Is Google Ads worth it for small businesses?
by Harry Watts · · updated · 7 min read
Ask an agency and it's worth it. Ask someone who lost £3,000 on it and it's a scam. Both answers are autobiographies. "Worth it" isn't a property of Google Ads — it's a property of your numbers, and you can compute it before spending a pound.
The platform itself is not in question. Google carries more than nine in ten UK searches, and an ad there reaches people at the exact moment they're looking for what you sell, the strongest targeting mechanism advertising has ever had. The entire variance in outcomes is on the advertiser's side of the auction: margins, conversion rates, market prices, and attention. So let's do your side properly.
The arithmetic that answers it
Four steps, five minutes, from the cost article, compressed here because it's the whole decision:
- 1. Value a customer. Revenue × margin, including repeat business. An average job worth £900 at 40% margin, where a third of customers return once: call it £480.
- 2. Set your acquisition price. A quarter to a third of that value is the usual band. Say £120.
- 3. Count clicks per customer. At a 5% site conversion rate with half of enquiries closing: 40 clicks per customer.
- 4. Derive your affordable click price. £120 ÷ 40 = £3. Now look up what clicks cost in your market.
If your market's clicks run £2, you have headroom: worth it, with room for error. If they run £8, the arithmetic fails before you start, and no bidding strategy negotiates with it. For e-commerce there's a faster gate: break-even ROAS is 1 ÷ gross margin. A 40% margin means 2.5× is where profit begins, not where success is declared.
The three honest no-cases
Most articles on this question are written by people selling the yes. Here's the no, properly argued:
1. Margins too thin for the market. A £25 product with £6 of margin cannot fund £1.50 clicks at ordinary conversion rates: every sale arrives at a loss, and volume just scales the loss. Without repeat purchases or basket-building to raise the value, this business shouldn't buy clicks; it should fix its margin or its average order first.
2. Lifetime value can't clear a high-CPC market. Some markets price clicks against the biggest winners in them. If established competitors profit from £8 clicks because a customer stays five years, and yours buys once for £60, you're bidding in an auction calibrated to someone else's economics. The fix isn't cleverer ads — it's either building the repeat relationship or conceding the head terms and fishing narrower.
3. The site can't convert. Ads charge full price for every click your page wastes. If the site loads slowly, buries the phone number, or asks for a form nobody finishes, paid traffic just measures the leak expensively. Structure before spend: the page is a prerequisite, not an optimisation.
And the soft fourth: nobody minding it. If there's no weekly half hour for the account and no budget for help, the maths may clear and the account still lose: an unattended account is the most expensive kind.
"I tried it and lost money" — what that usually means
I've audited a lot of accounts behind that sentence. Almost none of them were evidence about the channel. The recurring four:
- The defaults were on. Interest-based locations, Display pre-ticked, auto-apply recommendations: the standard leaks, running from day one.
- The tracking counted the wrong thing, so nobody could see what was true: page views dressed as conversions in both directions.
- Judgement at week two of what needs 60–90 days and ~30 conversions a month before smart bidding settles.
- No arithmetic beforehand, so there was never a number at which it would have counted as working, only a feeling that it wasn't.
The uncomfortable summary: the platform executed their mistakes efficiently. That's not a defence of Google: its defaults profit from beginners, and I've written exactly that. But it means "I lost money" is usually a fixable diagnosis, not a verdict.
What "worth it" actually looks like
Unspectacular. A cost per customer at or under the target you set in step 2, holding as you scale. Early months worse than the long-run rate, because the account is buying answers as well as customers. A weekly half hour of search-term pruning, whoever does it. Growth funded from proven headroom, including, eventually, the second platform.
Anyone promising a specific multiple before they've seen your margins, your market's click prices and your website is describing their marketing, not your prospects.
The verdict
Google Ads is a market, not a bet. Markets don't owe you a profit; they price what you bring to them. Bring an offer with real margin, a page that converts, honest measurement and a number agreed in advance, and it's one of the most controllable, provable ways a small business can grow. Bring none of those and the auction will find you out in weeks, at full price.
Run the four steps before anyone runs a campaign. Including me.
Quick answers
Is it worth it for a UK small business?
When the arithmetic clears: customer value → acquisition price → clicks per customer → affordable CPC vs your market's CPC. Where it clears, yes, provably. Where it doesn't, no optimisation rescues it.
When is it genuinely not worth it?
Margins too thin for the market's click prices; lifetime value too low for a high-CPC auction; a website that can't convert the traffic. Plus the soft case: nobody available to mind it weekly.
Why do so many small businesses lose money on it?
Defaults left on, tracking counting the wrong things, verdicts at week two, and no pre-agreed number for what "working" means. Fixable faults executed efficiently: rarely evidence about the channel.
Can profitability be checked in advance?
The four steps above, or for e-commerce the gate: break-even ROAS = 1 ÷ gross margin. At 40% margin, 2.5× is where profit begins.
What return is realistic?
Cost per customer at or under your pre-set target, stabilising over 60–90 days as the account earns ~30 conversions a month. Specific multiples quoted sight-unseen are marketing.
Want the arithmetic run with your real numbers
That's most of what a free intro call is: your margins, your market's click prices, your site, and a straight answer on whether the sums clear. If they don't, you'll hear "don't spend", and it will have been the cheapest half hour of the project. If they do, the build starts from a number we've already agreed.