Pay per click is exactly what it sounds like. Your advert appears for free and you are charged only when somebody clicks it. That single mechanic shapes everything else about the channel.
Whoever signs off the budget is choosing to buy results rather than exposure, because money only moves when somebody acts.
Why the pay per click model changes the incentives
Because impressions are free, appearing is not the achievement. Appearing to the wrong person is simply harmless, while appearing to the right person and losing them after the click is expensive.
Consequently the interesting work sits either side of the click. Targeting decides who sees the advert, and the landing page decides whether the click was worth paying for. The advert itself is the smallest part.
What decides your cost
Three things, and only one of them is under your control.
Competition sets the floor, since you are bidding against everybody else who wants that search. Relevance moves your cost up or down, because platforms reward adverts people actually click. Your bid sets a ceiling rather than a price, which surprises people who expect to be charged what they entered.
So the lever you own is relevance. A better matched advert and a better landing page lower what you pay for the same position.
Where pay per click fits against everything else
It is the fastest channel to start and the fastest to stop. Turn it on and traffic arrives the same day. Turn it off and it stops the same day.
That makes it excellent for testing demand, covering a slow season, or filling a gap while slower work builds. It also makes it a poor foundation on its own, because nothing accumulates. Meanwhile organic search and reputation keep working when the card stops being charged.
The searches worth paying for
Not all clicks are equal, and the difference is intent.
- Somebody ready to buy, using words like quote, book or near me.
- A person comparing two named options against each other.
- Researchers reading about a problem, who are months away.
- Anybody hunting for a free version of what you sell.
The first two are usually worth paying for. Researchers rarely are while the obvious demand is still uncovered. Free hunters almost never are, and they are exactly where negative keywords earn their place.
The arithmetic behind a pay per click decision
You need three numbers before deciding whether a click price is reasonable: what a customer is worth, how many enquiries become customers, and how many clicks become enquiries.
A worked example with invented round numbers, not a benchmark. Say twenty clicks produce one enquiry, and four enquiries produce one customer. That is eighty clicks per customer. If a click costs one unit, a customer costs eighty. Whether that is affordable depends entirely on what a customer is worth to you, which is a number only your own accounts hold.
Work that out before the first campaign rather than after the first invoice. Most disappointment with pay per click is really a mismatch between click price and customer value that was visible from the start.
Quality of leads, not just quantity
A campaign producing many cheap enquiries that never buy is worse than one producing few expensive ones that do.
So judge on qualified enquiries. That means somebody in the business marking which leads were real, and feeding that back into which campaigns get the budget. Without that feedback the account optimises towards whatever is easiest to generate, which is rarely what you wanted.
Common ways pay per click goes wrong
Five faults account for most of it.
Sending clicks to a homepage instead of a relevant page. Leaving matching loose with no negative keywords. Running so many campaigns that none gets enough volume to judge. Stopping after two weeks, before anything can be learned. And measuring form fills rather than enquiries worth having.
None of those is technical, and all five are cheaper to avoid than to diagnose later.
When to stop or scale
Scale when the cost per qualified enquiry is comfortably below what one is worth and has been steady for long enough to trust. Increase gradually and watch the cost rather than the volume, since efficiency falls as you push beyond your best audience.
Stop when the cost per result stays above what the outcome is worth after a genuine attempt to fix the page and the targeting. Stopping is a legitimate answer, and it is cheaper than a year of hoping.
Getting started without wasting a month
Set up conversion tracking first. Pick the one service that pays best. Use tight matching, an obvious negative list, and a page about that exact thing.
Then leave it alone long enough to produce a real answer, changing one thing at a time. Pay per click rewards patience and structure far more than it rewards clever settings.
Write down what you expect before you start, in one sentence. Comparing the result against that sentence is a far better test than reading a dashboard and deciding afterwards what you had been hoping for.
Frequently asked questions
What does pay per click actually mean?
Your advert appears at no cost and you are charged only when somebody clicks it. That mechanic shifts the important work either side of the click: targeting decides who sees it, and the landing page decides whether it paid.
What decides how much I pay per click?
Competition sets the floor, relevance moves your cost up or down, and your bid sets a ceiling rather than a price. Relevance is the only one you control, so better adverts and pages lower your cost.
How do I know if a click price is affordable?
Work out what a customer is worth, how many enquiries become customers, and how many clicks become enquiries. Those three numbers give you a cost per customer. Do that before the first campaign rather than after the first invoice.
Should I judge on enquiries or on sales?
On qualified enquiries at least, and on sales where you can. A campaign producing many cheap enquiries that never buy is worse than one producing fewer that do. Somebody has to mark which leads were real.
How is this different from organic search?
Pay per click starts and stops the same day, which makes it excellent for testing demand or covering a slow season. Nothing accumulates, though. Organic search and reputation keep working after the card stops being charged.
Which searches are not worth paying for?
People looking for a free version of what you sell, and usually people researching a problem months before buying. Negative keywords exist for the first group, and the second becomes worthwhile only once the obvious demand is covered.
How long before I can judge a campaign?
Long enough to gather results that mean something, which is rarely two weeks. Stopping early is one of the five common faults, along with loose matching, homepage landing pages, too many campaigns and measuring form fills.
When should I stop?
When the cost per qualified result stays above what that result is worth, after a genuine attempt to fix the landing page and the targeting. Stopping is a legitimate answer and it costs far less than a year of hoping.