In-House SEO vs Agency vs Freelancer: The Real Math

Somebody in your business has said it out loud by now: 'why are we paying an agency when we could just hire someone?' It is a fair question. It is also almost always priced wrong, because the number people compare against the agency invoice is the salary — and salary is the smallest part of what an employee costs you.

So here is the math with every input pulled from a live source you can open yourself. No vendor-flattering assumptions, no invented averages.

What does an in-house SEO hire actually cost, fully loaded?

Budget roughly $110,000 a year, not $72,000. Indeed lists the average US base salary for an SEO specialist at $71,632 per year (Indeed, 2026). That is the number that ends up in the spreadsheet. It is not the number that leaves your bank account.

The Bureau of Labor Statistics tracks what employers actually pay. In March 2026, employer compensation costs for civilian workers averaged $49.32 per hour worked — $33.72 in wages and salaries (68.4% of the total) and $15.60 in benefits (31.6%) (BLS, Employer Costs for Employee Compensation).

Read that ratio carefully, because nearly everyone gets it backwards. Benefits are 31.6% of total compensation, not 31.6% on top of salary. The correct move is to divide, not multiply: $71,632 ÷ 0.684 works out to roughly $104,700 in total compensation cost. That BLS benefits bucket already includes legally required items — the employer share of Social Security (6.2%) and Medicare (1.45%), a combined 7.65% (IRS Topic 751). So do not add FICA on top of the 31.6%. You would be counting it twice.

Then the tools. Ahrefs lists its Standard plan at $249/mo and Advanced at $449/mo (Ahrefs pricing). Semrush lists $139, $199, $299 and $549/mo across its SEO plans when billed monthly (Semrush pricing). Most in-house SEOs want both, because their crawl and link datasets disagree — and the argument for which one to buy is its own decision, covered in our Semrush vs Ahrefs vs Moz breakdown.

That is the honest floor: about $111,000 a year for one person and two subscriptions. Our own retainer starts at $2,500/mo — $30,000/yr, month-to-month, no minimum. The salary comparison is not close, and it is not supposed to be, because you are not buying the same thing. Which brings us to the part that actually decides this.

Can one person cover technical, content, links, local, and paid?

No — and the evidence is in how many distinct levers exist. Whitespark's 2026 Local Search Ranking Factors survey had 47 local-search experts score 187 separate factors, and found that three of the top five factors for AI search visibility are citation factors, leading Whitespark to conclude that 'in AI SEO, mentions (citations) are the new link' (Whitespark, Nov 2025).

Those 187 factors do not sort themselves into one job description. Technical SEO is engineering work. Link earning is outreach and PR work. Local is operations — categories, hours, review velocity, per-location pages. Content is writing and editing. Paid is media buying. These are five different people in most agencies for a reason. A full picture of what the work covers is in what SEO services actually include.

So the real in-house question is not 'can I afford $111,000?' It is: which two of the five pillars will this hire actually be good at, and what will I spend to fill the other three? Content production and link earning are usually the gap, and they are not cheap to buy piecemeal. Add that to the $111,000 before you compare anything.

When does hiring in-house genuinely beat an agency?

When your search work has enough recurring volume to keep one person busy 40 hours a week — which in practice means multiple locations, multiple service lines, or a site big enough that technical debt is a standing job. Below that, you are paying $111,000 for a person who is idle half the quarter and outsourcing the half you actually needed.

• You have 3+ physical locations, each with its own Google Business Profile, review flow, and location page to maintain.
• Your site has thousands of URLs and a release cycle — someone has to be in the standups, not in a monthly call.
• Search is your primary acquisition channel and you want the knowledge to live in the building, not in a vendor.
• You already have vendors for content and links, and you need an internal owner to run them.
• Compliance or confidentiality makes outside access genuinely hard — some legal, financial, and healthcare setups qualify.

Notice what is not on that list: revenue. A $12M single-location firm still has one Google Business Profile and one set of service pages. A $4M four-location HVAC company has four of everything. Volume of work tracks locations and service lines — not the top line.

Where does a freelancer beat both?

On a single bounded problem with a clear finish line — typically $1,500–$5,000 for a defined deliverable rather than an open-ended retainer. A good freelancer is the cheapest way to get one specific thing done well: a technical audit, a site migration plan, schema deployment, a set of location pages, a content refresh on decaying posts.

Freelancers are worst as a permanent, open-ended program. Not because they are weak — many are better operators than the account manager an agency will hand you — but because the scope of an ongoing search program exceeds what one contractor can hold while also selling their next contract. You get their best pillar and silence on the other four.

The tell that you are using a freelancer wrong: you cannot describe what you are buying next month. If the answer is 'SEO', you are buying an ongoing program from someone structured to deliver projects.

What is the hidden cost of a single point of failure?

One resignation stops your entire program while you keep paying the $6,576/yr tool bill that only they knew how to use. That is the bus factor, and it is the risk in-house buyers systematically underprice.

Run the honest timeline. Two weeks' notice. Then a hiring cycle — job post, screening, interviews, an offer, a notice period on their end. Then ramp: nobody ships meaningful technical or content work in month one on a site they have never crawled. In that window, rankings do not politely pause. Competitors keep publishing, links keep aging, and Google keeps recrawling.

Agencies fail differently. The risk there is that you are one of many clients and get handed a junior. That risk is real — ask who will actually touch your account, by name, before you sign. But it is a quality risk you can inspect in the pitch, not a continuity risk that shows up as a two-quarter hole. Vetting for it is a solvable problem — see the SEO agency vetting playbook.

The other continuity question people forget to ask: who owns the work when it ends? With us, you do — the content, the links, the ad accounts, the codebase. An employee walks out with the context in their head. Ask any vendor the ownership question in writing before you sign anything.

How does the math change with each new location?

Every new location adds a full unit of recurring work, and Google's own documentation explains why. Google states that local results rank on three factors — relevance ('how well a Business Profile matches what someone is searching for'), distance ('how far each business is from the customer who's searching'), and prominence ('how well-known a business is') — and that 'there's no way to request or pay for a better local ranking on Google' (Google Business Profile Help).

Distance is the one that scales your labor. Each location has its own proximity radius, and it can only be won from that address. So each location needs its own profile, its own categories, its own review velocity, its own city page, its own citations. None of it is copy-paste — and near-duplicate location pages get you thin-content problems instead of rankings.

This is the threshold nobody publishes: in-house SEO starts winning at roughly four locations, not at a revenue number. Multi-location work is per-unit labor. One-location work is not, no matter how large the business gets.

Which model should you pick?

For a single-location or single-market US service business, the agency or fractional model wins on cost and on coverage — you get all five pillars for less than a third of a fully loaded hire. Here is the three-way verdict without hedging.

Two rules to apply regardless of which you pick. First, no 12-month lock-in — a 12-month contract protects the agency, not you, and if a vendor cannot keep you month-to-month they are pricing in their own churn. Second, no ranking guarantees: Google says outright there is no way to pay for a better local ranking, so anyone guaranteeing position one is either lying or bidding on your brand name. If you want the full budget picture first, read how much SEO costs for a service business.

If you land on the agency side of this, we run SEO, GEO, and AEO as one program from $2,500/mo, month-to-month, and you own everything we build. Start with our SEO services, or skip the pitch and get my free audit — you will get the specific gaps in your site whether or not you hire anyone.