Six months into an SEO engagement for a heating and cooling company, every report I sent looked good. Organic traffic was up 40 percent. We were tracking 200 high-intent keywords, the kind of searches people type when the furnace quits in January, and most of them had moved from page two of Google to page one. Average position went from 15 to 8.
Then I counted the leads. One phone call. Two form submissions. In six months.
Not one a week. One. I have been doing this for twelve years and I still had to read that number twice.
The report looked great
If you have ever paid for SEO, you have seen a version of this chart. Clicks going up, impressions going up, average position getting better. It is a real result and it took real work to get there. It is also the wrong chart to run a business by, and it took me longer than I would like to admit to see that.
Keyword positions, month 1 200 tracked
Keyword positions, month 6 200 tracked
A dashboard like this is built to answer one question: is the SEO working? It is not built to answer whether the website is working. I had assumed those were the same question. They are not even close.
So where was everybody going?
Once I stopped trusting the top line, I pulled six months of data on every page of the site. Twenty-five pages, including a handful of blog posts that were getting no traffic at all. Then I connected three things that normally live in three separate tabs: what people searched for, which page they landed on, and what they did after that.
Two things jumped out.
About 70 percent of the searches were research. People wanting to know why their AC was blowing warm air, or what a new furnace costs. The other 30 percent were ready to hire someone. That is still a lot of people, and almost none of them called.
And most visitors landed on the homepage or one of a few service pages, then went nowhere. Very few clicked into the navigation. They arrived, looked, and left.
| Landing page | Sessions | Engagement rate | Pages / session | Calls | Forms |
|---|---|---|---|---|---|
| / | 3,412 | 41% | 1.3 | 0 | 1 |
| /ac-repair/ | 1,760 | 47% | 1.2 | 1 | 0 |
| /furnace-repair/ | 1,238 | 44% | 1.2 | 0 | 0 |
| /ac-installation/ | 862 | 39% | 1.4 | 0 | 1 |
| /heating-installation/ | 674 | 38% | 1.3 | 0 | 0 |
| /contact/ | 447 | 52% | 1.1 | 0 | 0 |
| /about/ | 318 | 35% | 1.5 | 0 | 0 |
| /blog/… (6 posts) | 31 | 22% | 1.0 | 0 | 0 |
| All organic landing pages | 8,742 | 43% | 1.3 | 1 | 2 |
The site was a brochure
Nothing was broken in the SEO sense. The problem was everything that happened after the click. Here is the homepage those visitors were landing on, with the things that were quietly costing the client work.
- 1The phone number was the smallest text in the header. On a phone it was not even tappable.
- 2One vague “Learn More” button. It did not say what you would learn or where it would take you.
- 3Six menu items and not one for the thing people were searching for. If your AC was out, you went hunting.
- 4The first thing under the banner was a company history. Nobody with a broken furnace is reading that.
- 5The only real ask was a form at the bottom of the page, with no reason to fill it out and no promise about what happens next.
Every step between a visitor and a phone call is a place to lose them. This site had four or five of those steps. The ready-to-hire visitors were doing exactly what you would expect. They looked, did not see an obvious way to get help, and went back to Google to try the next company.
One week, four people
We did not build a new page, add a keyword or touch anything that would affect rankings. We changed what the pages asked visitors to do.
- 1A call button with the number in it, in the header on every page. Tap it and the phone dials.
- 2The banner says what the company does and that they can come today. Two buttons: call now, or book online.
- 3The four main services right in the banner. One click from landing to the page that matches your problem.
- 4Reviews moved up under the banner, next to the ask, instead of living on a page nobody visited.
- 5A call bar that stays at the bottom of the screen on mobile. The contact page got the same treatment: a shorter form and a real reason to use it.
What happened next
Over the next four months the site produced 42 phone calls and 18 form submissions. That is against one call and two forms in the six months before.
Rankings barely moved. Traffic stayed about the same. The only thing that changed was the website.
To be fair to the client, he did not throw a party. He said he wanted to see what the next six months looked like, which was reasonable. He had been patient for six months already.
What did change was the conversation. We started meeting weekly, and the meetings were about conversion: which pages produced calls, what we were testing next, what else we could do to make the site convert better. Traffic and rankings came up less and less. After a while they stopped coming up at all. He did not care about them anymore, and he was right not to.
What I do differently now
The SEO was good. I would do the keyword work the same way again. Where I went wrong was letting the SEO report stand in for the business result for six months, because the report looked healthy and nobody was asking me a harder question.
Now the first thing on any report I send is leads by page. Traffic and rankings are still in there. They are further down.