Discussion starter: an agency question, posted with an answer from Niraj Raut to open the thread. If you have dealt with this on a site, reply with what you saw, especially where it differs.
- Agency
At what company size or spend does moving SEO in-house make more sense than an agency, in your experience?
Disclosure: Niraj Raut, who posted this answer, runs the SEO consultancy linked at the end of it.
1 reply
-
adminAdmin
Company size and spend are rough proxies for the real threshold. Moving SEO in-house tends to make sense when three things are true at once: the work is continuous rather than project-based, most of it needs internal access and context (engineering sprints, product data, content operations), and your agency spend is already close to the fully loaded cost of a senior hire. Even then, the first step should rarely be replacing the agency outright. It should be hiring one senior in-house owner and narrowing the agency to specialist work. A fuller in-house team makes sense later, when the site and the revenue behind it give technical, content and analytics roles a full workload each.
The decision turns less on how big the company is than on where the bottleneck sits. If good recommendations stall inside the company, an in-house owner removes the bottleneck. If the company does not know what to do, or needs several specialisms at low volume, an agency is still the better buy.
Why headcount and spend make weak thresholds
Consider two hypothetical companies. A 60-person online retailer with 150,000 SKUs, weekly releases and organic search as its largest acquisition channel needs SEO in product and engineering decisions every week. A 2,000-person professional services firm with a 300-page site that changes a few times a year may never need more than an agency and a marketing manager who can brief it well. Any headcount or revenue threshold would rank these two the wrong way round.
The variables that decide it:
- Rate of change. How often templates, navigation, the catalogue or content change. Each release is a chance to break something.
- Revenue at stake. What a modest swing in organic performance is worth in a year, compared with the cost of an in-house hire.
- How much needs insiders. The share of the work that requires engineering, product or editorial capacity only people inside the company can secure.
- Breadth of specialisms. Technical, content, digital PR, international and analytics. One person rarely covers all of them well.
- Management capacity. Whether anyone inside can hire, direct and judge an SEO.
Compare the full cost, and the hours it buys
Comparing a retainer with a salary understates the in-house cost. A fair comparison puts the retainer against the fully loaded cost of the role, and compares hours of useful work rather than invoices. An agency hour carries the agency’s overheads and margin, and part of every retainer goes on reporting and account management. An employee’s hours are cheaper, but they are only useful if there is enough work, direction and access to fill them.
Hypothetical, per year Agency retainer Senior in-house SEO Cash cost AUD 144,000 (AUD 12,000 a month) AUD 185,000: salary AUD 140,000, on-costs AUD 30,000, tools AUD 15,000. Year one adds about AUD 25,000 for recruitment Hours on your site About 800 at a blended AUD 180 an hour, some of it reporting and account management About 1,400 to 1,600 after leave, meetings and internal admin Specialist breadth A team: technical, content, PR, analytics One person’s strengths Context and access Briefed from outside; limited access to sprints and data In planning meetings, release reviews and data Continuity risk Account team changes Key-person risk if they leave All figures are illustrative; salaries and agency rates vary widely by market and seniority. The method is what matters. In this example the in-house hire costs more cash but buys nearly twice the hours, with far better access. That advantage is real only if the backlog is deep enough to fill those hours and someone can direct them. A company with a thin backlog, or one whose single hire would sit alone without a senior sponsor, pays more for less.
There is also an insurance argument the cost table misses. Suppose, hypothetically, organic search drives AUD 5 million a year. A release that noindexes a category directory, or a migration that loses a fifth of organic revenue for a quarter, would cost more than the in-house salary. An internal owner who reviews releases before they ship is partly insurance, and that is the part an agency outside the release process struggles to provide.
Signals that it is time to bring SEO in-house
- Agency recommendations sit in the backlog for months, and the reason is internal priority, not their quality.
- Releases introduce SEO problems faster than an external audit cycle catches them.
- A redesign, replatform or new site is coming. Google’s guide to hiring an SEO says a great time to hire is when you are considering a site redesign or planning to launch a new site, and the earlier the better. The guide covers hiring any SEO, but its point about early involvement is easier to meet with someone inside the room when requirements are written.
- Decisions need joined data (revenue, margin, CRM stages, product data) that an agency cannot easily reach.
- You pay the agency to re-learn your business each time its account team changes.
- Organic search is a leading acquisition channel, and a modest swing in it is worth more than a senior salary.
Signals that an agency is still the better buy
- The work is episodic. An audit, one migration or a content push, followed by maintenance.
- You need several specialisms at low volume. A digital PR campaign, an international hreflang build and a technical audit in the same year would need three specialists you could not keep busy.
- Nobody inside can manage an SEO. A junior hire reporting to a manager who cannot judge the work is set up to stall, and nobody will know whether it has.
- You want a wider comparison set. During a core update, an agency watching dozens of sites can separate an industry-wide shift from your own problem with more data than any single in-house team has.
- Digital PR at volume. Media relationships and campaign production are hard to justify for one brand’s output.
A staged path, and the triggers between stages
Stage 1: agency or freelancer, with a marketing manager who briefs and checks the work↓Trigger: the work becomes continuous, recommendations stall, or a redesign is coming↓Stage 2: one senior in-house SEO lead; the agency narrows to specialist execution and a second opinion↓Trigger: the backlog outgrows one person, markets or product lines multiply, organic becomes a leading channel↓Stage 3: a small in-house team (technical, content, analytics) with specialist agencies for digital PR, migrations and international workThe first hire should be senior. The job is less about doing tasks than about winning priority for them: getting SEO requirements into tickets, reviewing releases, and deciding what the agency should and should not do. Replacing a senior agency team with a junior executor removes the strategy and keeps the tasks. A fractional or contract head of SEO is a reasonable middle step when a full-time senior role is hard to justify.
A growing company can reach the first trigger long before it reaches the second, which is why the hybrid stage can last for years. That is not a failure to finish moving in-house. It can be the right end state.
What goes wrong in the handover
Watch for this:
Before changing anything, make sure the company, not the agency, owns admin access to Search Console and analytics, the tool accounts, crawl and rank-tracking history, dashboards and documentation. Export history from any platform the agency controls. Losing years of baseline data costs more than a month of overlap. And avoid switching during a migration or a peak trading period.
A newer trap is adding a separate “AI search” provider on top of whichever model you choose. Google’s May 2026 guide to its generative AI features says foundational SEO still applies and that you do not need special AI text files, special schema or AI-specific rewording. For Google’s AI features at least, that is the same work, so paying two providers for it duplicates effort. Visibility in other AI platforms raises measurement questions worth testing, but test them before committing a second retainer.
How to test the move before committing
Run a hybrid for two quarters before restructuring: bring in a senior lead, as a hire or a contractor, and narrow the agency’s scope. Two quarters covers onboarding and several release cycles. Organic results lag implementation, so judge the pilot mainly on operating measures:
- Recommendations shipped per month, and the median time from recommendation to release.
- SEO regressions caught before release compared with those found after.
- The share of agency hours spent on delivery rather than reporting and account management.
- Non-brand organic clicks and revenue, compared with the same period last year, as the lagging check.
If implementation speeds up and the agency’s work naturally shrinks to specialist projects, make the structure permanent. If the in-house lead spends most of the week managing the agency and producing reports, you have added a layer, not a capability.
The first thing I would do is sort the last two quarters of agency recommendations into three piles: shipped, stalled inside the company, and rejected on merit. If most stalled inside the company, the constraint is internal and an in-house owner is the fix. If most shipped and results still disappointed, the constraint is strategy or specialist capacity, and moving in-house will not solve it. Size and spend matter only because they tend to make the first constraint bite harder.
Need help with this on your own site?
Niraj Raut works with businesses in Nepal, Australia, the UK, Europe and the US on technical SEO, ecommerce SEO, local SEO and AI search.
Add a reply Cancel reply
Replies are open to approved members. Log in or apply to join.