How to choose a SaaS SEO agency, including whether to hire one at all

We are an agency, so treat everything here accordingly. I have tried to write the version I would want to read if I were on the buying side, including the parts that argue against hiring us.

Three cases where you should not hire anyone

If you have fewer than about thirty customers, you do not yet know who buys or why, and any keyword research anyone does for you will be guesswork dressed up in a spreadsheet. Sell more deals manually first. Six months of that will produce better positioning than any agency could.

If your ACV is under about $2,000 with high churn and no expansion, the arithmetic is difficult. You need volume that SEO in a small category cannot supply quickly, and paid acquisition will tell you more, faster.

And if you need pipeline this quarter, this is the wrong channel and anyone who tells you otherwise is selling. SEO in B2B pays back somewhere between month five and month twelve. That is not a caveat, it is the whole shape of the thing.

Generalist versus specialist, and how to tell quickly

Every agency has a page saying they work with SaaS. Three questions will separate the ones who do.

Ask them to name the three query patterns that convert best in B2B software and explain why. If the answer does not include comparison and alternative queries within the first fifteen seconds, they have not run a SaaS program that mattered.

Ask what they would do about your docs subdomain. A specialist will immediately have an opinion about whether it should be a subdomain at all, how it competes with the marketing site, and what happens to developer queries. A generalist will not know that you have one.

Ask how they would report to a CFO given a five-month sales cycle. If the answer is a traffic dashboard, they have never had that conversation.

You can run all three in ten minutes on a first call and it will save you a month.

What the pricing actually means

Roughly, and this is the market as of 2026 for US, UK, and EU agencies working with B2B software:

Under $3,000 a month generally buys you a junior executing a template, or an offshore content operation, or both. There are exceptions, usually a single very good freelancer, and those are worth finding.

$4,000 to $8,000 buys a real strategy and either content or technical work but rarely both at depth, plus little to no link budget.

$8,000 to $20,000 is where a full program lives: strategy, content, technical, and links running together with senior attention.

Above $20,000 you are usually buying embedded capacity or an enterprise-scale program with multiple workstreams.

The single most useful pricing question: is link acquisition budget included or separate? If an agency quotes $6,000 all-in including links in a competitive category, one of the two things is not really happening. Link acquisition in B2B costs $2,000 to $8,000 a month on its own.

Questions that produce useful answers

“Show me a program that did not work and tell me why.” Any agency with more than twenty clients has failed at least twice. An answer blaming the client entirely is a bad sign, as is an inability to name one at all. What you want is a specific diagnosis: the market did not have the demand we projected, or the client could not ship pages fast enough, or we were wrong about the competitive difficulty.

“Who will actually do the work?” Ask for names and ask whether the person on this call is one of them. The gap between the pitch team and the delivery team is the single most common source of disappointment in agency relationships.

“What is your client-to-strategist ratio?” Above about six accounts per strategist, you are getting a template. Ask directly.

“How many clients do you have in my exact category, and can I talk to one?” Specialists say yes quickly. Note that if they have a direct competitor of yours, that is a genuine conflict and you should ask how they handle it. We turn down direct competitors of existing clients, which costs us business regularly.

“What happens if I want to leave in month four?” The answer tells you about the contract and about the culture. Twelve-month lock-ins with a 90-day notice period are a retention mechanism, not a strategy requirement.

The proposal itself

A proposal that could have been written before meeting you is a bad sign, and most of them can be. The tell is that it contains no specifics about your site.

A good one names actual pages of yours and says what is wrong with them, names actual competitors and what they are doing that you are not, and contains at least one thing you did not know. It should also contain something you disagree with, because an agency that agrees with all of your existing assumptions is either lucky or telling you what you want to hear.

Be suspicious of guarantees. Nobody controls the ranking, so a guarantee is either meaningless in the fine print or attached to something trivial like “top 3 for your brand name.”

Agency, freelancer, or in-house

A senior in-house SEO lead in the US costs roughly $130,000 to $180,000 fully loaded, and they cannot write content, build links, and do technical implementation simultaneously. What they can do is know your product deeply and be in every planning meeting, which no agency will match.

The pattern that works best in our experience: in-house owns the writing and the product knowledge, external owns strategy, technical, and links. That splits the work along the line where each side has a real advantage.

A good freelancer at $4,000 to $7,000 a month can outperform a mediocre agency easily. The risks are capacity, holidays, and the fact that no one person is genuinely excellent at technical SEO, content strategy, and digital PR at once.

The test I would apply

If I were buying, I would pay for a scoped audit from two agencies before committing to a retainer with either. Five to eight thousand each, four weeks, and you learn more about how they think than any number of sales calls will tell you.

Then hire the one whose document told you something uncomfortable.