SEO Performance Verification

This question usually arrives four to six months into an engagement, from an owner who is receiving reports they cannot read and who has a quiet suspicion that nothing is happening.

The suspicion is sometimes wrong. SEO is genuinely slow, and a competent engagement can look like nothing for a while. But you should not have to take anyone's word for it either way, and almost every part of this is checkable yourself.

First: the one measure that matters

Tracked leads, trending, over a rolling three-month window.

Not traffic. Not impressions. Not rankings. Leads — counted with a tracking number and a form that records where the submission came from, compared quarter over quarter rather than month to month so that seasonality does not fool you in either direction.

Everything else in a report is either a leading indicator of that number or a distraction from it. Impressions can triple while your phone stays silent. Rankings can improve in a place none of your customers are standing. Traffic can rise from people who will never buy anything.

If you take nothing else from this: the question is not "did the numbers go up," it is "did more of the right people contact me."

The twelve checks

Run these yourself. Most take minutes.

What is actually happening

1. Is there a dated work log, at task level? Ask for the last ninety days of work, dated, task by task. "Content optimization" is a category, not a task. "Rewrote the water heater page title and meta description on 14 March" is a task. An agency doing real work already has this and can send it today. One that needs a week is assembling it for the first time.

2. Has anything on your website actually changed? Look at your own site. Are there pages that did not exist six months ago? Have titles changed? If nothing on the site has moved in half a year, very little has been done to it.

3. Is your Google Business Profile different? Open it. Check the primary category, the service area, the hours, the services list. If these are still exactly as you left them, the fastest and cheapest work in local SEO has not been done.

Whether it is working

4. Are tracked leads up quarter over quarter? The main event. If there is no tracking number and no form source recording, this cannot be answered at all — which is itself the finding.

5. Is position measured across a grid, or from one point? Distance is a ranking factor computed from the searcher's location, so a single screenshot is a result from one place, usually your own address. Ask which. A grid produces a sentence — winning Green Valley, losing Anthem — that a single point cannot.

6. Is the traffic commercial, and is it local? Traffic to your service pages, from your area, is worth something. Traffic to an explainer article from another state is not. Check the split before celebrating a total.

7. Are impressions rising while clicks are flat? This pattern usually means you are appearing for searches you do not deserve, or appearing in positions nobody scrolls to. It is frequently presented as progress. It is not, by itself.

8. Are you appearing but not being chosen? If you rank well and the phone is quiet, the problem may not be ranking at all. Look at your profile the way a customer would: review count and recency against the competitors listed beside you, photos, hours. Being seen and not picked is a different problem with a different fix.

Whether you are protected

9. Do you own your domain, hosting, and profile? Check that your own account — not the agency's — has owner-level access to the domain registration, hosting, the Google Business Profile, Analytics and Search Console. Do this now rather than at the point you want to leave.

10. Who owns the tracking number? If the agency owns it, leaving means losing the number and its call history. This is the asset most often lost in an agency change and the one that hurts most.

11. Has anyone bought links on your behalf? Ask directly and in writing. Purchased links are a liability that sits on your domain long after the agency has gone, and you are the one who inherits it.

12. Is anything in the reporting unverifiable? If a number in your report cannot be traced to a source you can independently open, treat it as decoration until proven otherwise.

How to run the grid check yourself, roughly

You do not need a tool to get a usable version of this, and doing it once yourself changes how you read every ranking report afterwards.

Pick one search that produces paying customers. Not the highest-volume term — the one that actually converts.

Search it from four or five places across the city, in a private window. Green Valley, Anthem, somewhere near Water Street, and wherever most of your customers actually are. If you cannot travel, ask a few customers or friends in those areas to search it and screenshot what they see.

Write down your position from each point. That is it. Five numbers.

What you will usually find is a spread rather than a number — strong near your own location, weaker as you move away, absent past a certain distance. That spread is your real ranking, and it is the thing a single screenshot conceals.

Two things follow from it. You now know whether your agency's reported position is representative or cherry-picked. And you know which parts of the city are worth contesting, which is a strategy question a single number cannot answer.

When the answer is "it is working, slowly"

Not every disappointing engagement is a bad one, and it is worth being able to recognise the difference before you act.

Signs it is genuinely working and simply early: leads are flat but impressions and map position are both moving; the work log is full and specific; you can see changes on your own site and profile; the agency has told you about something that did not go to plan. Local SEO takes 60 to 90 days for map movement and six to twelve months for competitive organic terms, and a category with entrenched competitors sits at the slow end of both.

Signs it is not: nothing has changed on your site or profile in six months; the report leads with impressions; no tracked lead count exists; the plan for month six is identical to the plan for month one.

The distinction is not speed. It is whether the engagement responds to evidence. An agency running an unchanged plan while the data says it is not working has stopped doing SEO and started doing invoicing, however busy it looks.

Four questions worth asking outright

If you only do one thing, send these four in an email and read how they are answered. The speed and specificity of the reply tells you nearly everything.

1. What did you do last week, specifically?
2. How many tracked leads did I get last month, and where did they come from?
3. Do you measure map position from one location or a grid of points?
4. Which accounts are in my name, and which are in yours?

An agency doing real work answers all four in a single reply, quickly, without discomfort. The answers may be imperfect — real work has bad months — but they exist.

An agency doing nothing produces delay, jargon, a rankings chart, or a meeting request. That substitution is the tell.

What to do if the answers are bad

Do not fire anyone the same afternoon. Do this in order.

Secure access first. Domain, hosting, profile, Analytics, Search Console, tracking number. Confirm your own account has owner-level rights to each. Access is dramatically harder to recover after a relationship ends badly.

Then export what you can. Historic Analytics data, call records, and any content you paid for.

Then have the conversation. Give them the four questions and a deadline. Some engagements recover at this point, because the problem was communication rather than work. Many do not.

If you are already past this point, recovering from a bad SEO agency covers what to check for damage and what order to undo it in. And if you would rather not be in this position again, The One Number Method is our answer to the whole problem — leads counted first, every task logged and dated, position measured across a grid, and a month-to-month agreement so the reporting has to stay honest.