An SEO budget is the amount a business puts into organic search each year: people's hours, content, technical work, links and software. Deciding how to invest in SEO means answering two separate questions. How much to put in. And where to put it.
On the first there are dozens of price lists, including our own on how much SEO costs in the UK. On the second there is far less. On 10 October 2026 we read the first page of Google UK for "seo budget" and "how much to spend on seo". There's an SEO cost calculator, UK agency pages with monthly price bands (£1,500 to £5,000, £500 to £5,000+), a 2022 page recommending 5-10% of revenue, and two guides, from Surfer and Search Engine Land, that break spend down by category. Surfer gives percentage ranges per category; Search Engine Land gives budget ranges by type of business. Neither changes the split for a new site or a site with technical problems, and neither says in what order the money goes out over the first twelve months.
This guide starts there. You'll find two calculations for setting the figure, a split for three different site situations and a spending calendar built on our projects.
The five items an SEO budget ends up in
Any SEO budget, in-house or handed to an agency, breaks down into five items: people who think, content, technical work, external authority and software. The proportions change, the list doesn't.
Search Engine Land uses a similar division in its guide to SEO budgets (people, technology, content, technical SEO, brand and off-page). The practical difference for a UK business is that the items don't carry a fixed weight: software costs the same in Manchester as in San Francisco, hours don't.
| Item | What it buys | A unit you can check | Sign it's underfunded |
|---|---|---|---|
| Strategy and analysis | keyword research, priorities, reading the data | hours of an experienced person | pages get produced without knowing which searches they're meant to capture |
| Content | new pages, rewrites, updates | 3 hours 20 minutes per article on average (Orbit Media, 2026) | the pages exist but none answers better than the competition |
| Technical | crawling, indexing, speed, structure | a crawler: Screaming Frog costs $279 a year (about £211), and without a licence it crawls up to 500 URLs | important pages out of the index or slow to load |
| Authority | editorial links, digital PR, citations | a single editorial link in the UK from about £325 (guest post) to over £2,000 | content better than competitors' that stays behind |
| Software | keyword, ranking and backlink data | Ahrefs Lite $129 a month (£97.54), Semrush SEO plan $139 a month (£105.10) | decisions made on gut feeling, with no history |
The software row deserves a sum of its own. An Ahrefs Lite subscription plus a Screaming Frog licence spread over twelve months comes to just over $150 a month, about £115. On a £700 monthly budget that's roughly a sixth of the spend, before a line has been written or an error fixed. On a £5,000 budget it's just over 2%. That's why on small budgets one complete tool is the better choice, or working with someone who already has the tools: we compare the options in our guide to the best SEO tools, and if you want to spend as little as possible you'll find the alternatives among cheap SEO tools.
How much to put in: two calculations that check each other
You find the right figure for SEO by crossing two calculations: one starts from the overall marketing budget, the other from the value of the searches you want to capture. If the two numbers are miles apart, one of the assumptions is wrong.
Top down: what share of marketing goes to organic
The most quoted reference on marketing budgets is Gartner's CMO Spend Survey. In the 2026 edition marketing is worth 7.8% of revenue, against 7.7% in 2025, and 56% of marketing leaders say they don't have enough budget for their strategy.
Watch the sample: 401 CMOs in North America, the UK and Europe, almost all at companies with more than $1 billion in revenue. It doesn't describe a UK SME, and we haven't found an equivalent published figure for UK small businesses.
The breakdown is more useful than the 7.8%. In the 2025 edition paid media took 30.6% of marketing budgets. It's a good place to start thinking: if you already spend on Google Ads, the question isn't how much to add, but which part of that spend you'd like to be coming from organic search, with no cost per click, in two years' time. We cover how the two channels relate in SEO vs PPC.
Bottom up: what the searches you want are worth
The second calculation starts from keywords. Shopify, in its guide to the return on SEO, uses this formula: search volume times CTR, times conversion rate, times order value. For a business that sells services, order value becomes the margin on a new customer.
Here's an example with made-up but plausible numbers. A B2B company identifies 20 commercial searches that bring in customers, 2,400 searches a month in total. If it reaches the top positions and takes 10% of the clicks, it gets 240 visits. At a 2% enquiry rate that's almost 5 enquiries a month; if it wins one in four and each customer is worth £2,500 in margin, the traffic is worth about £3,000 a month once it's up to speed.
That number needs correcting before you use it, because not every click reaches a website. According to SparkToro's study of Datos data (September 2022 to May 2024), 59.7% of Google searches in the European Union and 58.5% in the US end without a single click; the study has no separate UK figure. On informational queries the AI Overview adds to this: Ahrefs measured across 300,000 keywords that its presence cuts the CTR of the first position by 58% (desktop data, December 2025 against December 2023). We go into it in the article on zero-click searches.
The value once the traffic is up to speed is the ceiling on spend, not the target. Our rule: if the annual budget is more than half the value the traffic will eventually produce, the return arrives after the second year. In the example, £3,000 a month justifies up to about £1,500 a month of investment. Above that, the plan needs rethinking. We explain how to turn this number into a return calculation in our guide to SEO ROI.
How to split it according to where the site starts
How you split an SEO budget depends on the state of the site more than on the sector. A new site needs content, a site with technical errors first needs Google to read it properly, a healthy site in a crowded niche needs authority.
TripleDart, a US agency specialising in SaaS, proposes a fixed split: 30-50% on people, 20-30% on content, 10-20% on links, 10-20% on tools, 5-10% on technical work. Surfer's guide, on the first page of Google UK for "seo budget", uses the same ranges plus 2-5% for training. They make sense for US budgets of thousands of dollars a month. On a £1,000 UK budget the software share rises on its own, and one percentage for every site hides the thing that matters: where you start from.
This is the split we start from in the first twelve months, before adapting it to the data of each site.
| Item | New or rebuilt site | Site with technical problems | Healthy site in a crowded niche |
|---|---|---|---|
| Strategy and analysis | 15% | 20% | 15% |
| Content | 45% | 25% | 35% |
| Technical | 25% | 40% | 10% |
| Authority (links, digital PR) | 5% | 5% | 30% |
| Software | 10% | 10% | 10% |
Why so little authority on a new site? Because links point to pages. If the pages that answer the right searches don't exist yet, money spent on link building goes into strengthening a home page that won't rank for anything useful. In a niche where everyone has decent content, on the other hand, what separates position one from position five is often who gets cited from outside: that's where digital PR becomes the biggest item.
The site with technical problems is where the most money gets wasted. Articles are commissioned that Google doesn't index or that sit on slow pages. Fix the technical SEO first, then write. To know which of the three columns you're in you need an SEO audit: it's the first spend to make and the one that decides all the others.
The calendar: what to fund in the first twelve months
An SEO budget isn't spent evenly across the year. Technical work is concentrated at the start, content in the middle, authority once there are pages to push.
| Period | Where most of the spend goes | What to measure to see if it's working |
|---|---|---|
| Months 1-2 | audit, technical fixes, keyword research | indexed pages, errors resolved in Search Console |
| Months 3-6 | new pages and rewrites on commercial searches | impressions and keywords on page one |
| Months 6-12 | updates, links to the pages that are climbing | clicks, enquiries, rankings on the keywords that bring customers |
Google itself, in its SEO Starter Guide, warns that some changes take effect within a few hours and others take several months. That's why the first monthly fees feel like money with nothing coming back.
At Sunscape, a company that develops solar installations in Italy, organic impressions went from 329 in March to 5,809 in August 2026. Over the same period keywords on page one rose from 1 to 38 out of 143 tracked, and monthly organic clicks from 46 to 195. Impressions moved before clicks: anyone judging the budget on enquiries in month three would have closed the project just as it started to pay. The full case is on the Sunscape project page.
Over the long run the sums change. At Macropix, an Italian LED wall manufacturer, the keyword "monitor pubblicitario" (advertising monitor) went from position 88 to position 2 between 2020 and 2025 according to Semrush, as we describe in the case on SEO for manufacturing. Five years of continuous budget. How long it takes on average to see the first results we measured in a separate article on how long SEO takes; what matters here is a practical consequence: the first year should be funded in full, or not started.
When to move money from one item to another
The initial split should be reviewed every quarter, looking at three signals in the data. Each one points to an item to increase or cut back.
Impressions rising, clicks flat. The pages are seen but not chosen. The cause may be titles and descriptions, or an AI Overview answering in your place. In the first case you need hours of optimisation; in the second it pays to move budget from informational content to commercial content, where the user still has to land on a website. We explain how search changes with generated answers in the article on how AI Overviews affect SEO.
Lots of pages, few doing the work. If 20% of the content brings in almost all the traffic, the content budget should move from production to updating and pruning. The method is in our guide to content pruning.
Content better than the competition stuck on page two. That's the sign authority is missing. Take something away from new content and fund links to the pages that are already good.
There's also a signal coming from the industry. The Orbit Media 2026 survey says 92.4% of bloggers use AI to write and only 13.9% report strong results, six points below the previous low in twelve years of surveys. Producing more content costs less and less, and that's exactly why it returns less and less. Search Engine Land reaches the same conclusion: less budget on mass-produced content, more on original research, your own data and content written by people who know the trade.
Where an SEO budget gets wasted
Some costs turn up in almost every site we inherit from other suppliers. We list them because they're the first to cut.
- Duplicate subscriptions. Two tools that do the same thing, paid for by different departments. Search Engine Land suggests reviewing your tool stack every quarter: whoever does it almost always finds a licence too many.
- Fixed-price link packages. Twenty links from a list of sites that's the same for every client move nothing and can attract a penalty.
- Articles by volume. Four articles a month on topics chosen without keyword research, because the contract says four articles.
- Reports without decisions. A thirty-page monthly document of rankings that nobody uses to change the following month's split.
If you don't know whether your current budget is working, the first step is to find out whether your SEO agency is working and which SEO KPIs to watch. If you have to defend the budget in front of the board, the part on presenting the numbers is in how to prove the value of SEO.
If you'd rather hand the split to someone who reviews it every month against the data, our SEO service page explains how we work and on what timescales.
The item no quote includes
In almost every project the underfunded item doesn't appear in the budget: it's the hours of whoever in the business knows the product and the customers. Content that beats the competition comes from a phone call with the technical or sales lead, from an internal figure, from a real case. If that person has no time, the agency writes with the information it finds online, which is the same information the competitors have. The Orbit Media survey notes that among people writing for the web, the practices that have historically paid off most, such as working with experts and original research, are the ones in decline. The advice we give before signing is simple: budget two or three hours a month of someone internal as well. They cost nothing on the invoice and change the value of every other item.
Frequently asked questions
It depends: on the state of the site and on the value of the searches you want to capture. The UK pricing guides on the first page of Google put most retainers between £500 and £5,000 a month, as we show in how much SEO costs. The check to make is bottom up: estimate what the traffic from your commercial keywords will be worth once it is up to speed, and don't invest more than half of that value each year, or the return arrives after the second year.
Yes, but with the budget moved around. 59.7% of Google searches in the EU and 58.5% in the US end without a click (SparkToro on Datos data, 2024). On informational queries with an AI Overview, Ahrefs measured a 58% drop in first-position CTR. So it pays to give more weight to commercial searches and to content with your own data, which generated answers cite but cannot replace.
It depends: on timing. Google Ads brings traffic from day one and stops when you stop paying; SEO takes months and then stays. In the companies in Gartner's 2025 sample, paid media takes 30.6% of the marketing budget: the useful question is which part of that spend you would like to replace with organic traffic in two years. The full comparison is in SEO vs PPC.
Yes, if you concentrate it. On £700 a month, an Ahrefs Lite subscription plus a Screaming Frog licence already take about a sixth of the spend: one tool is enough, with a few commercial pages done properly and no fixed-price link packages. The low-cost options are in our guide to affordable SEO services.
It depends: on where the site starts and on the competition. At Sunscape, organic impressions went from 329 to 5,809 in five months, while monthly clicks rose from 46 to 195 in six. That is why the return should be calculated on the second year: the first is largely construction.
You shouldn't stop completely while searches are bringing in customers: competitors keep working, and rankings are lost faster than they are won. You can, however, reduce the budget and move it to maintenance. We cover this in when can I stop doing SEO.
