The three real options, stated plainly
You can hire a person to work for your store, hire an agency to work on it, or buy a service that does the work for you. People often frame this as in-house versus agency, as if those were the only two. There is a third, and it changes the math for a lot of stores. Before comparing them, here is what each one is.
In-house. One employee, or a small team, on your payroll. They know your products, sit near the people who decide what goes on the site, and work only for you. You pay salary, taxes, equipment and your own management time.
Agency. A company that assigns people to your account. You pay a monthly fee. They bring a mix of skills, but the person who sold you is usually not the person doing the work, and your store is one of many.
Done-for-you service. A provider that researches your store, then writes and publishes pages on it with little or no input from you. It is closer to buying a result than buying hours.
If you are new to all of this and want the broader picture of who does what, the overview of consultants, freelancers, agencies and done-for-you help comes first. This page answers one narrower question: given your store, which arrangement is a sensible fit?
What an in-house hire does well, and what it quietly requires
In-house wins on context. A person inside the company knows when the next product launches, why a collection was renamed, which supplier is late and what customers keep asking support. That knowledge shows up in better pages. It is hard for an outsider to get.
In-house also wins on speed of small decisions. If a title needs changing today, they change it today. And when something breaks after a theme update, they notice.
Now the requirements nobody mentions in the job post.
Ecommerce SEO is several jobs. Technical fixes, product data, writing buyer-focused content, planning internal links, reading Search Console, and now watching how AI assistants describe you. One person is rarely strong at all of them. Our page on specialist roles in ecommerce SEO breaks those jobs apart.
A single point of failure. When that person leaves, goes on leave, or gets pulled onto email campaigns, the work stops and the knowledge goes with them.
Management time. Someone has to set priorities and judge the output. If you are not an SEO yourself, you may not be able to tell whether the work is good, which is the same problem you have with an agency, only slower to notice.
Tool and training cost. Paid tools, courses and conferences are real line items on top of salary.
What an agency does well, and where it strains
An agency wins on range. You get access to technical, writing and link-related skills without hiring for each. It also wins on having seen many stores, so a pattern that looks mysterious to you may be routine to them.
It strains in three places.
Attention. Your account competes with others for the same hours. Fixed fees reward doing the minimum that keeps you from leaving.
Distance. They do not see your inventory plans or your customer emails, so content tends toward generic.
Volume. Content costs people-hours, so agency retainers cap how much gets published. The homepage figure we cite for scale: an agency doing this work charges about $3,209 a month (2026 average, 439 providers) and publishes 4 to 12 pieces. For a catalog with hundreds of pages to improve, that pace is slow.
If you go the agency route, working with an ecommerce SEO agency month to month covers how to judge one, and a fair side-by-side lives in RunOctopus compared with an SEO agency.
Which fits a store of your size
Size is a better guide than preference. Think in terms of how many pages need attention and how many different skills those pages call for.
A small store, a few dozen products
An in-house hire is almost never right here. There is not enough SEO work to fill a role, and one person cannot also run the store. Learn the free tools, keep the monthly routine, and bring in outside help for the writing. A short consultation to set direction, then doing the steady publishing through a service, covers it.
A mid-sized store, hundreds of products
This is the hard zone. The work is large, the skills are wide, and one salaried person is stretched. A common pattern that works is an in-house owner of the priorities, someone who knows the products and keeps the plan, paired with an outside team or service that produces and publishes the pages.
A large catalog, thousands of products
Now an in-house specialist earns their place, because templates, crawling and governance need someone embedded who talks to developers daily. Most large stores still use outside help for volume and for second opinions.
The grid above shows the logic. Little work and narrow needs mean advice is enough. Lots of work and wide skills mean a team. That is why so many mid-sized stores land in the top right, where no single person is enough.
Budget, risk and what you can undo
Money matters, but how reversible each choice is matters more.
A hire is the hardest to undo. If it does not work, you have a recruiting cycle, notice periods and months of ramp-up behind you. A bad hire also costs the time of the person managing them.
An agency is easier to leave, but check the contract. Look at the term and notice period, and confirm that everything published stays yours. Content on your own domain is yours. Content stuck in a vendor's system might not be.
A service is the easiest to test. You can look at the first pages it publishes and judge the output directly, rather than waiting for a quarterly meeting.
On budget, ask what a dollar buys in each model. Salary buys hours of one person across all their duties. A retainer buys a slice of a team's hours across many clients. A service buys finished pages. For the economics of any of these against sales, what to expect from ecommerce SEO return gives a way to think about it without invented numbers.
Last, watch the hidden cost on the in-house side: the work you do not see. A salaried employee asked to "do SEO" may spend half the week on reporting and meetings. Ask for the output you expect in pages and fixes, not in hours.
A short way to decide
Answer these in order. Stop at the first line that fits.
Do you have fewer than a hundred pages that matter and an owner who can spend a few hours a month? Do it yourself with free tools, buy a one-time review, and consider a service for the writing.
Do you have the skills in the room but not the hours? A service or contractor that produces pages, with you or a staff member setting priorities.
Do you need many skills at once and have no one to run it? An agency or a done-for-you service, with a named person on your side who checks the output monthly.
Is the catalog large, with developers and several teams touching the site? An in-house specialist, plus outside help for volume.
Whatever you pick, decide in advance how you will judge it. Clicks to the pages that sell, by page type, over a few months, compared with before. The guide to what drives the price of ecommerce SEO helps you check whether the cost lines up with that.
The hybrid that many stores settle on, and how to switch later
The choice is rarely permanent, and it is rarely pure. Many stores end up with a blend, and most change arrangements as they grow.
The owner of priorities plus outside hands
The most common blend gives one internal person the job of deciding what matters. They know the catalog, the margins, the seasons and the launch calendar. Outside help does the volume work: writing, publishing, fixing and reporting. The internal person spends a few hours a week reviewing, not producing. It keeps the context of in-house with the capacity of a team.
The agency for a project, the in-house person for the rest
Another pattern uses an agency for a one-off push, such as a platform move, a site restructure or a first big content push. Afterward, an in-house person maintains it. This works when the project has a clear end and the handover is written down.
Starting outside and bringing it in
A store that outgrows its outside help may hire an in-house lead later. The cleanest path is to make sure that from day one you own the accounts, the documents and every page published. Then nothing is lost in the handover.
How to switch without losing ground
Whichever direction you move, do four things. Export Search Console and analytics data before the change. Ask the outgoing party for a written list of what was done and what was planned. Keep the publishing going during the handover, because search rewards steady output and punishes gaps. And set the first review for about two months after the change, since early movement in search is slow to show.
If your priority is to compare an agency directly with a service, when a freelancer works for a store and where it breaks covers the middle path between a hire and an agency.
Five questions that settle most cases
If the sections above left you still unsure, five plain questions settle most cases. Answer them on paper.
1. How many pages on my store need real work in the next year? Count collection pages, key product pages and missing articles. If the number is under a few dozen, you do not need a team. If it is in the hundreds, you need capacity more than advice.
2. Who inside my business knows the products and can answer questions in a day? If no one, in-house SEO will not fix that, and outside help will need a contact from you anyway.
3. Can I judge the work? If you cannot tell a strong collection page from a weak one, you need a way to check output. That could be a second opinion once a quarter or reports you understand. Hiring a person you cannot judge is the same risk as hiring an agency you cannot judge.
4. How steady is my budget? A salary is a fixed commitment. A monthly service can often scale up or down, which matters for stores with uneven cash flow.
5. What happens if the person or provider disappears next month? If the answer is "everything stops and we lose what we learned", pick the arrangement that leaves written plans and published pages behind, and own the accounts.
Many owners find that question 1 does most of the work. The size of the job, counted in pages, points to the answer more reliably than any general rule about agencies or hires.
Three mistakes that show up whichever way you go
Whichever option you pick, three errors repeat across stores.
Hiring before knowing the job. Owners write a job description or sign a retainer before they have counted what needs doing. Count first. A list of the pages that need work, even a rough one, makes every later decision easier.
Expecting the hire to learn the store alone. A new employee or a new agency needs your input in the first month: your best sellers, your margins, your customers' questions, your plans. Give it freely. The speed of the first quarter depends on it.
No handover plan. People leave and contracts end. Keep a shared folder with the plan, the change log, the account logins in your name and a list of what is published. Then no arrangement is a single point of failure.
How RunOctopus handles this
Many stores end up between the two options, with too much work for one person and too little control inside an agency retainer.
RunOctopus is a done-for-you service. Its AI, Ollie, researches your business, then writes and publishes pages directly on your own website. A team plans, writes, reviews and publishes alongside Ollie. It works with Shopify, WordPress, WooCommerce and Wix.
There is nothing to install, nothing to approve, nothing to upload and nothing to schedule. You do not approve each page, because pages go live on their own. In the first week you get 8 long-form articles, 6 pages and a custom interactive tool built for your niche. After that it is 175+ pieces of content a month, written, published and linked. You also get reports on where ChatGPT, Perplexity, Claude and Gemini name your business.
There is a 90-day promise: more visibility on Google and in AI answers within 90 days, measured against the 90 days before you joined, or every dollar back. For scale, an agency doing this work charges about $3,209 a month (2026 average, 439 providers) and publishes 4 to 12 pieces.
Paste your address into the free site read and Ollie reads the site and shows the plan. No account, no call.