Are SEO consultants worth it? Do the arithmetic

I searched this in Canada on 21 September 2026 and read page one. The first organic result was a Reddit thread from a small business owner. Below it Google placed a whole “discussions and forums” block: another Reddit thread with more than 150 replies, and two Quora threads. Two of the top four slots went to buyers talking to each other rather than to anyone selling the service.

Then came the vendors. I took the sentence Google chose to highlight from each:

• “The Verdict, Is SEO Worth It in 2026. Yes.”
• “SEO is worth it for any business where your target audience is searching online”
• “These experts can bring in more organic traffic”
• “one of the highest-ROI channels”
• “You have the budget to do so”

Every page published by someone who sells search marketing concluded yes. Not one described a business that should not buy it. The last of those is listed as a reason to hire somebody, which is an interesting definition of a reason.

One page stood out because it was the only neutral source on the screen: Google’s own documentation, “Do You Need an SEO?”, which is written as a genuine question and spends much of its length on how to avoid being ripped off.

My favourite was a title promising “63% of Businesses Think So”. That is a popularity measurement. What a majority of businesses believe about their own marketing spend is not evidence that the spend paid, and if it were, no marketing channel would ever have been abandoned.

I sell this service. So rather than give you my opinion, here is the sum, and then the cases where it comes out against me.

The sum

You need four numbers. You already have two of them.

1. What a customer is worth to you over twelve months. Not one transaction. The full year, including repeat business. Most owners know this within about twenty per cent.
2. What share of enquiries you convert. Phone calls, forms, walk-ins. Most service businesses land somewhere between a third and two thirds.
3. The monthly fee.
4. Whether demand exists in your market. This one is measurable before you spend anything, and it is the one people skip.

Break-even enquiries per month = fee divided by (customer value times close rate).

That is the whole thing. It tells you how many extra enquiries a month the work has to produce before you are level, and you can decide for yourself whether that number is plausible in your market.

Two worked cases, same fee, opposite answers

Take a $750 a month retainer, which is the middle of what I charge and roughly the middle of the Canadian market for a small business.

A trade. A plumbing customer is worth about $1,200 over a year once you count the callback and the eventual bigger job. You close about half of the calls that reach you. Each enquiry is therefore worth about $600, and break-even is:

$750 divided by $600 = 1.25 extra enquiries a month

Slightly more than one. If the work cannot produce one extra call a month, something has gone badly wrong. The decision here is easy, and it stays easy even if my estimate of customer value is out by half.

A cafe. A regular is worth perhaps $60 a year, and there is no meaningful close rate because people either walk in or they do not. Break-even is:

$750 divided by $60 = 12.5 extra customers a month, every month, attributable to search

Possible, but you can no longer tell whether it happened. You have crossed from a number you can verify into a number you have to believe. The fee has not changed. The business has.

The variable that decides this is customer value, not the quality of the SEO. That is why “is it worth it” has no general answer, and why every page that gives one is answering a different question than the one you asked.

What the delta actually looks like

For scale, one published example. A laundromat’s Google Business Profile sat at about eleven calls a month for three years. It reached twenty-six in July 2026 after profile work, and the full work log is published including what did not move.

That is roughly fifteen extra calls a month. Against the trade break-even above it clears the fee many times over. Against the cafe break-even it clears too.

Now the caveat, which matters more than the number. That profile had three years of no maintenance behind it, which is a large gap and an unusually easy one to close. A business already doing the basics properly will see a smaller and slower move. One account is an anecdote, and I am showing it to you as a sense of scale, not as a forecast.

The cases where the answer is no

I turn work down on all four of these, and it costs me the sale every time.

There is no demand to capture. This is the most common one and nobody checks it first. Outside the five largest Canadian cities, a whole category in a single town often runs ten to a hundred and thirty searches a month. If your category in your town is forty searches a month, then ranking first wins you a share of forty. Do that arithmetic before you spend, because no amount of skill creates demand that is not there.

The customer value is too low, as in the cafe. Search may still help, but it will not be provable, and an unprovable monthly fee turns into an argument in month six.

Search is not your constraint. If enquiries already reach you and die, because the page is slow, the phone number is buried, or nobody answers after five, then more traffic multiplies a leak. Fix the leak first. It is cheaper and the result is immediate.

The budget is below about $500 a month. At $300 a month a supplier can afford you roughly two hours, and two hours a month cannot run a search programme. It can produce a report, which is why the cheap end of this market is full of reports. I have written about what the market charges in more detail.

What “worth it” should mean in the agreement

If you do go ahead, make the question answerable rather than arguable.

• A baseline measured before any work starts. Otherwise improvement gets compared to a memory.
• The break-even number written down at the start, so both of you know what success is before anyone is defensive about it.
• A named month where you decide. Around month four the work is either paying or it is not.
• Accounts in your name, so leaving costs you an adviser and not your data.

A supplier who will not put a break-even number on paper either has not done this sum for you or does not want you doing it.

The honest summary

Sometimes. It depends on one number you already know and one you can measure in an afternoon. For a trade with a customer worth four figures, the bar is about one extra job a month and the answer is usually yes. For a low-value transaction in a thin market, the answer is often no, and anybody telling you otherwise has not asked what a customer is worth to you.

If you want the demand side measured rather than guessed, that is the first thing the audit does, and what SEO consulting covers and costs is on the page rather than behind a quote. You may also want the consultant against agency comparison, which is the same arithmetic applied to who does the work.

Method

Search results were recorded for Canada in English on 21 September 2026 to a depth of twenty and they move. The highlighted sentences are the fragments Google emphasised in each result on that date, not my summary of those pages. Customer values and close rates in the worked cases are illustrative figures chosen to be realistic, not measurements of a real business; the point is the structure of the sum, and you should run it with your own numbers. The search volume range for Canadian towns comes from Google Ads data for Canada pulled the same day. The laundromat figures are from a published work log.