Nobody senior has looked at your account

There is a version of a failing SEO account where nothing obvious is wrong. Work is definitely happening. Pages get published, tasks get ticked, the checklist advances, and this month's report has more on it than last month's. What is missing is harder to spot, because it is an absence rather than a mistake: no sign that an experienced person has looked at the whole account and made a call. The plan never changes shape. Nothing has ever been stopped. Nobody has said out loud that one part of the work is not worth doing.

This is one of the failures I cover in my longer guide to why SEO agencies fail, and it is the most structural one on the list. It is not carelessness and it is not dishonesty. It is what you get when the person touching your site follows a process instead of making decisions about your business, and nobody experienced ever reviews the result.

What judgement actually looks like in SEO work

Most people picture SEO as a body of work: keyword research, page optimisation, content, links, technical fixes. That is true, but it is only half the job. The other half is deciding which of those things to do on your specific site, in what order, and - the part that gets skipped - which of them to leave alone.

SEO is a competition for a small number of positions, and every hour spent on the wrong thing is an hour not spent on the thing that would have moved you. So the value of an experienced person is not that they know more tactics. Everybody knows the tactics; they are written down all over the internet. The value is that they can look at your site, your market and your budget and say "this one thing matters and the rest can wait", then defend that call when it is unpopular.

I use checklists myself, and so does every competent operator. They handle the boring, universal things well: title tags carry the service and the location, the site is crawlable, the internal links exist. But a checklist only encodes what was true when it was written, about sites in general. It cannot know that your industry has a high-volume search term that produces nothing but tyre-kickers, or that your two best pages are quietly eating each other, or that your domain has a history that makes link building a waste of money until something else is fixed first.

So the list gets worked through, faithfully and on schedule, and the account drifts. Judgement shows up as subtraction, and subtraction is the one thing a template has no mechanism to produce.

A checklist tells you what to do. It cannot tell you what not to bother with, and on most accounts that is the more valuable half.

Four decisions that need someone who has done this before

Abstract talk about judgement is easy to nod along to and hard to use, so here are four specific decisions that a template will not make for you. None of them are exotic. All of them involve saying no.

Killing a page that is cannibalising another

Cannibalisation is when two of your own pages target the same search, so Google has to choose between them and ranks whichever it likes slightly better - usually worse than a single strong page would have ranked. The fix is often to merge one into the other and redirect it. That decision is uncomfortable for three reasons: somebody was paid to write the page being killed, removing a page feels like going backwards, and the improvement shows up as one page moving up rather than as a new deliverable. Bulk content plans actively create the problem, which is why I have written separately about the cannibalisation trap that AI content buildouts walk into.

Declining to chase a keyword that will never convert

Some searches look wonderful on a spreadsheet and are worthless in a bank account. A broad, high-volume term that mostly attracts people looking for a DIY answer rather than a tradesperson can absorb a year of effort and return nothing. Sometimes the term is dominated by directories you have no realistic chance of displacing, so even winning would cost more than it earns. Saying "that one is not worth going after" costs an agency something, because the client wanted it and big volume numbers are persuasive. It is much easier to add it to the tracking list and let it sit at position 40 forever, quietly padding the report.

Consolidating rather than publishing

The default answer to almost every SEO question is "write something". Often the better answer is to take four thin, overlapping pages and rebuild them as one genuinely good page, then point the internal links at it. That produces no new content for the report, uses a chunk of the month's budget, and makes the site smaller. It is also, on a lot of the accounts I inherit, the single highest-value thing available.

Stopping link building until the technical work is done

Links are the most expensive part of SEO. Pointing them at a site with a broken structure, duplicate pages, or a target page that does not yet deserve to rank is like paying for a full page advert and printing the wrong phone number on it. The right call is often to pause the link spend for a quarter, fix the foundations, then start buying authority once it has somewhere to land. That decision shrinks this month's invoice, which is exactly why it is so rarely made inside a business built on retainers.

Additions only: the fingerprint of a template

Here is the pattern that gives the whole thing away, and it needs no SEO knowledge to see. Sort everything your provider has done in the last six months into two piles: things that added something, and things that stopped, removed, merged or reversed something.

New page. New blog post. New link. New keyword added to tracking. New location page. Those all go in the first pile. Now look at the second pile. On a healthy account it will not be empty. On an account being run to a template it is completely empty, every time.

The reason is simple. Additions are safe, visible and billable. Subtractions require a judgement call, an awkward conversation, and confidence that you are right. Someone six months into the job, working from a documented process, has no basis on which to argue for taking something away - and, to be fair to them, probably should not be arguing for it unsupervised. That is what senior review exists for.

The additions-only pattern also explains a common and miserable outcome: a business that has paid for SEO for a year and gone backwards. Very little of that damage comes from doing nothing. It comes from doing more and more, confidently, on a site where somebody should have stopped and thought.

The number that predicts most of this: accounts per strategist

If you only get one number out of your provider, make it this one: how many accounts does the person making the decisions on your site look after?

It matters because judgement is expensive in time, not just in skill. To make a genuine call, someone has to hold the whole account in their head at once: your market, your competitors, what has already been tried, what moved and what did not, and what your business actually earns money on. That is a couple of hours of proper thinking, not ten minutes before a call.

There is no magic number that separates good from bad, because it depends on account size and on how much support sits underneath the strategist. But the direction is not ambiguous. At a handful of accounts each, a strategist can genuinely know your business. At several dozen, they physically cannot, and the process quietly changes: instead of deciding what your account needs, they apply the same plan to everything and spend their time on whichever client is complaining loudest this week. Nothing dishonest has to happen. The arithmetic does it on its own.

A template scales beautifully across fifty accounts. Judgement does not scale at all. So a business that has sold more accounts than its senior people can think about drifts towards the template every time, not because anyone chose that, but because it is the only thing that fits in the day.

This is about seniority, not geography

I want to be completely direct here, because this is the point where the conversation usually goes off the rails. This is not an argument about offshore versus onshore, and it is not code for one. It is about experience and workload, full stop.

A junior following a template produces the same output whether they are sitting in Auckland, Melbourne, Manila or Manchester. A ten-year strategist with a sensible client load produces good judgement from any of those places too. The variable that matters is whether the person making the calls has enough experience to make them and enough time to make them properly. Where their desk sits tells you nothing about either.

Fixating on location also gets the diagnosis wrong in a way that costs the client money. Move the same template to a local office at twice the price and you have bought a nicer accent on the same result. The fix is a person with experience making the decisions and reviewing the work, and that fix is identical everywhere. If your provider uses an offshore team, the useful questions are exactly the ones below, and they are worth asking of a team in your own city too.

The ten-minute self-check

This takes one email and one honest read of the reply.

1. Ask who makes the strategic decisions on my account. Not who does the work and not who sends the report - who decides what gets worked on and what does not. You want a name and a role.
2. Ask how long that person has been doing SEO. Years in the industry, not years at that company, so that a vague answer is obviously a vague answer.
3. Ask how many accounts that person carries. This is the question most likely to be dodged, which is itself informative. A provider comfortable with the answer gives you a number.
4. Ask for one example, in writing, of a recommendation from the last six months that stopped or reversed something rather than adding to it. One example, from your account, not a case study about somebody else.
5. Sort the last six months of work into the two piles - additions, and things that were stopped, merged or reversed - and count each pile.

That fourth question does the real work. Experienced people say no to things: they kill a page that is cannibalising another, they tell you a piece of work is not worth the money, they pause a spend that is not landing. Anyone with a few years behind them can produce an example without effort, because they had the conversation and they remember it. A file of pure additions is what a checklist produces, and six months is long enough that an empty second pile is not bad luck.

How to read the answers you get

Good answers are specific and slightly boring. A name, a number of years, a number of accounts, and a concrete example with detail attached: this page was competing with that page, here is what happened after they were merged. It might come with an admission that the call did not fully pay off, which is a good sign rather than a bad one. Real decisions have mixed results and the people who make them talk that way.

Weak answers share a shape. "A team looks after your account." "Our strategists are all highly experienced." "We do not really share internal resourcing." None of those are lies. They are the sound of a question being absorbed rather than answered, and if every specific question has to leave the room and come back paraphrased a week later, nobody you can speak to is close to the strategy.

One fair caveat. If your account is genuinely new, three months in, there may honestly be nothing to stop yet, and a pile of additions is the correct output at that stage. This check is for accounts that have been running long enough for the first plan to have met reality.

What to do with what you find

If the answers are good, you have learned something reassuring and it cost you one email. If they are thin, the move is not necessarily to leave. Sometimes the fix is to ask for senior involvement explicitly, and pay for it: a monthly hour of a genuinely experienced person reviewing the account is worth more than several extra blog posts, and plenty of providers will arrange it if a client asks directly. What you should not do is assume more of the same work will eventually produce a different result. That is often the real story behind a page that has sat at position three or four for months: the work continues, none of it addresses the actual constraint, and nobody senior has ever said so.

And if you are still choosing who to hire, test this before you sign rather than after. A willingness to say no is one of the strongest signals available, which is why it also sits near the top of my guide to choosing an SEO provider. Ask a prospective provider what they would decline to do on your site. Anyone experienced will have an answer ready, because they have already spotted the thing on your site that is not worth doing.