Most of Google Ads is an auction you influence but do not control.
You choose keywords, and Google decides which searches match them. You set bids, and an algorithm adjusts them. You write ads and a system assembles them.
Get Found, Get Chosen, Get Your Time Back
A free one-day marketing seminar in Milton Keynes · 10 November 2026
Then somebody clicks, and for the next few seconds everything is yours. Your page, your words, your form, your decisions.
Get your free Marketing Action PlanNot sure where to start? Answer a few questions and get your plan in about a minute. No email needed.
Get my plan →That’s the one part nobody else has a say in, and it’s the part most businesses have never examined.
The click is already paid for
Here’s the framing that changes how this feels.
By the time somebody lands on your page, the money is gone. You paid for that click whether they enquire or leave.
So the page isn’t where you save money. It’s where you decide whether the money you already spent turns into anything.
Two identical accounts, same keywords, same bids, same budget. One converts at four per cent and one at eight. The difference is entirely on the page, and it halves the cost of every enquiry.
That’s why I’d look here before touching bids again. Bidding changes fight the auction. Page changes change the arithmetic.
The double cost of a weak page
Most articles cover either conversion rate or Quality Score. The compounding of the two is what makes this expensive.
Quality Score has three components. Expected click-through rate, ad relevance, and landing page experience.
That third one is your page. It’s not a minor input; it’s a third of the calculation, and it feeds directly into what you pay per click.
So a weak page bills you twice
The first bill is obvious. Fewer visitors turn into enquiries, so each enquiry costs more.
The second is quieter and compounds. Poor landing page experience drags Quality Score, and a lower Quality Score means you pay more for the same position on that keyword.
You end up paying a premium for clicks that then convert badly. Both effects come from the same page, and businesses usually only notice the first one.
Which way round to fix it
Improve the page and both numbers move together. Conversion rate rises, and over the following weeks the Quality Score on the keywords pointing at that page tends to follow.
That’s the closest thing to a free lunch in a Google Ads account, and it doesn’t require a bigger budget.
What Google actually requires
There’s a lot of noise about landing page policy at the moment, some of it invented, so let me be precise.
Google publishes a set of destination requirements. They govern where your ads send people. There are eleven of them.
The ones worth knowing are these.
Destination mismatch means your ad has to accurately reflect where it sends people. No misleading display URLs, no redirecting to a different domain.
Destination not working covers destinations that return errors or don’t work on common browsers and devices.
Destination not accessible means the page has to be reachable in the location you’re targeting. Geo-blocking your own advert is more common than you’d think.
Destination experience requires the page to be easy to navigate and safe. It prohibits difficult navigation, direct downloads and disruptive ad formats.
Destination not crawlable means Google needs to be able to read the page to verify the ad is accurate.
What is not in the policy
This matters because several articles ranking for this topic say otherwise.
Google’s destination requirements page does not mention page load speed. It does not mention privacy policies. It does not mention cookie consent.
We checked the source before writing this, because a business acting on those claims would be solving a problem that doesn’t exist while ignoring the ones that do.
To be clear, speed and privacy both matter. Speed affects conversions and feeds the landing page experience. A privacy policy is often required for other reasons entirely, including data protection law and specific ad formats like lead forms.
They’re just not destination requirements, and treating an invented policy as urgent is how businesses end up rebuilding the wrong thing.
The policy nobody mentions
Buried in that list is a rule with real consequences, and I’ve never seen an agency blog discuss it.
It’s called insufficient original content.
The policy prohibits pages built primarily to carry adverts, content replicated from elsewhere without adding anything, redirect-only sites, and pages that make no sense to a visitor.
Why that should worry you
Look at what most landing page guides recommend building.
Strip out the navigation. Remove the links. Cut the content to a headline, three bullet points and a form. Make it as focused as possible.
Taken far enough, that produces exactly what the policy describes. A page with almost no original value, existing only to receive advert traffic.
I’m not saying a focused page is against the rules. Focus is good, and it converts. But there’s a floor, and the thinnest version of the advice on page one sits uncomfortably close to it.
The practical version
Your page should be able to justify its existence to somebody who arrived from a Google search rather than an advert.
Does it answer the question properly? Does it contain information a real customer would find useful? Would it be a reasonable page to land on if there were no advert at all?
If yes, you’re comfortably clear and you’ll probably convert better too, because thin pages under-perform with humans as well as policies.
If your page is a headline and a form with nothing else, that’s worth revisiting for both reasons.
Message match, the fix with the best return
If you only change one thing, change this.
Message match means the page continues the sentence the advert started. Somebody searched for a thing, clicked an ad about that thing, and should land on a page about that thing.
It sounds obvious. It’s broken in most accounts.
The common failure
A plumber bids on “emergency boiler repair”. The advert says emergency boiler repair. The click lands on a homepage listing bathroom installations, underfloor heating, a photo of a van and a company history.
The visitor has to work out whether this company does the thing they urgently need. Most won’t do that work. They’ll go back and click the next result.
Nothing was wrong with the advert. The handover failed.
How to check yours in five minutes
Take your three highest-spending keywords. For each, open the advert and then open the page it points to.
Ask one question. Does the top of that page, without scrolling, repeat the language of the search?
If somebody searched “commercial waste collection Leeds” and the first thing they see says “Commercial waste collection in Leeds”, that’s a match. If it says “Welcome to our website” or lists eleven services, it isn’t.
Fix the headline before anything else. It’s usually a ten-minute change, and it moves conversion rate more reliably than anything else on this list.
Speed, honestly assessed
Speed is on every list, usually with dramatic statistics attached. Let me be measured about it.
Google’s research on mobile page speed has long shown a steep relationship between load time and conversions, with pages loading in about a second converting at roughly three times the rate of pages taking five.
Treat the exact multiple as directional rather than a promise for your business. The direction is not in dispute.
What to actually do
Test your page on a phone, on mobile data, not on your office wifi. That’s the experience most of your visitors get.
If it takes noticeably longer than about three seconds to become usable, it’s costing you.
The usual culprits in small business sites are large uncompressed images, a page builder loading half a dozen scripts, chat widgets, and tracking tags stacked up over the years.
Images are almost always the biggest and the easiest to fix.
Where I’d put it in the queue
Below message match, above form length.
A fast page with the wrong headline still fails. A slightly slow page with a perfect headline usually converts. Fix the message first, then the speed.
The form, and why length is the wrong question
Every guide says shorter forms convert better. Broadly true, and it’s not the whole story.
The right question isn’t how many fields. It’s whether each field earns its place.
The test for each field
Ask what you’d do with the answer.
Name and a way to contact them. Obviously required.
A short description of the job. Usually worth it, because it saves a phone call and helps you judge whether the enquiry is worth chasing.
Company size, how they heard about you, budget range, a dropdown of eleven service categories. Each of these costs you completions. Some of them earn it back by qualifying the lead. Most don’t.
The link to lead quality
There’s a real tension here, and it’s worth naming.
A very short form gets you more enquiries and more rubbish. A longer form gets you fewer and better.
Which you want depends on where you sit. If you’re drowning in unqualified enquiries, adding a qualifying question is a sensible filter. If your phone barely rings, cut the form to the bone.
That’s a decision about your business, not a best practice. Anyone telling you three fields is universally correct hasn’t asked what your problem is.
The mobile point
Most of your traffic is on a phone. Fill out your own form on a phone before you decide it’s fine.
Fields that need precise typing, dropdowns that behave badly, a date picker built for a desktop. These kill completions, and they’re invisible from a laptop.
Trust, which matters more for small businesses
A visitor who has never heard of you is deciding whether to hand over their phone number.
Big brands get this for free. You have to earn it on the page.
The things that do the work are unglamorous. A real address. A landline and a mobile. Photographs of actual work or actual people rather than stock images. Review scores with the source named. Accreditations that mean something in your trade.
Stock photography of a smiling call centre does the opposite of what it’s meant to do.
LIVELive every Tuesday at 2pm UK. Bring your marketing questions.
Get your Google, local SEO and AI questions answered live.
The specific version beats the impressive version
“Over 500 satisfied customers” reads as marketing.
“We’ve been fitting flat roofs across West Yorkshire since 2011” reads as true, and it’s checkable.
Specific, modest and verifiable will out-convert broad and impressive almost every time, particularly for a local service business where the buyer is worried about being let down.
The twenty minute audit
In order of what actually moves the numbers.
- Open your three highest-spending keywords, then open the page each one points at. Check whether the top of the page, before scrolling, repeats the language of the search. Fix the headlines that do not.
- Load each page on a phone using mobile data. If it takes much more than three seconds to become usable, note it. Check image sizes first.
- Fill in your own form on a phone. Every field that annoys you will annoy a customer more, because they care less than you do.
- Count the form fields and ask what you would do with each answer. Remove any you would not act on.
- Look at the page as a stranger. Can they tell what you do, where you do it, and what happens if they get in touch, without scrolling?
- Check the trust signals. Real address, real phone number, real photographs, named review sources.
- Ask whether the page would justify its existence with no advert pointing at it. If it is a headline and a form and nothing else, add the substance a real customer would want.
- Check landing page experience in the Google Ads keyword view. Below average on a high-spending keyword tells you where to start.
Most accounts get the majority of the available gain from the first item on that list.
The visitor who is not ready yet
Most landing page advice assumes one outcome. Somebody arrives, they’re convinced, they enquire.
That’s a minority of your traffic on almost any page, and the rest get ignored entirely.
Think about who else lands there. Somebody comparing three suppliers. Somebody whose contract ends in two months. Somebody researching on behalf of a boss who’ll make the decision.
None of them is going to ring you this afternoon. All of them might buy eventually.
Give them something smaller to do
A page with one call to action treats every visitor as ready. A second, lighter option catches the ones who aren’t.
That might be a price guide, a straightforward explanation of how the work is done, a checklist of what to ask any supplier in your trade, or simply a phone number for a no-obligation question.
The point isn’t the format. It’s that somebody unwilling to request a quote today might still take a step that lets you speak to them later.
Don’t let it compete with the main action
The mistake is giving both options equal weight, which just creates a decision.
Make the primary action obvious and the secondary one clearly smaller. A prominent enquiry form, and a quieter line underneath offering the guide.
Ready buyers won’t be distracted. Unready ones get an exit that isn’t the back button.
Why this matters for a small budget
You paid for every one of those clicks, including the ones from people who weren’t ready.
Capturing even a fraction of them changes the economics of the whole campaign, because the same traffic now produces two kinds of value rather than one.
This is also where a landing page connects to the audience work. Somebody who downloaded a guide is a person you can reach again cheaply, which is a far better position than paying to introduce yourself twice.
Testing changes when you have a small budget
Every conversion article tells you to A/B test. Almost none of them are honest about what that needs.
Proper split testing wants a lot of traffic. If your page gets a couple of hundred visits a month, a formal test comparing two versions will take months to say anything, and by then the season has changed, and the result is worthless.
So here’s what I’d actually do instead.
Make obvious fixes without testing
If the headline doesn’t mention what the person searched for, change it. Don’t test it.
Some changes are corrections rather than hypotheses. A broken handover, a four megabyte image, a form asking for a job title on an emergency callout. Fix those and move on.
Reserve testing for genuine unknowns, like whether a longer or shorter form suits your business.
Use before and after carefully
With small traffic, sequential comparison is usually your only option. Change one thing, then compare a month against the previous month.
It’s imperfect, and you should know why. Seasonality moves, competitors change bids, and a good month can be luck.
Two rules make it more honest. Change one thing at a time, so you know what caused any shift. And give it a full month at minimum, because a fortnight of small numbers tells you almost nothing.
Know when you cannot conclude anything
If a page got sixty visits and four enquiries before, and seventy visits and six enquiries after, that isn’t a result. It’s noise, and treating it as evidence leads to confident nonsense.
Be willing to write down inconclusive. It’s a more useful answer than a decision built on eleven data points.
And none of this works at all if the conversion tracking is wrong, which is the subject of the previous post in this series and the reason it comes first.
A worked example, start to finish
A hypothetical to show how the gains stack. A commercial electrician on twelve hundred pounds a month, sending everything to the homepage.
They convert at about two per cent. Roughly two hundred clicks a month, four enquiries, three hundred pounds an enquiry. The account looks like a failure and the keywords get blamed.
They start with message match.
Their top keyword is “emergency electrician [their city]”. The homepage headline says the company name and “Electrical Contractors Since 2009”, followed by six services.
They build one page for emergency callouts. The headline repeats the search. The first line says they answer the phone twenty-four hours and typically attend within two hours. The phone number sits at the top.
Conversion roughly doubles on that keyword alone, because the handover finally works.
Then speed.
The new page loads in about six seconds on mobile data, because the header image is a four-megabyte photograph. Compressed properly, it drops under two.
Then the form.
The old one asked for company name, job title, service type from a dropdown and preferred callback window. For an emergency callout, that’s absurd. It becomes name, phone number and a box for what’s wrong.
Then trust.
They add photos of their own vans and engineers, their Checkatrade score with the source named, and their actual address rather than a contact form alone.
Where it lands
Conversion on that keyword goes from two per cent to somewhere around six. Same clicks, same spend, three times the enquiries.
Over the following weeks, the Quality Score on that keyword improves, so the cost per click drifts down as well. The second gain arrives quietly, and nobody notices it happening.
Nothing in that sequence required a bigger budget or a new agency. It required somebody to look at the page a stranger sees.
Homepage or dedicated page
The question I get asked most, and the answer has a nuance worth knowing.
For most advertised services, a dedicated page wins. It can repeat the search language, answer one need properly, and remove the choices that distract.
But the thin, navigation-free version has two problems. It can bump against the insufficient original content policy, and it removes the reassurance some buyers need before they’ll trust you.
What I’d build
A proper service page rather than a stripped landing page.
It keeps your navigation, because a nervous buyer often wants to check your about page before enquiring. It carries real content about that service. It answers the obvious questions. And it leads with the thing the searcher typed.
That page serves your advert, and it can rank in organic search too, which is a better return on the work than a page that only exists while a campaign runs.
When a stripped page is right
Short campaigns, single offers, events with a deadline. Cases where there genuinely is one action and nothing else matters.
For an ongoing service that you advertise every month, build something with substance.
The mistakes that waste the most money
- Sending every keyword to the homepage, so the visitor has to work out whether you do the thing they searched for.
- A headline that talks about your company rather than repeating the search.
- Judging page speed on office wifi rather than a phone on mobile data.
- Adding form fields nobody acts on, then wondering why completions dropped.
- Stock photography instead of your own work, which reads as generic to exactly the buyer who is worried about being let down.
- Building a headline-and-form page with no substance, which underperforms with humans and sits near the insufficient original content policy.
- Believing the agency blogs claiming page speed and cookie consent are destination requirements. They are not, and it distracts from the things that are.
- Testing page changes while conversion tracking is unreliable, which makes every result meaningless.
Frequently asked questions
Does the landing page affect Quality Score?
Yes. Landing page experience is one of three Quality Score components, along with expected click-through rate and ad relevance.
That’s why a weak page costs twice. You lose conversions, and you pay more per click on the keywords pointing at it.
Should I send Google Ads traffic to my homepage?
Usually not. A homepage serves everybody, so it can’t answer the specific question one searcher typed.
A page about the service being advertised almost always converts better, because the headline can repeat the search and the content can address that single need.
What are Google Ads destination requirements?
The policies governing where your ads send people.
They include destination mismatch, destination not working, destination not crawlable, destination not accessible, destination experience and insufficient original content, among others.
Note that page speed, privacy policies and cookie consent are not among them, whatever some articles claim.
How fast should a Google Ads landing page load?
Under three seconds is a sensible minimum, and faster is meaningfully better.
Google’s mobile page speed research has long shown a steep relationship between load time and conversion rate.
Test it on a phone using mobile data rather than your office connection, because that’s what your visitors actually get.
How long should a landing page form be?
Short enough that people finish it, long enough to qualify the enquiry.
Ask what you’d do with each answer. If you wouldn’t act on it, remove the field.
If you’re getting too many poor enquiries, one qualifying question is a reasonable filter. If your phone barely rings, cut the form back.
Why is my landing page experience rated below average?
Usually a mismatch between the keyword and the page, or a page serving a broad audience rather than that specific search.
Difficult navigation and thin content also contribute.
Start by checking whether the top of the page repeats the language of the keyword. That’s the most common cause and the quickest fix.
Do I need a privacy policy on my landing page?
Often yes, but not because of destination requirements.
Data protection law and certain ad formats, including lead forms, have their own requirements.
It’s worth having regardless. Just don’t let anyone tell you it’s a destination policy, because that’s not what the policy says.
Where to start this week
Open your three highest-spending keywords and look at the page each one lands on.
Ask the only question that matters at first glance. Does the top of this page repeat what the person searched for?
Fix the headlines where the answer is no. That’s usually an afternoon; it costs nothing, and it moves conversion rate more reliably than any bidding change you could make instead.
One caveat before you measure any of it. If your conversion tracking isn’t trustworthy, you’ll have no way of knowing whether the change worked. Sort that first, then test the page.
The click is already paid for by the time somebody arrives. What happens next is the only part of the whole system that belongs entirely to you.
Written by Uzair Kharawala
Partner at SF Digital Studios, a Google Premier Partner agency. Uzair has been helping businesses grow with Google Ads since 2002 and teaches it every week on YouTube.
Get our best Google Ads tips by email
New articles and videos, straight to your inbox. Unsubscribe any time.