Search advertising is the quickest way to appear at the top of Google for the things you sell, and the quickest way to pay for people who were never going to buy from you. Both are true at once, which is why the same tool produces one business that swears by it and another that lost a thousand pounds in a fortnight and never went back.
The useful question is not whether Google Ads works. It plainly does, for a great many small businesses. It is whether it works at your budget, in your sector, pointed at your website — and most of that you can work out before you spend anything.
Ads and organic search answer different questions
Ads buy attention today and stop the moment you stop paying. SEO earns it slowly and keeps working after you stop commissioning it. Choosing between them is usually a question about timing rather than quality.
Ads suit you when you need enquiries this month, when you are launching something with no search history behind it, or when the work is seasonal and the season is now. They are also the fastest piece of research available: a few weeks of ad data tells you which searches actually produce customers, and organic search would take far longer to tell you the same thing. Most small businesses end up running both, with ads carrying the load and coming down as the organic work matures.
Do the arithmetic before you open an account
There is no minimum spend on Google Ads. You can start with a few pounds a day, and plenty of guides will encourage you to. The trouble with a very small budget is not that it fails outright — it is that it teaches you nothing. A handful of clicks a week produces no pattern anyone can act on, so you end up guessing with real money.
So work four numbers through first, in your own terms:
- —What a click costs for your services in your area — Google's Keyword Planner gives a rough estimate for your terms and locations before you have spent anything
- —How many website visitors currently become enquiries, which is the number everything else depends on
- —How many enquiries become customers
- —What a customer is worth over the whole relationship, rather than on the first job
What that arithmetic tends to show
Multiply those through and you get an approximate cost per customer. The step people skip is the middle one: ads deliver visitors, not customers, so a site turning one visitor in fifty into an enquiry needs five times the clicks of one managing one in ten.
What usually falls out is a sense of the monthly figure below which there is not enough traffic to learn anything — much higher in legal, cosmetic and some trade markets where a single click can cost several pounds. Occasionally what falls out instead is a cost per customer above what a customer is worth to you. If that is the answer, you have learnt something valuable for twenty minutes of arithmetic, and the right move is not to start.
Three leaks account for most of the waste
When we are handed an underperforming account, the same three faults turn up, and all three matter more than anything to do with bids.
The first is matching. Unless you tell it otherwise, Google is generous about what counts as a match for your keyword, and without a list of negative keywords you pay for searches containing free, jobs, salary, DIY, a competitor's name or a town you do not serve. Half an hour listing the words you never want to appear for, before launch, is the highest-return task in the whole exercise.
The second is measurement. With no conversion tracking installed, the account cannot see which clicks became enquiries, so it optimises towards the only thing it can see — clicks — and Google will cheerfully buy you a great many cheap, useless ones.
The third is the landing page. An ad pointed at your homepage asks the visitor to go and find the thing the ad just promised them. Most go back to the results and click the competitor who did not make them work for it.
The search terms report is the screen that matters
Keywords are what you asked to appear for. Search terms are what people actually typed. The gap between the two is where small budgets disappear. Read the report every few days for the first fortnight, then weekly, adding anything irrelevant as a negative keyword.
Two caveats. Google only shows search terms that enough people searched for, so some of your spend will always sit behind queries you never see — a tightly defined keyword list limits what that can cost you. And resist excluding every term that has not converted yet: on a small budget most terms have not had the chance, and pruning hard in week two can strangle a campaign before it has told you anything.
The enquiries that arrive as phone calls
For a plumber, a solicitor or a garage, most enquiries are not form submissions. They are phone calls, often placed straight from the ad without the website being opened at all. An account measuring only form fills will conclude that the campaigns producing your best work are failing.
Google Ads can record calls placed from an ad and, with a little setup, calls to a number on your site after an ad click. Set that up before you spend rather than after, because the data cannot be recovered retrospectively. A related caution: bidding strategies that optimise towards conversions need a steady flow of conversions to learn from, and at low volumes there is little for them to work with.
Your website is part of what the ads cost
What you pay per click is not decided by your bid alone. Google weighs the quality of the ad and the experience of the page behind it, so a slow, vague landing page raises your cost per click as well as losing you the visitors it does attract. You pay twice for the same weakness.
That surprises people who treat the ads and the website as separate purchases. The cheapest way to bring a cost per enquiry down is frequently not a lower bid or a cleverer keyword — it is a faster page that says plainly what you do, where you work and how to reach you, with the enquiry route visible without scrolling on a phone.
What the first six weeks should look like
Clicks arrive the day your ads are approved. Judgement should not. Expect four to six weeks before performance settles enough to mean anything, and treat the first month as tuition — you are buying data about which searches convert, and some of it is expensive.
Judge it on cost per enquiry, and then on cost per customer. Not impressions, not click-through rate, not the position your ad appears in. If enquiries are arriving and nobody can say what each one cost, the reporting is the first thing to fix.
Change one thing at a time and give each change a fortnight. The commonest way a small account fails is being restructured every few days by an anxious owner, which guarantees no version of it ever gathers enough data to assess.
When not to run ads at all
Saying this costs us work occasionally, and it is still the right advice. Ads are a poor idea, for now, in any of these situations:
- —Your website cannot take an enquiry well — a slow site, no clear call to action, or a contact form nobody has tested this year
- —Nobody can answer the phone while your ads are running, because a paid click that reaches voicemail is simply bought and discarded
- —The arithmetic puts your cost per customer above what a customer is worth
- —Almost nobody searches for what you sell, which is common for genuinely new products — search ads harvest existing demand rather than create it
- —Nobody will look at the account weekly for the first couple of months
How we handle it
We set up and manage Google Ads for small businesses, and the arrangement is deliberately plain: the account is in your name, your budget goes to Google rather than through us, the management fee is quoted separately, and there is no minimum term. If you leave, you keep the account, its history and its data.
We also run the estimate above before anyone spends anything, and say so when the numbers do not yet support advertising — sometimes the honest recommendation is to fix the page the ads would point at, or to put the money into organic search instead.
If you are weighing it up, or already spending and unable to tell what it is producing, tell us what you sell and what you are spending and we will give you a straight read on it. Often the finding is a negative keyword list and a better landing page rather than a bigger budget, and when that is the answer we would rather say so.