was wrong, and we stopped rather than spending more to find out again.
If you want the sums run against your real numbers instead of a general case, the free 30-minute audit covers it, and a fair share of those calls end with us saying not yet. Our own Google Ads management starts with an audit for that reason: we would rather not point paid traffic at a site that does not convert, so if the pages leak we fix that first, or tell you to hold off, before we spend. The management fee is flat and agreed in writing, never a percentage of your spend.
Worth reading next: how much Google Ads cost in the UK for the real click prices by trade, Google Ads vs SEO: which should a small business do first?, what SEO costs in the UK and how long SEO takes if you are weighing the slower channel against this one.
