Cost and budget · 11 min read
What actually drives the price of SEO, and how to work out yours
This page contains no price ranges and no industry averages, because there is no honest source for either. What it does contain is the list of things that genuinely move the price, what each pricing model quietly rewards your agency for doing, and a method for arriving at a defensible number for your own situation.
Written by Zubair Afzal, FounderUpdated
Before we start
Why this page has no price ranges in it
Every page that answers this question with a range is guessing. The figures circulate between articles, get cited as if they were research, and end up anchoring people to a number that was invented to fill a table. We are not adding to that.
The reason a range cannot be honest is that the phrase "SEO" describes an amount of work that varies by more than an order of magnitude between two businesses that would describe their needs identically. The variables that move it are knowable, and they are listed further down this page.
Skayle does not publish pricing, and the reason is the same. A published price has to assume a scope we have not seen. It would overprice the straightforward work and underprice the difficult work, and underpriced difficult work is how engagements fail.
What follows is more useful than a number: the things that actually drive the cost, what each pricing model rewards your supplier for doing, and a method for arriving at a figure you can defend internally before you speak to anyone.
The variables
Nine things that genuinely move the price
Work through these against your own situation. Each one that applies moves the number up, and the combination explains most of the spread between quotes.
- Starting position. A healthy site needs improvement. A site carrying a bad migration, a manual action or years of accumulated technical debt needs repair first, and repair is slower than improvement.
- Competitive intensity. Not how many competitors, but how well funded they are and how long they have been at it. Displacing a rival who has invested for a decade is a different job from outperforming one who has not started.
- Number of markets and languages. Each additional market multiplies research, content and technical complexity, and international configuration is easy to get wrong and expensive to unpick.
- Site size and how many templates it has. Cost tracks unique templates and systems, not page count. A large site with consistent templates is often cheaper to work on than a small handmade one.
- How hard it is to ship a change. If a recommendation takes a quarter to reach production, the programme spends its budget writing documents. Development access is a cost variable disguised as a process detail.
- Content production volume, and who supplies the expertise. Content written from your specialists is faster and better. Content that requires the agency to acquire the expertise first is slower and costs more.
- Whether earned coverage is needed. In competitive categories, digital PR and link acquisition are frequently the largest single line, and they are the hardest to do at low cost without cutting corners that create risk.
- Speed. Compressing a plan into half the time costs more than double, because parallel work needs more people and more coordination between them.
- Governance overhead. Legal review, brand approval, multiple stakeholders and formal reporting are real work. They do not improve results and they must be paid for.
Four pricing models and what each one rewards
Every model creates an incentive. Knowing which one you have created is the most useful thing you can take from this page.
| Dimension | What you are buying | What it quietly rewards | Best fit | Watch for |
|---|---|---|---|---|
| Monthly retainer | Continuous capacity and attention across a programme. | Staying busy and staying engaged. Not necessarily finishing. | Ongoing programmes where priorities shift and continuity matters. | Activity reports replacing outcomes. Agree what the retainer is for, and review against that. |
| Fixed-scope project | A defined deliverable with a start and an end. | Efficiency, and staying inside the written scope. | Audits, migrations, restructures, one-off builds. | Scope-shaving, and a change-request process that reprices anything unexpected. |
| Hourly or day rate | Access to a specialist for a measured amount of time. | More hours. The unit of value is time spent, not progress made. | Advisory work, second opinions, supporting an in-house team. | Cost drifting upward without a scope to anchor it. Cap it. |
| Performance-based | A share of an agreed result rather than the work itself. | Whatever is easiest to attribute, including demand you already had. | Rare. Narrow, well-instrumented situations with a clean baseline. | Brand searches counted as wins, and avoidance of slow structural work with no attributable result this quarter. |
Reading a proposal
What a defensible SEO quote contains
Should be in there
- The work expressed as days by discipline, with the seniority of who does each part.
- What specifically gets delivered in month one, month three and month six.
- A named starting position and what has to be repaired before improvement is possible.
- What is excluded, and what would trigger a change in price.
- Who owns the accounts, the documentation and the content at the end.
Should make you ask questions
- Tiered packages named after metals, with the difference expressed as a number of keywords.
- A price that scales with your revenue rather than with the effort involved.
- Any predicted ranking position, traffic figure or return, presented as a forecast.
- A monthly fee with no description of what a month actually contains.
- A twelve-month lock-in with no defined deliverables inside it.
Do the working
How to arrive at your own number before you get a quote
This produces a figure you can defend internally and compare quotes against. It takes an afternoon.
Write down what the work actually is
Not "SEO". The specific jobs: fix the crawl issues, restructure the service pages, produce twenty pages of content, earn coverage in three publications, implement conversion tracking properly. If you cannot list them, buy an audit first — that is a smaller purchase with a clear output.
You get: A task list, not a service name
Estimate each task in days of skilled work
You will be wrong, and it does not matter. What you need is an order of magnitude. Is the content programme two days a month or twelve? Is the technical work a week or a quarter? Suppliers can correct your estimate, which is a far better conversation than asking what it costs.
You get: A rough monthly day count by discipline
Decide what a customer is worth to you
Take the value of a customer and your gross margin on it, and be explicit about whether you are using first-order value or lifetime value. Spending against lifetime value requires the cash to survive the gap between the two.
You get: A contribution figure per customer
Work out how many extra customers would justify the spend
Divide your candidate annual spend by contribution per customer. That gives the number of additional customers the programme must produce to break even. Illustration only, with placeholder numbers: if a customer contributes £2,000 and you are considering £40,000 a year, you need twenty. Put your own figures in.
You get: A break-even customer count
Sanity-check it against available demand
Ask whether twenty additional customers plausibly exist within the searches you could realistically compete for. This is the step that stops well-run programmes from being aimed at markets too small to pay for them, and it is the question most quotes never raise.
Set your budget by payback period, not appetite
Work out how many months you can fund before the programme needs to pay for itself. That number, not enthusiasm, is what determines whether you should be buying organic search at all right now.
You get: A funded runway in months
Questions
Cost questions people ask before buying SEO
Why will nobody publish SEO prices?
Because the same words describe wildly different amounts of work. "SEO for a ten-page site" and "SEO for a ten-page site with a broken migration, three languages and a CMS nobody can deploy to" are the same brief and not remotely the same job.
A published price has to assume a scope. That either overprices the simple cases, which loses good clients, or underprices the difficult ones, which produces underfunded work and a bad outcome for everyone. We would rather scope first and quote second.
How do I compare two very different quotes?
Convert both into days of work by discipline. Ask each supplier how many days per month go to technical work, content, links or digital PR, analytics and account management, and who does each. Then compare on that basis rather than on the headline figure.
A quote that cannot be broken down that way is not a quote, it is a price. The difference between two proposals is nearly always scope and seniority rather than margin.
Is a cheap SEO package ever worth buying?
Sometimes, for a narrowly defined job. A small local business that needs its business profile corrected and a handful of pages written can get real value from a modest, well-scoped engagement.
It stops being worth it when the price is low because the work is automated, outsourced without review, or consists of reporting on things that were going to happen anyway. Ask what would be delivered in month three specifically. Vague answers are the signal.
Which pricing model should I choose?
Match the model to the shape of the work. A defined piece with a clear end — an audit, a migration, a rebuild of the information architecture — suits fixed-scope project pricing. Ongoing programmes suit retainers. Occasional advice suits hourly.
Treat performance-based pricing with care. It sounds aligned but tends to reward whatever is easiest to attribute, which is often demand you already had.
What makes SEO more expensive for one business than another?
Mainly: the state of the site you are starting from, how well funded your competitors are, how many markets or languages you need, how difficult it is to ship a change, whether links and earned coverage are required, and how fast you want it done.
Speed is the one people underestimate. Compressing twelve months of work into six costs more than twice as much, because it needs more people working in parallel and more coordination between them.
Should the price go up as my business grows?
Only if the work does. Pricing based on your revenue rather than the effort involved is common and hard to defend, because it means paying more for the same days of work.
It is reasonable for costs to rise when scope expands — more markets, more pages, more channels to coordinate. Ask which of those changed before accepting an increase.
Want a scope before you talk about a price?
We would rather look at the site, the competition and the starting position and then tell you what the work is. If the honest answer is that the budget is better spent somewhere else, we will say that instead.
If we don't deliver the work we agreed to deliver for reasons within our control, you don't pay for the undelivered work. Read our guarantee
Last updated