Your Google Ads landing page is where most of the budget is decided, and it usually gets the least attention. Traffic gets tuned endlessly while the page it arrives at has not been touched in two years.
The gap is often large. Unbounce’s data puts the median landing page at 6.6%. Plenty of the accounts I open are converting at around two. That difference means paying three times as much for every enquiry, at the same click price.
This is for you if your click costs look reasonable and your cost per lead does not.
What a Google Ads landing page is competing against
For scale, Unbounce analysed 41,000 landing pages, 464 million visitors and 57 million conversions, putting the median conversion rate at 6.6% on Q4 2024 data.
Two caveats before you use that. It covers Unbounce customers, who are by definition people paying attention to landing pages, so it likely runs optimistic. And it is Q4 2024 data, so it is not this year’s figure.
Even allowing for both, the direction is clear. If you are at 2%, the problem is not that the market is hard. Something specific on your page is losing people, and it is usually one of four things.
The Google Ads landing page mistakes that cost the most
In the order I find them.
Sending paid traffic to the homepage. A homepage serves everybody, so it commits to nothing. Someone who searched for one specific service arrives at a page listing nine and has to work out where to go next. A meaningful share will not bother.
No message match. The search said emergency boiler repair and the headline says Welcome to our website. There is a half-second of doubt about whether this is the right place, and doubt is expensive.
Too many form fields. Every field costs completions. Asking for company size, budget and timeline before anyone has spoken to you filters out people who would have converted on the phone.
Slow on mobile. Most paid traffic arrives on a phone, often on a patchy connection, and a page that takes five seconds loses people who never saw your offer at all.
Why the Google Ads landing page is the strongest lever you have
Cost per lead is click cost divided by conversion rate. That is the whole relationship.
At $5 a click and 2% conversion you pay $250 a lead. At $5 and 6% you pay $83. Same bid, same auction, same competitors, a third of the cost.
Nobody bids against you on conversion rate. It is the one improvement in paid search with no auction attached, which is why it beats bidding harder almost every time.
The whole market demonstrated this in 2026. WordStream’s study of 13,474 US search campaigns found average cost per lead falling for the first time in five years, driven by conversion rate rising for 87% of industries. Not by anyone winning a bidding war.
Google Ads landing page message match, in practice
The page should repeat the words of the search that brought someone there.
That means the headline uses the phrase, not a synonym. Someone searching emergency boiler repair should see emergency boiler repair, not rapid heating solutions. The second is a marketing department talking to itself.
Your search terms report is the source for this. It shows the actual words people typed, which are rarely the words the business uses internally, and lifting them into the headline closes the gap between what they asked for and what they see.
You do not need one page per keyword. You need one page per intent. Group the searches that want the same outcome and serve them one page that speaks their language.
Form fields and the cost of asking
Cut the form to what your sales process genuinely uses.
The test is simple: if nobody looks at the answer before the first call, delete the field. I have removed three fields from a form and watched completions rise by a third, with no change in lead quality, because the fields were collecting information the salesperson asked for again anyway.
The exception is deliberate qualification. If you are drowning in unqualified enquiries, one well-chosen question that costs you volume and raises quality is a good trade. Just make it a decision rather than an accident.
Speed, and why it costs twice
A slow page loses people before they see the offer. That much is obvious.
The second cost is less obvious. Page experience feeds into ad rank, so a slow page can raise what you pay per click as well as lowering what you get from it. You lose at both ends.
Test on a phone, on mobile data, not on your desk over office wifi. That is the experience most of your paid traffic is actually having, and it is usually worse than anyone in the business realises.
What to test, and in what order
Headline first, because message match is the cheapest fix and often the largest.
Form length second, because removing a field takes a minute and the effect is immediate.
The offer third. What happens after someone enquires is the thing you are actually selling at that moment, and making it concrete beats making it prettier.
Layout and design last. It is the most fun and the least productive, and it is where most people start.
Give each test two weeks and a couple of hundred conversions if you can. Small samples produce differences that look decisive and are random.
What to change this week
Four steps.
Load your own page on a phone, on mobile data, and time it. Then read the headline and ask whether it repeats what somebody searched for.
Count your form fields and delete any whose answer nobody reads before the first call.
Check where your paid traffic actually lands. If any campaign points at the homepage, that is your first build.
Then work out what a doubling of conversion rate would do to your cost per lead, so the priority is obvious to whoever approves the work.
The wider piece on judging lead cost explains the ceiling this feeds, negative keywords stop the wrong people arriving in the first place, and instant forms against landing pages covers the same trade on Meta. If you are wondering whether one person can handle the ads and the pages together, I have written about exactly that. For the build itself, web development is part of what I do, and you can get in touch to have your page looked at.