Skip to content
Rank on Maps Free scan

Local SEO

Website traffic but no phone calls: the two leaks and how to find yours

Usually a contractor with this problem is ranking for searches that never lead to a hire, or has a page that never tells the visitor what to do next. We answered it on our weekly Q&A, and the fix starts in Search Console, not in a redesign.

Published · 10 min read · Answered by Daniel Girmay and Darin Pirkey

Website traffic but no phone calls usually means one of two leaks: the visitors want something you do not sell, or they land on the right page and it never tells them what to do. On our weekly Q&A we described a client who told us traffic was up 30 percent under a previous agency while the phone stayed silent.

Why is my website getting traffic but no leads?

Because the traffic is the wrong traffic. That was the first thing said on the call, and it is usually where the problem comes from.

“It depends on the type of traffic that you’re getting,” as we put it on our weekly Q&A. Some of it is bots. Some of it is people outside your service area. Usually, for a contractor in this situation, much of it is people who typed a question, got their answer, and had no reason to call anyone.

The example from the call was an HVAC contractor who, before working with us, had been publishing a lot of content. The articles were good. Genuinely good, well written, on topics people search for. One of them explained the difference between types of AC filters. It brought real traffic.

The problem is that most people do not buy an AC filter from an HVAC company. The contractor was not even selling filters. So the visitor read the article, learned what they came to learn, and left. Multiply that across a lot of posts and you get a traffic chart going up and a phone that stays quiet.

This is the mechanism behind most contractor websites that are not converting. The content is answering questions the business does not get paid to answer.

How do I tell if my traffic is the wrong traffic?

Open Google Search Console, go to Performance, set the date range to the last three months, and sort queries by clicks. Then label each query with one of three words: hire, how-to, or unrelated.

Hire queries name a service and a place, or a service and “near me”, or an emergency: “water heater replacement” plus your city, “roof leak repair”, “electrician near me”. How-to queries ask a question: how to reset a breaker, what size AC do I need, why is my faucet dripping. Unrelated queries are the ones that make you squint, usually a product name or a term from a blog post that drifted off topic.

If the top twenty queries are mostly how-to and unrelated, you have found the leak. Nothing on the website will fix that, because the people arriving were never going to call.

This is exactly what we saw with the client who told us traffic was up 30 percent with the previous agency. When we looked at the Search Console data, in the words of one of us on the call, “the traffic that they’re getting is just crap. And that’s just a result of just poor quality keyword research.” Harsh, and accurate. The reports looked great because the reports measured visits. Visits were never the goal.

There is a second point from the call worth holding onto. The pool of local search for your trade is fairly steady outside of seasonal swings. When your share of it goes up, you are taking visits that previously went to a competitor. So the question is never “how do I get more traffic.” It is “which traffic in my market leads to a hire, and how much of it am I getting.” That question has an answer in Search Console. The first one does not.

For the search side of this, our local SEO service is built around hire-intent queries first, and if a competitor is winning those calls from the same pool, why your competitor is getting more calls than you walks through what to compare.

Why am I getting website visits but no sales?

Because the page never asks for the call. This is the second leak, and it catches contractors whose keyword targeting is actually fine.

On the call we were blunt about this: “You could be getting a lot of content that’s actually good traffic, but you’re forgetting the step of telling them what to do.” The visitor reads the service page, agrees that you do the thing they need, and then looks for the next step. If it is not obvious, they go back to Google and tap the next result. That happens more than you think.

Here is what “telling them what to do” means on a contractor service page:

  • A phone number at the top of the page that dials when tapped. Not in the footer. Not in an image.
  • One clear action per page. “Call for a same-day estimate” or “Request a quote”. Not four buttons that all go somewhere different.
  • A form with three fields: name, phone, what you need. Every extra field can cost you a visitor you already paid for.
  • Your service area and hours stated in plain text, so the visitor does not have to wonder whether you come to them.

Then there is the trust side. One of us on the call described what he calls inverse thinking. Pretend you are the customer, land on your own site, and ask: why wouldn’t I call this company? Is the site humanized, or is it stock photos and AI-generated images of a crew that does not exist? Are there real selling points, or just “quality service you can trust”? If you have done five thousand roofs in ten years, does the page say so? Because the competitor’s page does, so the customer clicks them instead.

Google’s own guidance on helpful content says its systems try to prioritize content that shows experience, expertise, authoritativeness and trustworthiness, and that of those four, trust is the most important. Google also notes E-E-A-T is not itself a single ranking factor. That is Google describing what its systems look for, but it is also a decent description of what a homeowner with a leaking ceiling is looking for. Real photos. Real reviews with names. A specific number about your work. A face.

Why am I getting clicks but no phone calls, and how would I even know?

You would not know, unless you are recording what visitors do. Many contractors are not. A lot of SEO agencies are not either.

On our weekly Q&A: “You put so much effort in getting that visitor on there and you’re just going to close your eyes when they jump on your site.” We record every visitor session on our clients’ websites. Heat maps show which buttons get clicked and which get ignored. If a button breaks, we see it quickly instead of finding out a month later when the owner asks why leads stopped.

You do not need our setup to get the basics. You need three things:

  1. Call tracking that tells you which page the call came from.
  2. Session recording or heat maps on your top five pages.
  3. A habit of submitting your own form and calling your own number after every site change.

The third one surprises people. A form can look fine, pass every visual check, and send nothing. Plugin updates, theme changes and relaunches are where this happens. If you have not tested your form in the last 30 days, do it before reading further.

Is AI killing website traffic?

It is taking a share of informational clicks, and that is mostly the traffic that was never going to call you. On the call we described it as a shift where people provide more context before they get a result. A homeowner who used to search “how to size an AC unit” and click three articles now asks an AI assistant and reads the answer in place.

If your traffic was built on how-to content, that drop looks scary on a chart. It usually changes little about the phone, because those visitors were rarely callers. The searches that produce calls, a service plus a city, a near me query, an emergency, still send people to businesses. Some of them now pass through AI-generated local results that can name fewer companies, which we covered in why the AI Local Pack names far fewer contractors.

So the honest answer: AI is killing the traffic you should not have been counting, and it is raising the stakes on the traffic that matters. Our AI search visibility work exists for that second part.

Where does the Map Pack fit in all of this?

The Map Pack sends you the best visitor you can get, because the person who taps it usually typed a hire query. Google says local results are mainly based on relevance, distance and prominence, meaning how well your profile matches the search, how close you are to the searcher, and how well known the business is, including reviews and links. None of that involves a blog post about filters.

A contractor whose website gets traffic but no calls can be strong in organic blog rankings and weak in the Map Pack. That is backwards for a home service business. The fix is to put the effort where the hire-intent searches land, which is what our Map Pack ranking service is for.

Why do SEO agencies ignore conversion?

The mistake is treating the visit as the finish line, with nobody responsible for the phone.

One of us on the call said it plainly: it is weird how many SEO agencies do not pay attention to conversion. We have had agencies tell us directly that it is not something they need to look at. Their job ends at the visit. The owner is left with a rising traffic chart and no explanation for the silence.

That mindset comes from an older version of search, where getting a visitor onto the site was most of the work and the site would convert reasonably well on its own. That is not true anymore. Earning the visit is one job. Earning the phone number is a second job. An agency that only does the first one will keep sending you reports that look like success.

When is fixing the website not worth it?

Three cases.

If almost nobody with buying intent is reaching the site, conversion work is premature. A better form on a page that gets two hire-intent clicks a month will not move anything. The problem is visibility, and the fix is the Map Pack and your service pages. If you are not sure whether you are even showing up, start with why your business is not showing up on Google.

If the traffic is wrong, a redesign is wasted money. A prettier page for a how-to query still gets zero calls. Fix the keyword targeting first. The redesign can wait, and it may turn out you only needed a phone number moved to the top and a form shortened.

And do not delete the informational content in a panic. It is usually not hurting you. It is just not helping. Stop adding to it, and redirect the writing budget. The one exception: if you actually sell the thing the article is about, such as a subscription plan, that content can convert. We have a client on a subscription plan, so for them that topic is a real product. Many contractors are not in that position.

One more thing worth saying, because word count comes up a lot in SEO advice: Google says it has no preferred word count. Padding a service page to hit a number does nothing for the phone. Answering the hire query and asking for the call does.

What would we do this week?

In order, with the cheapest fix first:

  1. Test your own form and phone number from a cell phone on cellular data. Fix anything that fails. Thirty minutes.
  2. Pull 90 days of queries from Search Console and label them hire, how-to, unrelated. Count the clicks in each column. One hour.
  3. If hire clicks are low, list the five services you most want to sell and check whether each one has its own page that names the service and your city in the title. Any that do not get written this month. Stop writing how-to posts until that is done.
  4. If hire clicks are healthy, open your top three service pages on a phone. Move the number to the top, cut the form to three fields, add one specific proof point about your work, and replace any stock or AI photo with a real one.
  5. Install call tracking and a session recorder on those three pages so next week you can see whether anything changed.

What should you do today?

Open Search Console, sort your last 90 days of queries by clicks, and write hire, how-to or unrelated next to each of the top 20. That one list tells you which of the two leaks you have, and it is the same Search Console data we looked at for the client who reported traffic up 30 percent.