How Do You Measure SEO and AI Visibility?

ClearFlow has done the work. Is it helping?

Think back to ClearFlow Plumbing, the fictional business in Austin we've been following. We've improved its service pages, made its website easier to use, and explored how customers might discover it through Google, Maps, other websites, and AI answers.

Now imagine ClearFlow's owner opens the business laptop after a busy week. The website looks better, but the owner still has a fair question: “Are more people actually finding us, and is that leading to useful enquiries?”

You can't answer that question by looking at the website alone. You need some evidence about what happens when people search, visit, and get in touch. That is what measuring SEO means: checking useful information over time so you can understand what's happening and decide what to investigate or improve next.

But the numbers need care. Seeing a business online isn't the same as clicking its website, and clicking the website isn't the same as becoming a customer. We learned that in Lesson 2. Now we'll learn how to recognize the numbers behind those differences.

First, did ClearFlow appear in Google Search?

Imagine a homeowner in Austin searches for “water heater repair near me.” ClearFlow's service page is among the unpaid results. The person sees its link but decides not to open it.

Google may count that appearance as an impression. An impression means a link to the website appeared in a Google result under that report's counting rules. It does not mean someone visited the website, and it doesn't necessarily mean a different person each time.

Now picture a second search. This time, someone clicks ClearFlow's organic result and opens its page. Google can count that action as a click.

A free tool called Google Search Console helps website owners see figures like these for their Google search results. For our made-up reporting period, let's say ClearFlow's water heater repair page recorded 1,600 impressions and 64 clicks.

That doesn't mean 1,600 people became customers, or that 64 different people called. It tells us the page appeared in Google results and received clicks. One useful question now is whether the searches were relevant to the service ClearFlow offers.

You may also see click-through rate, or CTR. That's the share of impressions that led to a click. In our example, 64 clicks from 1,600 impressions is 4%. It's one clue, not a grade for the whole business. Google Search Console also shows average position, but that number is an average across changing searches, not a promise that everyone sees a page in exactly the same spot.

What happened after someone clicked?

A click from Google can bring someone to ClearFlow's website. The business might use another tool, such as Google Analytics, to understand website activity. It can report visits from organic search, meaning visits associated with unpaid search rather than paid advertising.

In website analytics, a session is a period of activity by a visitor. It's a useful way to talk about visits, but it is not the same measurement as a Google Search Console click. The two tools have different ways of counting and may not show matching totals.

Let's say ClearFlow's website analytics shows 58 organic search sessions during the same fictional period. We shouldn't try to turn that into 58 customers or assume exactly which of the 64 clicks produced those sessions. These figures come from different reporting systems.

Now imagine ClearFlow's website records five activations of its Call link and three contact-form submissions. Those actions may be tracked, if the website has been set up to measure them. But a tap on Call doesn't prove a phone conversation happened, and a form submission isn't automatically a genuine new sales opportunity.

To understand real enquiries, ClearFlow also needs to check its received messages, answered calls, and customer records. Someone might even call directly from a Maps listing, without visiting the website. Website traffic, contact actions, and confirmed enquiries are related, but they're not interchangeable.

And can we measure when AI makes ClearFlow visible?

Remember Lesson 4: AI can mention a business, cite a webpage as a source, or recommend a business as an option. Those aren't the same outcome, so measuring them isn't one simple number either.

There are some useful clues. For example, Google Search Console now includes a Generative AI performance report. For eligible websites with enough data, it can show impressions when links to their pages appear in Google's AI Overviews or AI Mode. That is evidence of visibility in those Google features. It is not a count of every AI recommendation on the internet. It is a view of Google visibility, not an extra total to add to your other Google figures.

A website's analytics may also show visits coming from an AI service when someone follows a link. That depends on source information being available to the analytics tool. But reading an AI answer without clicking a link leaves no website visit to count. And the traffic figure alone doesn't show every time a business was mentioned or recommended.

ClearFlow could also occasionally check whether a specific AI answer mentions its services correctly. A check like that is just a snapshot: answers can change with the question, location, provider, and time. It's not proof that the business is always recommended.

Why does this matter? Because without these distinctions, a business might mistake one AI mention for dozens of customers or mistake a lack of website clicks for a complete lack of visibility.

How can ClearFlow use the numbers to make a better decision?

Imagine the owner compares two equal-length periods. Organic search visits went up, but the business received about the same number of confirmed enquiries from all sources. Is that good news or bad news?

It's a reason to ask another question. Perhaps people are finding a helpful article but aren't looking for a repair service yet. Perhaps the service page could explain the next step more clearly. Maybe more people are searching this month for reasons unrelated to ClearFlow's changes.

The figures don't tell us which explanation is correct. Instead of celebrating a bigger number or changing the whole website in a hurry, the owner can look at the searches, pages, and actual customer enquiries to understand what happened.

This is why we measure SEO. Not to collect impressive-looking charts, but to decide what needs attention next. Blue Kite SEO can help organize a strategy, show supported search and website activity, and record completed work. Where it tracks a phone or email link click, that still isn't evidence of a completed conversation or booked job. The owner remains the person who knows whether an enquiry was real and useful.