Your ads are specific. Your homepage is not.
Somebody searches for one precise thing, clicks your ad, and lands on a page that tries to introduce your whole company. They scan, do not find the thing they searched for in the first screen, and leave. You paid for that click. This is the most common and most expensive setup error in US paid search, and it is usually inherited rather than chosen. Here is how to tell whether it is costing you, and how many pages you actually need.
The homepage is written for a different visitor
A homepage has an impossible job: it serves someone who typed your brand name, someone who arrived from a referral, someone comparing three companies, and someone who wants a phone number. It answers all of them a little and none of them well. That is a reasonable compromise for organic traffic, where visitors arrive with mixed intent.
Paid traffic is the opposite. You know exactly what the person typed, because you bought that query. Sending them to a page that does not repeat their words in the first screen throws away the one advantage paid search gives you. The fix is not a prettier homepage. It is a page whose headline continues the sentence the ad started.
- The ad promised something specific; the page has to confirm it above the fold
- One audience, one offer, one action, measured on its own
- Navigation that invites wandering works against a paid visitor
- The phone number and the form belong where the thumb already is
It costs you twice, and the second cost is invisible
The obvious cost is conversion: fewer of the people you paid for take action. The second cost sits inside the auction, though not where most people look for it. The 1-10 Quality Score in your account is not what runs in the auction: Google states plainly that it is a diagnostic tool and not an input in the ad auction. What gets evaluated every time someone searches is Ad Rank, which Google calculates from your bid, the quality of your ads and landing page at auction time, the Ad Rank thresholds, the competitiveness of the auction and the context of the search. Landing page experience is one of the three components Quality Score reports on, alongside expected click-through rate and ad relevance, so a low score is a useful warning sign. But it is the live assessment of that same page, not the number in the column, that moves what you pay and whether you show.
So a mismatched destination can push your cost per click up while lowering the share of clicks that convert. Both movements push cost per acquisition in the same direction, which is why accounts pointed at a homepage tend to feel like they are getting quietly more expensive without anything visibly changing.
How many pages you actually need
The useful unit is not the campaign and not the keyword: it is the promise. If two ad groups make the same promise to the same kind of buyer, they can share a page. If they make different promises, they need different pages. A company advertising emergency repair and scheduled installation is making two different promises to two people in two different states of mind, and one page cannot serve both.
In practice most local service businesses need a small number of pages, not dozens. Start with the campaigns carrying the most spend, build a real page for each promise, and leave the rest pointed at the relevant service page until the data justifies more. Building thirty near-identical pages with a city swapped is the failure mode on the other side, and it produces thin pages that help nobody.
- One page per promise, not per keyword
- Start with the campaigns spending the most money
- A service page can be the destination when the promise matches it
- Do not mass-produce city variants of the same page with a find-and-replace
- Track the form and the call separately, per page
When a dedicated page will not fix anything
Honesty is worth more than another invoice here. If your campaigns are attracting the wrong searches, a better page will convert the wrong people slightly more efficiently and your business will not improve. Check the search terms report before building anything: if you are paying for queries you would never quote, the problem is targeting, not design.
The same applies to volume. If a campaign produces a handful of clicks a month, no page change will produce a readable difference, and you would be optimizing noise. And if the enquiries do arrive but nobody follows up quickly, the bottleneck is operational. A landing page cannot outrun a phone that rings out.
Frequently asked questions
Can I just use my existing service page instead of building a new one?
Often yes, and it is the cheaper first move. If the service page opens with the same promise as the ad and has a clear action, point the campaign there and measure. Build a dedicated page when the promise is narrower than the service page can honestly carry.
Will landing pages hurt my SEO?
They serve a different job. A page built for paid traffic is not competing for organic rankings, and a single page rarely ranks for competitive terms anyway. The risk to avoid is publishing many near-duplicate pages, which is a content quality problem regardless of the channel.
How do I know the homepage is the problem and not the ads?
Look at the search terms report first. If the queries are ones you would happily quote and they still are not converting, the destination is a fair suspect. If the queries are irrelevant, fix targeting before touching any page.
How many should we build to start?
As many as you have distinct promises carrying real spend, which for most local businesses is a small number. Build those, measure per page, and expand only where the numbers ask for it.
Paying for clicks that land on the wrong page? Tell us which campaigns carry the spend and we will tell you which need their own page and which do not.