Four Hours a Week: A Prioritisation System for People Doing SEO Alone

Most published SEO advice is written for people whose job is SEO. If that is not you — if search is one of eleven things you are responsible for, and it gets Thursday afternoons when nothing else catches fire — then the standard recommendations are not just impractical, they are actively demoralising. A comprehensive audit produces four hundred findings and you have four hours.

This article is a system for that situation, built around the Semalt platform because consolidated data is what makes a four-hour week viable at all. The premise is simple: with limited time, what you skip matters more than what you do.

Where the four hours actually go

The allocation described below; adjust the proportions, but keep monitoring separate from project work.

The split matters more than the total. Teams with limited time usually spend it all reactively — reading, tweaking, responding to whatever someone mentioned in a meeting — and end the quarter having done a great deal without moving anything.

Protect the monthly project time above all. It is the only part of the week that produces compounding results, and it is invariably the first thing sacrificed when something urgent appears.

The weekly thirty minutes

This is monitoring, not work. The goal is to notice problems while they are still cheap, and to notice nothing the rest of the time.

1. Read the change, not the state

   Open the dashboard and look only at what moved since last week. Everything stable can be ignored entirely — it is not asking for your attention.

2. Check the technical changelog

   New crawl errors, new 404s, pages that dropped out of the index. Deployments break things, and a fault caught within a week is a small fix; the same fault found in month three is a recovery project.

3. Scan the commercial group only

   Not every keyword. The terms that produce revenue. A drop there is worth investigating today; a drop on an informational term can wait for the monthly review.

4. Write one line in the log

   Date, what moved, what you did or deliberately did not do. Thirty seconds, and it is the thing that makes the quarterly review possible.

One thing per month

The monthly project is where progress actually comes from. One substantial piece of work, finished, beats five started.

A realistic list of monthly projects, roughly in order of return for a site that has never had sustained attention:

Six months of that, at two and a half hours a week, will move most small sites more than a year of scattered activity. None of it is advanced, and that is the point: the advanced work is not where the returns are when the basics have never been done.

What to ignore without guilt

This is the part that makes the system work, and the part nobody writes down.

The second item on the right saves the most time. Following industry news feels productive and changes almost nothing about what you should do next Thursday. If something genuinely significant happens, it will still be significant in a month, and by then the useful analysis will exist instead of the speculative kind.

We cover this in detail in Local SEO for Real.

Making the tooling do the remembering

A person with four hours a week cannot also be a monitoring system. The setup should surface problems without being asked, so that the weekly half hour is reading rather than searching.

Three configurations do most of that work. Scheduled crawling, so technical changes appear as a diff rather than requiring you to run something. Keyword grouping by commercial intent, so the weekly scan is one chart instead of a table. And a tracked competitor set, so that when your numbers move you can immediately see whether the market moved with you — which turns "why did we drop" from an afternoon's investigation into a five-second observation.

Set all three up once, properly, and the ongoing cost of the system approaches zero. Set none of them up and every week starts with reconstruction.

Reporting to someone who is not you

Even a one-person operation usually has to justify the time to a manager or an owner. Keep it to one page, monthly, with three sections: what moved, what I did, what is next and what I need. It takes twenty minutes to write and it prevents the conversation where someone asks what SEO has produced this year and nobody has a record.

There is a full walkthrough in Redesigning Without Losing Google.

Include the things you decided not to do and why. It reads as judgement rather than omission, and it heads off the recurring suggestion to chase whatever a competitor was seen doing last week.

Working with an agency on four hours a week

Many people in this situation are not doing everything alone — they have an agency, and their four hours are spent managing it. That is a different job with its own failure modes, and the same time constraint applies.

The most common waste is the monthly call spent listening to a report being read aloud. If the reporting is accessible in a shared dashboard, that hour should be a decision meeting instead: here is what moved, here is what we recommend, what do you approve. Insisting on that format is the single highest-return change available to a busy client, and most agencies welcome it.

The second waste is the blocked ticket. In our experience the majority of stalled SEO work is not stalled because the agency is slow but because a change is waiting on someone internal — a developer, a content owner, a legal review. If you have four hours a week and an agency, spend a meaningful share of them unblocking rather than reviewing. Nothing an agency does compensates for a fix that sits unshipped for five months.

See also: The Blog That Stopped Paying.

Third, keep your own log even when someone else keeps theirs. Agencies change account managers, and institutional memory of what was tried and what happened is worth more than it sounds when a new person arrives and proposes the thing that was already ruled out last year.

One useful check on the relationship: ask what they recommend that you have not approved, and why it has not happened. If the answer is a list, the constraint is on your side and you now know where the four hours should go. If there is no list, that is worth a conversation too.

When four hours is genuinely not enough

The system above holds for a small site with a clear structure and a handful of commercial pages. Three situations break it, and recognising them early is more useful than working harder.

A large catalogue with thousands of URLs generates structural problems faster than four hours a week can resolve them. A competitive market where established players invest continuously will not be moved by six hours a month, regardless of how well those hours are spent. And a migration or replatform is a project with a fixed deadline and a high cost of failure, which is exactly the wrong shape for occasional attention.

In those cases the honest answer is that the constraint is capacity, not method, and the useful thing to do is to say so clearly with the evidence to support it. A well-documented six months of work that shows steady progress and a visible ceiling is a much stronger argument for resources than any amount of asking.

Start here

If you have four hours this week and nothing set up, spend them like this. One hour configuring: scheduled crawl, keyword groups by intent, three competitors. One hour on the first monitoring pass, reading what it tells you. Two hours fixing whichever indexation or technical fault it surfaced first.

Then keep the rhythm: thirty minutes weekly, one project monthly, one page of reporting. It is unremarkable, and after six months the difference between it and the alternative — reacting, reading, tweaking — is not subtle.

If the setup part is what has been postponed, that is the hour worth spending now. Open the dashboard, add the domain, group the commercial terms, and let the first crawl tell you what your month one project should be.