Local SEO

How do you know if your SEO is actually working?

Todd, Founder — Hitman MarketingPublished July 24, 2026

You know SEO is working when enquiries from search go up — calls, forms, direction requests, booked jobs — not when impressions or a keyword position do. For a local business the honest test is a rising trend in contacts attributable to search over months, judged against a baseline recorded before the work started.

Key takeaways

The only outcome that counts is enquiries you can trace to search. Everything else is a leading indicator at best.
Record a baseline before the engagement starts, or nothing afterwards can be evaluated.
Impressions, keyword counts, and 'positions improved' are the three most common vanity metrics in local SEO reporting.
Six months is the fair judging window for competitive local terms; profile and listing work should show movement in weeks.
If your provider cannot tell you which pages produced enquiries, the reporting is decoration.

What is the first number to check when judging SEO?

Check contacts from search: calls, form submissions, messages, and direction requests that can be traced to organic search or your Google Business Profile. Everything else in an SEO report is a step on the way to that number, and a report full of intermediate metrics with no contact trend is describing activity rather than results.

For a local service business the profile is usually the larger half of that total, and it reports its own version — calls, direction requests, website clicks, and how people found the listing. Those figures are available to you directly, without any agency intermediary, which makes them the hardest number in the whole conversation to spin.

Then take one step further, to booked jobs. Enquiries are not revenue, and a rise in low-quality enquiries is a real failure mode of aggressive SEO. The provider is accountable for the enquiries; you are accountable for knowing which of them turned into work.

Which SEO metrics are vanity metrics for a local business?

Three recur in local SEO reporting and mean almost nothing on their own: total impressions, the number of keywords a site 'ranks for', and a count of positions improved. Each can rise substantially while enquiries stay flat, because none of them distinguishes a buying search from an irrelevant one.

None of these are lies, and each has a legitimate diagnostic use. They become vanity metrics when they are reported instead of outcomes rather than alongside them. A report that leads with impressions and never mentions calls has chosen which story to tell.

Metric	Why it appears in reports	What to ask instead
Impressions	Rises easily and looks like growth	Did clicks and contacts rise with it?
Keywords ranked for	A large number that always goes up	How many of those terms would a buyer type?
Positions improved	Counts movement from 87 to 61 as a win	How many terms moved onto page one or into the pack?
Average position	One tidy figure across everything	What is the position on the handful of terms that convert?
Backlinks acquired	Countable and easy to invoice against	Did the pages those links point at produce enquiries?

How long should you give local SEO before judging it?

Give listing, profile, review, and technical work a few weeks — those affect people already searching for you and show up fast. Give competitive rankings and content six months, which is roughly when local search work produces results either party can defend. Judging competitive terms at month two produces a wrong answer.

The two halves are worth tracking separately for exactly this reason. If nothing has moved on the fast half by week six — profile interactions flat, listings still inconsistent, technical errors unfixed — that is a legitimate early warning, and it does not require waiting six months to act on.

What the six months buys is a fair test of the slow half. Content has to be published, indexed, and gather signals; category-level competition does not turn over in weeks. A provider asking for that window is being reasonable; a provider asking for two years before any evaluation is not.

What should a monthly SEO report contain?

A monthly report should contain four things: what was delivered, what moved, what it produced, and what happens next. Deliverables as counts rather than adjectives, ranking and visibility movement on terms that matter, contacts from search, and a prioritised plan for the coming month.

One thing a good report also contains is bad news. A month where a page dropped, a listing broke, or a planned deliverable slipped should appear in the report, because a document that only ever contains good news stops being information.

If your work includes AI search visibility, that is measured differently — share of voice across a stable prompt panel, treated as a trend, never a single query result. The mechanics and the honest caveats about under-counted AI referral traffic have their own page.

Deliverables completed, as countable items: pages published, citations built or corrected, posts, reviews handled, technical fixes.
Movement on a fixed set of buyer-intent terms — the same set every month, so the trend means something.
Contacts from search and from the Google Business Profile, with the previous months alongside for comparison.
Listing health across every source that matters, not only Google.
What is planned next month, and why that rather than something else.

How do you connect a phone call back to search?

Connect calls to search with tracked numbers and consistent instrumentation: a dedicated number on the website that forwards to the main line, the Google Business Profile's own call reporting, and form and click-to-call events recorded in analytics. Without that, attribution is opinion and every channel claims the same call.

Two cautions specific to local search. First, changing the phone number displayed in your listings to a tracking number is how businesses break their own citation consistency — the safe pattern is a tracking number on the website with the real number kept identical everywhere else. Second, ask new customers how they found you and write it down; it is imperfect and it catches what analytics never will.

Also accept a permanent ceiling on precision. A meaningful share of visits arrive without a referrer and land in analytics as direct traffic no matter how the setup is configured, so measured search performance is a floor rather than a count. Any provider who presents attribution as exact is overselling the tooling.

What does it look like when SEO genuinely is not working?

SEO that is not working looks like this: months of reports with no contact trend, deliverables described in adjectives rather than counts, no baseline anyone recorded, ranking terms that change whenever the previous ones stop improving, and a provider who cannot say which pages produced enquiries. Any two of those is a conversation; all five is an answer.

Before ending an engagement, run one check on your own: is the work visibly there? Look at the site for the pages that were supposedly published. Look at the profile for the posts and the review replies. Search a few of your own buying terms from a phone away from your office. Reports can be arranged; the artifacts either exist or they do not.

If you do move on, take the accounts with you first. The domain, hosting, Google Business Profile, analytics, and ad accounts should be in your name with the provider holding delegated access — and if they are not, sort that out before giving notice rather than after.

Common questions

We have no baseline — is it too late to measure?: No. Start one today: this month's profile interactions, search contacts, and position on your five real buying terms. You lose the ability to judge what has already happened, but you gain the ability to judge everything from here. Most owners who feel stuck are stuck precisely because nobody wrote down the starting point.
Our rankings improved but the phone did not ring. What happened?: Usually one of three things: the terms that improved are not terms buyers use, the ranking gain happened outside the map pack where local demand actually lives, or the site is ranking and failing to convert. Check which terms moved and whether the traffic reaches a page that asks for the call. It is a diagnosable problem, not a mystery.
Should we get a second opinion on our SEO?: If the doubt has lasted more than a quarter, yes. An independent audit costs a fraction of a year's retainer and gives you a prioritised list you can hand to your current provider, an in-house person, or anyone else. A provider confident in their work is not threatened by one; a defensive reaction is itself information.
Related: The AI Visibility Teardown — $497, credited if you continue

Find out whether your territory is open

One contract per industry per city. If yours is open you can execute at the published price today; if a competitor already holds it, the nearest open market is the one to look at.

See open contracts Request a territory — $497

The survey is credited in full against the contract if your territory opens and you take it.