Google Ads is worth it when three things are true at once: people are already searching for what you sell, your profit per customer can absorb the cost of winning one, and you can tell which enquiries came from the ads. Miss any of the three and the spend is a guess. Here is how to decide before you fund anything.
Is anyone actually searching for what you sell?
Search advertising captures demand that already exists. It does not create it. So the first question is not what to spend, it is whether people are typing anything at all, and whether what they type describes a problem you get paid to solve rather than a topic you happen to know about.
The distinction that matters is intent, not volume. Someone typing a phrase that names an urgent, specific, paid job is a different prospect from someone researching a category. Google’s own Keyword Planner, inside a Google Ads account, is the place to check this, but read it for the shape of the demand rather than the size of it: the terms that describe a job someone wants done now are usually lower in volume and higher in value than the broad category terms sitting above them. If nothing in your category returns meaningful search activity, that is a real answer and a useful one. It means the demand you want has to be created somewhere else first, and paid search is the wrong instrument for creating it. KM Digital Solutions treats a thin keyword picture as a reason to stop rather than a reason to widen the targeting until something moves.
Three shapes of demand, and what each one implies.
| What people type | What it means | Is paid search the right tool? |
|---|---|---|
| A named urgent job, with a place or a qualifier | Ready to buy, comparing suppliers now | Yes, this is what search advertising is for |
| A category or a how to question | Researching, months from a decision | Rarely. Content and search optimisation cost less |
| Almost nothing | The demand does not exist yet | No. Something has to create it first |
Do your numbers support paying for clicks?
You need two figures before anyone quotes you: the gross profit a new customer leaves behind, and the share of enquiries that become customers. Multiply them and you have the most you can pay for an enquiry before the campaign is losing money. Everything else is downstream of that one calculation.
Work it as arithmetic rather than as a target. As a hypothetical teaching example, and not a result from our work or from any client’s account, suppose one enquiry in five becomes a customer. Then five enquiries buy you one customer, so the most you can pay per enquiry before you break even is one fifth of the gross profit that customer leaves behind. If a click converts to an enquiry one time in twenty, twenty clicks stand behind each enquiry, and your ceiling per click is that enquiry ceiling divided by twenty. Run your own numbers through the same two divisions. There is no defensible South African benchmark to substitute for them: cost per click is settled in a live auction, conversion rates vary by industry, offer and page, and any figure presented as the national average for your sector was almost certainly assembled from someone else’s account.
The second thing to get straight is that a Google Ads budget is two separate costs, not one.
- Media spend. This goes to Google. You set it, you can change it whenever you like, and Google publishes no minimum.
- Management. This goes to whoever builds and runs the account. It is a separate published line on our rate card and it does not come out of the media budget.
- Once-off account setup. Account structure, tracking and the first campaign build. A once-off account setup fee applies; it is quoted per account.
- The landing page. If the page the ads point at does not exist yet, it is a build cost and it belongs in the decision.
Confusing the first two is the most common way a small business misjudges affordability. What a campaign costs to run nationally, and how media spend behaves across a month, is a longer subject with its own guide.
Can you measure what happens after the click?
If you cannot tell which enquiries came from the ads, you cannot tell whether the ads work, and you will end up arguing about impressions instead. Conversion tracking is not a reporting nicety. It is the input the bidding itself runs on, and without it the system is optimising toward nothing.
Three things have to be working before the first rand is spent, and they are cheaper to fix beforehand than to diagnose afterwards. Conversion tracking has to record the actions you actually care about, which Google Ads documentation describes as choosing what you identify as valuable, such as a purchase, a sign-up or a phone call. The page the ad points at has to keep the promise the ad made, because Google’s published Quality Score diagnostic is scored partly on how relevant and useful that landing page is to the search. And someone has to answer the enquiries quickly, because an unanswered lead costs exactly the same as an answered one. All three sit outside the ad account, which is why a KM Digital Solutions audit usually starts there.
How long before you know whether it is working?
Long enough to accumulate conversions, not a fixed number of weeks. Google Ads recommends judging performance over a period containing at least thirty conversions, and at least fifty for Target ROAS. How long that takes depends entirely on your volume, so the honest answer is a range you calculate rather than a promise anyone makes.
That reframes the whole trial. A campaign generating four enquiries a month needs the better part of eight months to reach thirty, which means a three-month trial on that volume proves almost nothing and a decision taken at week six is a coin toss dressed as analysis. A campaign generating forty a month gets there inside a month. Work out which you are before you agree a review date, because the review date should follow the arithmetic rather than the invoice cycle. Google Ads publishes no minimum budget at all, stating that the amount is entirely up to you, and it publishes the two mechanics that actually bound your spending: on any single day your cost may reach twice your average daily budget, and across a month you will not be charged more than 30.4 times it. Those two published rules, read from Google Ads Help in September 2026, are the only spend guarantees in the system, and our cost guide works them through at four budget levels.
A bounded first period, run properly, looks like this.
- Fix the measurement first. Tracking, the landing page and the enquiry follow-up, all confirmed working before anything is funded.
- Calculate your ceilings. The most you can pay per enquiry and per click, from your own two figures.
- Set a review point in conversions, not weeks. Thirty conversions, or fifty if the bidding targets return on spend.
- Change one thing at a time. Keywords, then page, then offer. Simultaneous changes make the result unreadable.
- Decide out loud. Continue, change, or stop, against the ceilings you wrote down at the start.
When is Google Ads the wrong answer?
When nobody searches for the thing, when the margin cannot carry the cost of a customer, when nothing after the click is ready, or when the budget is too small to reach a readable number of conversions in a reasonable time. Any one of those is enough to say no.
Saying no is a legitimate outcome and it should be available to whoever is advising you. A business in Cape Town selling a low margin product to an audience that does not search for it will lose money faster with a well built campaign than with a badly built one, because the good campaign spends the budget more efficiently on the wrong thing. In those situations the money usually does more in search optimisation, in a better site, or in simply answering enquiries faster. KM Digital Solutions would rather turn down an advertising brief in the Western Cape than run one whose arithmetic never worked, and the decision framework above is the same one we apply before quoting. Whatever the answer, get the term and the notice period in writing before you sign, from us or from anyone else, because that is the clause most often left to a verbal assurance.
Work out whether paid search fits your numbers
Bring your gross profit per customer and your enquiry to customer rate, and we will tell you honestly whether search advertising can clear them. If it cannot, we will say so and point you at what would work better.
Campaign build and ongoing management sit in our paid advertising service. Where the answer is that demand has to be earned rather than bought, search optimisation is the longer route, and if the problem is the page rather than the traffic, that is website design and development work. Two companion guides cover what Google Ads management includes each month and what happens after the click.
Frequently asked questions
Is Google Ads worth it for a small South African business?
It is worth it when people already search for what you sell, when your gross profit per customer can absorb the cost of winning one, and when you can measure which enquiries the ads produced. Those three conditions decide it, not the size of your budget. If any one of them fails, the money almost always does more elsewhere, and a campaign built well will simply lose it more efficiently.
What is the minimum I need to spend?
Google publishes no minimum, and states that the budget amount is entirely up to you. The practical floor is set by arithmetic instead: your spend has to be large enough to accumulate a readable number of conversions in a period you are willing to wait through. Work out how many enquiries a month you expect, then check how long thirty conversions would take at that rate.
How long before Google Ads shows results?
There is no fixed learning period that applies to every campaign. Google Ads recommends evaluating performance over a window containing at least thirty conversions, and fifty for Target ROAS, so the honest answer depends on your conversion volume. At forty enquiries a month that is roughly a month. At four a month it is most of a year, which usually means the trial itself is the wrong shape.
Do I need conversion tracking before I start?
Yes, and it is not optional in any meaningful sense. Conversion data is what the bidding optimises against, so without it the system has no signal to improve toward and you have no way to separate enquiries the ads produced from ones you would have received anyway. Set it up, confirm it records a real enquiry end to end, and only then fund the campaign.
Should I run Google Ads or Facebook Ads?
Start with the one that matches your demand. If keyword research shows people already searching for what you sell, search advertising captures that demand and is usually the better first test. If almost nobody searches, paid social can put an offer in front of people who were not looking, which suits new offers, visual products and local awareness. How Meta prices that attention is covered in our guide to Facebook ads costs.
Google Ads Help pages on conversion measurement, Smart Bidding, managing spend, Quality Score and Keyword Planner are the sources for this article and were all accessed on 16 September 2026. The months-to-thirty-conversions chart is arithmetic on hypothetical volumes, not a result from any account.
