How to tell whether your SEO consultant is working, without understanding SEO
You do not need to learn SEO in order to supervise it. Five questions do most of the work, and none of them require technical knowledge. What did they decide not to do. Who raised the bad news. And what would have to happen for them to tell you to stop.
The awkward thing about hiring an SEO consultant is that you are buying something you cannot evaluate directly. You can taste the food. You can sit in the chair the carpenter made. You cannot look at a month of SEO work and know whether it was good.
So most owners fall back on the two proxies available to them: is traffic up, and does the report look thorough. Both are unreliable. Traffic moves for reasons nobody controls, and a thorough looking report is the easiest artefact in this industry to produce.
You do not need to learn the discipline. You need five questions.
1. Can they tell you what they decided not to do, and why?
This is the highest signal question in the list, and almost nobody asks it.
Real strategic work is mostly the elimination of options. Somebody who has genuinely thought about your business can tell you, without preparation, which obvious moves they rejected and what reasoning ruled them out. Somebody executing a checklist cannot, because no options were ever considered.
What should worry you: an answer that lists everything they are doing. That is a menu, not a decision.
2. Does the reporting connect to your business, or only to the tool?
Impressions, positions and traffic are inputs. They are worth tracking. They are not results, and a report made entirely of them is a report about the software rather than about your company.
A useful monthly report tells you: what moved, what did not, what it means for your business, and what happens next. Four things, in plain language. If you have to translate the report before you can act on it, it is not finished.
What should worry you: a dashboard export with a paragraph glued on top.
3. When something did not work, did you hear it from them first?
Every engagement has months where the plan does not survive contact with reality. That is normal and it is not a scandal.
What matters is who raises it. If bad news always arrives because you noticed and asked, you do not have a partner, you have a supplier managing you. A consultant whose reporting only ever contains progress is either extraordinarily lucky or editing.
4. Do they know what your business actually sells, and to whom?
Ask your consultant to describe your ideal client. Not your audience. Your best client, the kind you want more of, and why that person chooses you over the alternative.
If the answer is vague, the work above it is guesswork, however technically competent it is. Search work is the practice of anticipating decisions. You cannot anticipate a decision you cannot describe.
What should worry you: an answer that describes a demographic rather than a decision.
5. What would they have to see to tell you to stop?
A good consultant has a threshold in mind, even if it is rough, and can say it out loud. It might be: if this has not moved by month five, the problem is not visibility, it is demand, and we should say so.
Somebody with no such threshold has no mechanism for ending an engagement that is not working, which means the engagement ends when you get tired of it rather than when the evidence says it should.
The five questions on one page
If you keep one thing from this, keep the table. It is meant to be read in a meeting.
| Ask this | A good answer sounds like | Worry if you hear |
|---|---|---|
| What did you decide not to do? | Two or three rejected options, with the reasoning | A list of everything they are doing |
| What did the last report mean for my business? | What moved, what did not, what it means, what is next | Impressions and positions, with no translation |
| What has not worked recently? | A specific thing they raised before you asked | An unbroken run of good news |
| Describe my best client and why they choose me | A decision, in the buyer’s own terms | A demographic, or your own website read back |
| What would make you tell me to stop? | A rough threshold and a date | Nothing, or a change of subject |
The three answers that should end the conversation
- A guaranteed ranking or a guaranteed AI answer. Nobody controls those systems. A guarantee is either ignorance or a lie, and both are disqualifying.
- Volume as the plan. If the strategy is a number of articles per month and the reason is that more is better, there is no strategy, only a production schedule.
- Access you cannot have. If your analytics, your search data or your content live somewhere only they can reach, you are not a client, you are a hostage. Ownership is not a detail to sort out later.
What a reasonable first ninety days looks like
It helps to have a rough shape in your head, so that you can tell drift from normal pace.
The first month is mostly diagnosis and foundations: understanding the business, fixing what is structurally broken, and getting measurement in place so the following months mean something. Very little of this is visible from the outside, and that is not a warning sign, it is the sequence.
The second and third months are where the plan meets reality. Pages get rebuilt or rewritten, structure gets put in, and the first movement usually shows up in technical and visibility measures rather than in enquiries. If somebody promised you enquiries in month two, that promise was made to close the sale.
By the end of the third month you should be able to answer three questions without help: what did we decide to focus on, what has been done, and what are we watching to know whether it worked. If you cannot answer those, the problem is not the pace of the work. It is that nobody has been explaining it to you, which is a different failure and an easier one to fix.
Two things that look like problems and are not
First, slow early months. Organic growth compounds, which means the first stretch looks flat while the foundations go in. Movement in the first two to three months is usually technical and structural rather than commercial. That is expected.
Second, boring work. A lot of what actually moves the needle is unglamorous: fixing structure, clarifying pages, making the same business look like the same business everywhere it appears. If your consultant seems to be doing dull things carefully, that is frequently a good sign rather than a bad one.
Questions owners ask about this
How long should it take before I see anything?
Expect the first two to three months to move technical and structural measures rather than enquiries. Anyone who promised commercial results in month two was closing a sale, not describing the work.
My traffic dropped. Should I end the engagement?
Not on that alone. Traffic moves for reasons nobody controls, including seasonality and changes to the search engines themselves. The question that matters is whether your consultant explained the drop before you noticed it, and what they proposed to do.
Is it reasonable to get a second opinion on work already in progress?
Yes, and it is more common than people assume. A consultant confident in their reasoning has nothing to lose from another practitioner reading it.
What if they will not answer these five questions?
Refusing to answer is itself an answer. None of the five requires disclosing method or trade knowledge, so an evasion is about the thinking behind the work rather than about protecting it.
How much should a small business pay for this?
There is no single right number, and any figure quoted without scope attached is meaningless. What you can insist on is that the price and what it includes are both stated in writing before you commit. Ours are on the pricing page.
The one sentence version
You are not evaluating whether they know SEO. You are evaluating whether they are thinking about your business, and whether they will tell you the truth when it is inconvenient. Those two things you are entirely qualified to judge.
If you want a second opinion on work already in progress, that is a legitimate reason to book a call, and it happens more often than people expect. How this practice works, who does the work and what gets reported is on the services page and in more detail on the FAQ page. Two engagements are described in full on the work page, including what did not work. The call itself is thirty minutes, and it is fine to use it to ask about somebody else's work.
