Is your SEO attracting
the right customers?

Define a suitable inquiry first

For a selective service business, a suitable inquiry might involve the right service, a workable project scope, a relevant location and a realistic decision timeline. Define your own criteria before looking at the results. A form submission alone cannot tell you whether someone is a good fit.

At OBVS, we would start the measurement plan with those criteria and the action that begins the relationship. Keep newsletter interest separate from a project request. Keep a project request separate from a qualified opportunity and a client.

This is our business measurement approach, not a set of search ranking factors. Its purpose is to make better decisions about where to invest attention.

Keep four layers of evidence separate

Visibility

Where pages appear in search, and which queries or surfaces expose them.

Visits

Which visits reach the site, including identifiable search and AI referrals.

Actions

Relevant events such as a successful inquiry submission.

Business fit

Which inquiries become suitable opportunities and customers.

Search Console reports search performance, including impressions and clicks. These measures answer different questions from a website’s lead records. Google: Search Console performance report ↗

In Analytics, a key event marks an action important to your business. It does not independently establish lead quality. Google also documents separate lead-generation and qualification events for implementations that connect those stages. Google Analytics: About key events ↗ Google Analytics: Recommended lead events ↗

Test the path before trusting the report

Choose the intended action, complete it yourself and confirm that the correct event is recorded once. Test a failed submission too. A click on “Send” should not be mistaken for a successfully received inquiry.

Record the page, event definition, counting method, responsible person and review date. Keep personal contact details in the appropriate customer system; use the analytics system for measurement rather than storing the contents of a customer’s message.

Check a sample against actual inquiries. Remove internal tests and obvious spam from your business assessment. Document tracking changes so an increase caused by a new event setup is not presented as new demand.

Measure AI visibility without inventing certainty

Google has introduced dedicated generative AI performance reports in Search Console. Use the dimensions available in your account and read the report definitions before combining them with other totals. A reported appearance is not the same thing as a visit or a qualified lead. Google: Generative AI performance reports ↗

Record identifiable AI referral traffic separately where your analytics exposes it. Also ask new inquiries how they found you, with an optional free-text answer. Neither method captures every interaction; someone may read an answer and return later through another route.

You can maintain a dated sample of relevant prompts to inspect how your brand is described. Record the tool, prompt, date and cited URLs. Treat it as an observation sample, not a universal visibility score: responses can differ by context and time. Never claim that a mention in a private chat proves a ranking improvement.

Read the result in context

In this example, traffic fell 20%, while suitable inquiries doubled. The proportion of inquiries judged suitable rose from 20% to 50%. That may be a better business result if opportunity value and delivery fit remain healthy.

It does not prove SEO caused the improvement. The small sample, a different campaign, seasonality, a change in qualification or the sales cycle could influence the result. Compare like-for-like periods, document concurrent changes and inspect the underlying inquiries.

Do not force a winning narrative. If visibility grows but suitable inquiries do not, investigate intent, offer clarity, audience and the next step. If qualified opportunities increase but sales do not, review the sales process and offer fit as well.

Follow the opportunity through to revenue

Our recommended scorecard follows suitable inquiries into qualified opportunities, won customers and revenue. Record the originating page or reported discovery source when you know it, along with dates and your attribution assumptions. Keep pipeline estimates separate from revenue actually earned.

For a service business, compare qualification rate, close rate, sales-cycle length, customer revenue and delivery costs. For a store, consider purchases, refunds, repeat purchases and margin. Revenue growth that brings poor-fit projects or unsustainable delivery costs may be the wrong growth for your business.

Google Analytics documents distinct events for leads, qualified leads, converted leads, purchases and refunds. These require appropriate implementation; this article does not install tracking or connect your sales records. Google Analytics: Recommended events ↗

Do not credit every sale after an SEO change to that change. Reconcile your analytics with customer records, consider other channels and allow for the full sales cycle. Helpful educational and support content can also reduce uncertainty or improve retention without producing an immediate sale.

Use the review to choose the next action

Maintain a simple review record: what changed, what evidence moved, what remains uncertain and what you will do next. Select a period long enough for your normal demand and sales cycle; there is no universal 30-day SEO verdict.

Compare search queries and landing pages against your desired work. Read the questions suitable prospects ask. Improve the page that is failing a valuable decision, not automatically the one with the fewest visits.

Use the prioritization guide to choose the next fix. Use the content blueprint when the missing piece is a useful answer. Each review should lead to a reasoned decision, even when that decision is to collect more evidence.