Should You Do Ecommerce SEO Yourself or Hire an Agency?
A small shop with a manageable catalogue and an owner willing to learn can do a great deal of this alone. That is worth saying by somebody who sells the alternative, because the answer depends on your catalogue rather than on who is asked.
What You Can Genuinely Do Yourself
Five things a store owner can do well without help, with what each takes. Several of them are the highest value work available on a shop, which is the part suppliers tend not to mention.
Category page content. The most valuable of the five.
You know your products better than any agency will. Writing what helps somebody choose is knowledge work. Per our category page guide it is what those pages need. Allow a couple of hours per category, done properly.
Product descriptions on the ones that matter. Compatibility, exclusions and the question customers keep asking are things only you have. An hour per product, on the few dozen that justify it.
Image discipline. Resizing before uploading rather than after.
A habit rather than a task. It prevents the largest speed problem most shops develop.
Review collection. An automatic request after every order. An afternoon once, then almost nothing. Per our reviews guide, this is the cheapest content programme available to a shop.
Internal linking. Connecting categories the way customers shop.
You know what goes with what. Nobody external does. Most shops never write it down.
What those five have in common. They depend on knowing your products, which is the thing an agency cannot supply and has to extract from you anyway, which is why doing them yourself is frequently better rather than merely cheaper.
What Small Catalogues Change
Almost every difficult argument in ecommerce search is a problem of scale. Below a few hundred products, most of them stop applying. A great deal of published advice becomes irrelevant to you.
What stops mattering. Four things. They are the hard four.
Crawl efficiency. A shop of two hundred pages gets worked through regardless of how efficient it is.
Faceted navigation at scale. Fewer products means fewer combinations. The problem is one of volume.
Product prioritisation. With eighty products you can work on all of them, so the skill of deciding which few hundred matter has nothing to decide.
Depth. A small catalogue has nothing far from anywhere.
What still matters at any size. Whether your categories match how people search, which our structure guide covers. Whether those pages give anybody a reason to choose them. And whether anything is preventing the shop being found at all.
Where the threshold sits. Roughly a few hundred products. It is a range rather than a line.
Variants shift it downwards, since a hundred products in six sizes and four colours behaves like a much larger catalogue. Filters shift it downwards too. A flat range of simple items shifts it upwards.
What this means for the decision. A small shop is buying less. Not nothing, though considerably less than the sector generally implies, which is why block six exists.
What Needs Real Technical Skill
Five areas where inexperience is expensive rather than merely slow. In each, the cost of getting it wrong is larger than the cost of the help.
Faceted navigation. Deciding which filter combinations should exist as pages. Getting it wrong in one direction buries the shop under generated pages. In the other it removes routes people were using. Both are recoverable and both take months to notice.
Crawl at scale. Understanding where attention is going on a large catalogue.
This is diagnosis rather than configuration. The same symptom has several possible causes.
Duplication across variants. Which of six near identical pages should be the one.
The wrong answer here splits a product against itself. The shop looks fine while nothing ranks.
Migrations. The most expensive item here by a distance. A shop that changes addresses without mapping them properly discards everything it earned, in an afternoon, permanently unless somebody repairs it. Our migration guides exist because this happens constantly.
Structured data. Where an error is a false statement about price or availability.
Per our product page guide, this is a consumer protection matter before it is a search one, which puts it outside experimentation.
The pattern. These are the ones where trying and failing costs more than not trying. Everything in block one is safe to attempt. Everything here is not. That distinction is the useful version of this whole question.
The Time Cost
Doing this yourself is not free. It is paid for in hours, which for an owner-operator come from somewhere specific.
What it takes weekly. On a small shop, a few hours a week to make steady progress on block one, plus an initial period of considerably more while you learn what you are doing. That is not a large commitment. It is a permanent one.
What the learning costs before the work does. The part nobody budgets.
Most of the first month goes on establishing what matters. A good deal of published advice is written for ordinary websites and will send you the wrong way.
What it displaces. The question that decides it.
For an owner-operator, hours spent here are hours not spent buying stock, talking to customers, negotiating with suppliers or working on margin. Those activities have returns too, frequently faster ones.
How to think about it. Compare against your best alternative use of the time. Not against the agency fee. Against what those hours would otherwise produce. An owner whose time is worth a great deal in buying should probably buy the help. One who enjoys this and has quiet afternoons should probably not.
The trap. Owner time feels free because nobody invoices for it, while being the most expensive resource in a small business.
Where DIY Goes Wrong
Owners worry about breaking something dramatically. That is rare. Four quieter failures account for almost everything.
Not doing anything for months. The commonest by a wide margin.
An owner reads, plans, intends. Then a busy quarter arrives. The shop is unchanged a year later. The cost was the year rather than any mistake.
Doing the easy things repeatedly. The one that feels productive.
Rewriting the same descriptions, adjusting titles, checking positions daily. All of it feels like work and none of it addresses whatever is actually limiting the shop, which is usually structural and therefore harder to look at.
Following advice written for a different kind of site. The one that wastes the most effort.
Most published guidance assumes a site with a few dozen pages. Applied to a catalogue it produces work that is correct in principle and useless in practice, which is why this cluster exists separately.
Breaking something and not knowing. The one with the longest tail.
A setting changed, an application installed, a category renamed without a redirect. Nothing appears to happen, traffic declines slowly over months, then by the time anybody investigates nobody remembers what was altered. Our diagnostics guide covers finding it afterwards.
What prevents all four. Writing down what you change, with dates. A dated log costs nothing and it is the most useful habit an owner working alone can adopt.
The Middle Options
The question is usually posed as all or nothing. Four arrangements sit between. Most shops should be in one of them rather than at either end.
An audit, implemented by you. Buy the diagnosis, do the work.
Suits an owner with time and a willingness to follow instructions precisely. The risk is per our agency guide: audits that never get implemented. Ask for it prioritised rather than exhaustive, so the first three items are obvious.
Training. Somebody teaching you to do it on your own shop.
Frequently better value than it sounds, since the knowledge stays with the business. It suits an owner who intends to keep doing this rather than one hoping to hand it over later.
Occasional consulting. A few hours when a decision arises.
The arrangement we recommend most often for small shops. You do everything in block one. You buy an opinion before touching anything in block three. It costs very little and it prevents the expensive failures.
Split responsibility. An agency for the technical work, you for the content.
The most efficient division available, because it puts each part with whoever is genuinely better at it. You know the products. They know the platform. It requires both parties to be clear about who owns what, which is where it fails when it fails.
What to avoid. Paying for a full service and then doing the content yourself anyway.
Common. It means paying for something you are also providing.
What Changes The Answer
Five things move this decision. The last one is more predictive than most people expect.
Catalogue size. The largest factor, per block two.
Under a few hundred simple products, doing it yourself is genuinely viable. Above a few thousand the technical work becomes the whole game, as it does wherever variants and filters are heavy.
Growth. Where the catalogue is going rather than where it is.
A shop adding a thousand products this year should make its structural decisions now, since they are cheap to make and expensive to change afterwards.
How contested your market is. Whether adequate is enough.
In a market where competitors have done nothing, doing something reasonable wins. Where several have invested properly, adequate does not place.
Margin. What the work has to return to be worth buying.
Per our page on whether this is worth it, thin margins can mean no external help is affordable at any competence level, which makes the decision for you.
Whether you enjoy it. More predictive than any of the above.
This work is continuous rather than a project. An owner who finds it interesting will still be doing it in two years, which is what produces results. One who finds it tedious will stop in month three regardless of how viable it was, having paid the learning cost for nothing.
If You Are Doing It Yourself
Six steps, in sequence. Working in this order means each stage makes the next one worth doing.
One. Record where you are. Before changing anything.
Non-brand organic revenue for the last twelve months, indexed pages against total pages, plus how you measured both. Per our measurement guide, this hour is what makes next year's answer a fact rather than an opinion.
Two. Check the shop can be found. Pages that exist against pages indexed. A large gap either way is your first job, possibly your only one.
Three. Look at your categories against how people search. Not against your supplier's catalogue. This is the highest value thinking available and it costs an afternoon.
Four. Write the categories that matter. Three to start, not thirty.
Finishing three properly beats starting thirty. It tells you whether this is working.
Five. Turn on review collection. Per block one.
Six. Wait, then look. Six months, then compare against step one.
What to avoid throughout. Checking positions daily, which produces anxiety and no information, since daily movement is noise. Look monthly at impressions rather than positions, then leave the rest alone.
When To Hand Over
Five moments where the answer changes. Each is a specific event rather than a feeling, so you can decide in advance and act when it arrives.
The catalogue passes a few hundred products. Per block two.
You are planning a migration or replatform. The clearest trigger of the five.
This is the one place we would tell almost any shop to buy help, because the downside is total and permanent while the cost of doing it properly is modest.
Traffic declines and you cannot find out why. After a fortnight of looking.
Diagnosis is the skill hardest to acquire from reading. A decline continuing while you learn is expensive.
You have done everything in block eight and stalled. The encouraging trigger.
An owner who completed that sequence and stopped progressing has usually reached the technical ceiling in block three, which is a good position to be buying help from because you know what you need.
The hours stopped being available. The commonest and the least dramatic.
Per block four, a permanent few hours a week is a real commitment. When it stops happening, the choice is to buy the help or to accept the shop will not progress. Either is better than the third option of intending to get back to it.
What to bring. Your log and your baseline. An owner arriving with a dated record of what was changed and a measurement from before it started is a considerably cheaper client, because half the first month is normally spent establishing exactly that.
Some of this
you should keep.
Category content, your best product descriptions and internal linking all depend on knowing your products. That is the thing an agency cannot supply and has to extract from you anyway. What we are better at is everything that breaks quietly.
What is included every month:
£350 per month, one target area. No setup fee, nothing billed separately.
Twenty-two guides.
One subject.
This guide covers doing it yourself. The rest of the series covers the sequence, structure, category and product pages, technical health, measurement and everything a store owner has to decide.
