We turn down roughly one in four inbound enquiries, and it is almost never about budget. It is because the company asking has a demand problem, a product problem, or a timeline problem, and SEO is the wrong tool for all three. Saying so on the first call costs us a deal and saves them nine months.
This page is that conversation, written down. It covers the conditions that make an agency worth it, the conditions that make it a waste, and the cheaper things to buy while you wait for the first set to be true.
When do you actually need an SEO agency?
When three things are true at once: search already produces something you can measure, you know broadly what needs building, and your team cannot get it done fast enough. Two out of three is not enough.
- Organic search already drives leads or sales, and the growth curve flattened because nobody has time to work on it
- You are replatforming or migrating, and the downside of getting it wrong is bigger than the retainer
- Competitors are being named in AI answers and buyer shortlists while you are invisible
- You have budget for content but nobody who can decide what to write, for whom, and why
- Search is a board-level channel and no single person in the building owns it
- You need specialists in four disciplines and cannot justify four salaries
That last point is the real economic case for agencies. Technical SEO, content strategy, writing and digital PR are four different skill sets. Hiring one generalist to cover all four is how companies end up with a person who is decent at writing, out of their depth on log file analysis, and quietly miserable.
If you want the mechanics of what that team actually does with your money, what is an SEO agency breaks down the four workstreams and where the hours go.
When do you not need an SEO agency?
When you need revenue this quarter, when nobody can approve content, or when nobody is searching for what you sell. Each of these kills a retainer, and none of them is the agency's fault.
Situations where hiring an SEO agency reliably disappoints, and what solves the actual problem instead.
| Your situation | Why an agency fails here | Buy this instead |
|---|---|---|
| You need pipeline in the next 60 days | Meaningful organic movement takes three to six months minimum | Paid search, outbound, or partnerships |
| Nobody has searched for your category yet | You cannot rank for demand that does not exist | Category education through paid social, PR and community |
| No internal approver for content | Briefs and drafts stall for weeks and the retainer burns anyway | A part-time editor or content owner first |
| Under 500 organic sessions a month with no product-market fit | You will optimise a funnel that does not convert | Conversion work and customer research |
| The retainer would be most of your marketing budget | Single-channel dependency with a six-month payback | A one-off audit plus in-house execution |
The approver problem is the most common and the least discussed. An agency can produce twelve excellent pages a month, but if nothing publishes without a founder's review and the founder is closing deals, you are paying for a queue. Fix the bottleneck before you add supply to it.
Ask yourself one question before signing anything: if the agency delivered perfect work next month, could we get it published, deployed and linked within two weeks? If the answer is no, that is the problem to solve first.
What should you do instead if you are not ready?
Buy diagnosis, not delivery. A one-off audit plus a few hours of senior direction costs a fraction of a retainer and tells you whether the retainer would have worked.
Cheaper alternatives to a full retainer, in rough order of how early you should consider them.
| Option | Typical cost | What it gets you |
|---|---|---|
| One-off technical audit | $2,000 - $8,000 once | A prioritised list of what is broken and what it is costing you |
| Consulting hours | $200 - $450 per hour | Direction for an internal team that can already execute |
| Freelance writer with a strategist's brief | $1,000 - $3,000 a month | Steady publishing without agency overhead |
| AI visibility baseline | $1,500 - $4,000 once | Where you stand in ChatGPT and Perplexity before spending anything |
| Doing it yourself for two quarters | Time only | Enough knowledge to hire well later, and better questions on sales calls |
The audit route is the one we recommend most often to companies that are not ready. A proper SEO audit tells you whether your problem is technical, structural or editorial, and you can hand the output to a freelancer, an in-house hire, or an agency later without repeating the work. Pair it with a few hours of consulting if your team needs direction more than hands.
If AI visibility is the specific worry, the cheapest useful step is a baseline: a locked prompt set, sampled across engines, so you know whether you are absent or merely under-cited. That standalone exercise is described in which agency is best for AEO reporting, and the vocabulary in what is AEO.
Is an agency, a freelancer, or an in-house hire the right shape?
Freelancer for one clear gap. In-house when search is a permanent core channel. Agency when you need several disciplines at once and cannot keep four specialists busy.
The three delivery shapes compared on annual cost and what each one is genuinely good at. Fully loaded means salary plus tools, benefits and management overhead.
| Shape | Annual cost | Strong at | Weak at |
|---|---|---|---|
| Freelancer | $10,000 - $50,000 | One discipline, low overhead, fast start | Coverage, holidays, and anything outside their specialism |
| In-house generalist | $85,000 - $170,000 fully loaded | Institutional knowledge, availability, cross-team relationships | Depth in four disciplines nobody can hold at once |
| Agency retainer | $30,000 - $180,000 | Parallel workstreams, senior strategy, pattern recognition across clients | Context depth, and staying interested in a small account |
| Hybrid: in-house lead plus agency | $120,000 - $250,000 | Most reliable model above roughly $5m revenue | Costs the most, and needs the internal lead to be genuinely capable |
The hybrid row is what most companies end up with once search matters enough to defend. One internal owner who knows the business, plus an agency supplying specialist capacity, beats either alone. The internal person stops the agency drifting; the agency stops the internal person guessing.
How much do you need to be able to spend?
Enough to fund twelve months, not three. SEO spend that stops at month four buys you the cost with none of the compounding, which is the most common way this budget gets wasted.
The affordability test worth running before your first call
Minimum viable commitment = monthly retainer x 12 Payback check = (target monthly organic revenue x 12) / minimum viable commitment
Target monthly organic revenue is what you realistically expect a year out, not today's number. If the payback check comes out under 2, either the retainer is too big for the opportunity or the opportunity needs a bigger site than you can build in a year. Both are reasons to start smaller.
Sanity-check the retainer size against your category too. Enterprise programs cost more because coordination costs more, and the arithmetic behind those numbers is in enterprise SEO pricing. Smaller companies looking at the same page usually discover their real budget belongs one tier down.
What being ready looks like on your side
Readiness is mostly organisational, not financial. The companies that get the most from a retainer have four things sorted before the kickoff call, and none of them cost money.
- One named internal owner with authority to approve content and prioritise tickets
- Analytics and conversion tracking that you trust enough to argue with
- A developer or agency partner who can ship changes within a two-week cycle
- Agreement at leadership level on what success looks like twelve months out
Miss the first one and everything else slows down. An agency without an internal counterpart spends its hours chasing approvals instead of doing the work you are paying for, and by month three both sides are frustrated for the same reason.
How long should you expect to work with an agency?
Eighteen months to three years for the model to pay off properly, with a fair break clause after the first six. Anything shorter is a project, and should be priced as one.
What a healthy engagement looks like over time, and the question to ask yourself at each stage.
| Stage | What should be happening | The honest check |
|---|---|---|
| Months 1-3 | Audit, fixes shipped, baseline set, first content live | Is work actually shipping, or only being recommended |
| Months 4-6 | Ranking movement on secondary terms, indexation improving | Can they show leading indicators, not just activity |
| Months 7-12 | Compounding traffic, conversions traceable to organic | Is the trend line above where we would have been anyway |
| Year 2 onward | Defending position, expanding into new clusters and AI surfaces | Are they still bringing ideas, or maintaining a spreadsheet |
Year two is where most engagements quietly deteriorate. The exciting fixes are done, the account gets handed to a more junior team, and the monthly call turns into a status update. The signal is easy to read: when the agency stops proposing things you had not thought of, the relationship has become maintenance, and you should either reset the scope or move on.
How will you know it is working?
Leading indicators in the first quarter, traffic in the second, revenue in the third and fourth. If none of the leading indicators move by month three, something is wrong regardless of what the report says.
- Month 1-2: crawl errors down, indexed pages up, page speed improved, tickets actually merged
- Month 2-3: impressions rising in Search Console even before clicks do
- Month 3-4: non-branded keywords entering the top 20, first AI citations appearing on tracked prompts
- Month 5-8: organic sessions and conversions trending up against the pre-engagement baseline
- Month 9-12: organic contribution to pipeline or revenue large enough to argue about in a board meeting
Impressions are the underrated early signal. They move weeks before clicks and tell you the site is being surfaced for new queries. On the AI side, tracked citation rate plays the same role, which is why AI visibility reporting belongs in the monthly pack from month one rather than being added later when someone asks about ChatGPT.
What if you hire one and it is not working?
Have the direct conversation at month four, not month nine. Most underperforming engagements are fixable, and the ones that are not should end before you have spent another $40,000 finding out.
Ask three questions in that conversation. What did we ship versus what we planned? Which of the delays were on your side and which on ours? What would you change if this were your budget? A good agency answers all three specifically, including the parts where they were the bottleneck. A weak one talks about algorithm updates.
If you do move on, take the assets with you: content, tracked keyword and prompt sets, dashboards, and any accounts created in your name. Getting that written into the contract on day one costs nothing and saves an awkward negotiation later. The clauses worth insisting on are listed in how to choose an SEO agency.
So do you need one?
If search already makes you money and you cannot keep up, yes. If you are hoping an agency will create demand that does not exist yet, no, and no amount of budget changes that. The distinction is worth being honest about before anyone signs anything.
When you are ready to compare providers, the category framework is in the best SEO agency for the US market, and named providers are surveyed in our best SEO agencies roundup. Other questions people ask before hiring are collected in the answers library.
And if you would rather have someone look at your numbers before you decide, ask for a proposal. We will tell you if the honest answer is to wait six months and spend the money elsewhere.