Most businesses cannot tell you what their SEO returned last year. They can tell you what it cost and show you a rankings report, but SEO ROI, the money that came back against the money that went out, usually goes unmeasured.
What SEO ROI actually measures, and why most reports get it wrong
SEO ROI is the profit generated by organic search, set against everything you spent to generate it. It is the standard return on investment calculation applied to one channel. That sounds simple, but the reason so few businesses can produce the figure is that both halves of the sum are harder to pin down than they look.
On the return side, organic search rarely gets clean credit. Someone finds you through a non-branded search, leaves, clicks a retargeting advert two days later and buys. Under a last-click model the advert takes all the credit, and the search that introduced you to that customer gets none. On the cost side, businesses routinely count the agency retainer and forget the developer time, the content production, the tools and the staff hours spent reviewing it all.
The result is a number that is wrong in both directions at once, which is why so many SEO conversations collapse into arguing about rankings instead. Rankings are not a return. A page can rank first for a term nobody searches with commercial intent and earn nothing. Since AI Overviews arrived, a page can also rank well and lose clicks to the summary above it. What follows is how to build the figure properly, how to forecast it before you spend, what a realistic result looks like, and when the honest answer is that SEO is not worth it for your business.
Six inputs you need before you can calculate SEO ROI
Total organic revenue
Revenue attributable to organic search over a defined period, not sessions or conversions. For ecommerce this comes straight from analytics. For lead generation you need to trace enquiries through to closed deals, which usually means connecting your CRM rather than trusting the analytics figure alone.
Your true gross margin
ROI on revenue flatters every channel. If you sell at a 30% margin, £100,000 of organic revenue is £30,000 of gross profit, and that is the number to compare against your costs. Skipping this step is the single most common reason SEO ROI claims look implausible.
Fully loaded SEO cost
Agency fees or salaries, plus content production, plus developer implementation time, plus tools. If an in-house marketer spends two days a month on SEO, that is a real cost and belongs in the calculation.
A defined measurement window
SEO spend and SEO return do not land in the same month. Comparing this month's cost against this month's revenue understates early performance and overstates it later. Use a rolling twelve months once you have the data.
A baseline to measure against
Some organic revenue would have arrived regardless, particularly branded search from people who already know you. Separating branded from non-branded traffic is what stops you claiming credit for demand you did not create.
Customer lifetime value
If customers buy repeatedly, first-purchase revenue badly understates what organic search delivered. A subscription or repeat-purchase business measuring only the first transaction will conclude SEO lost money when it comfortably did not.
The SEO ROI formula and how to apply it
The calculation itself is simple once the inputs are honest:
SEO ROI = (Gross profit from organic search - Total SEO cost) / Total SEO cost x 100
Worked through with realistic numbers: a business generates £120,000 in organic revenue over twelve months at a 40% gross margin, giving £48,000 in gross profit. It spent £18,000 on an agency retainer, £4,000 on content and £3,000 of internal developer and marketing time, so £25,000 all in.
That gives (£48,000 - £25,000) / £25,000 x 100, which is 92%. Just under two pounds back for every pound spent, measured on profit rather than revenue.
Run the same figures on revenue instead of margin and you would report 380%. That is the version most online calculators and AI-generated answers give you, because the common formula puts revenue in the top line. It is the sort of number that circulates in case studies and means very little. Always state which basis you are using, and if a proposal quotes an ROI figure without saying whether it is revenue or profit, ask before you sign anything.
A simple SEO ROI calculator you can build in a spreadsheet
You do not need a paid tool. Ten rows in a spreadsheet will give you a more honest answer than most online SEO ROI calculators, because you control the inputs. Set it up like this, one row per line, with a column per month:
- Non-branded organic revenue for the month, from analytics or your CRM.
- Gross margin %, your blended margin on the products or services organic search sells.
- Gross profit from organic: row 1 multiplied by row 2.
- Repeat revenue uplift, if customers buy again: expected repeat purchases in twelve months multiplied by margin.
- Agency or freelance fees.
- Content production, including writers, design and video.
- Developer and internal staff time, at a realistic hourly cost.
- Tools and subscriptions used for SEO.
- Total SEO cost: rows 5 to 8 added together.
- ROI %: rows 3 plus 4, minus row 9, divided by row 9, multiplied by 100.
Add two summary cells. The first is a rolling twelve-month ROI, which smooths out the front-loaded costs. The second is your break-even organic revenue: total cost divided by margin. In the worked example above, £25,000 divided by 40% is £62,500. Below that level of organic revenue the work is losing money; above it, every extra pound of organic revenue is earning 40p of profit for a cost that does not rise with it.
The break-even figure is the most useful number in the sheet, because it turns a vague question ("is SEO working?") into a specific one ("will organic search produce more than £62,500 of revenue this year?").
How to forecast SEO ROI before you spend
A forecast will never be precise, but it tells you whether the opportunity is big enough to justify the cost. Work from your own data wherever you can rather than industry averages:
- List the commercial queries you want to win and their monthly search volumes from a keyword tool. Leave out informational queries that are unlikely to lead to a sale.
- Estimate a click-through rate for the position you can realistically reach. Use your own CTR at that position from Search Console's performance report, not a generic curve. CTRs vary heavily by query type, and AI Overviews have pushed them down on many informational searches.
- Apply your organic conversion rate and average order or deal value from analytics and your CRM.
- Apply your gross margin to get forecast gross profit.
- Compare against the cost and the break-even figure from your spreadsheet.
An illustrative example: a cluster of commercial queries with 2,000 searches a month, where your own data shows a 5% CTR at the position you expect to reach, gives roughly 100 visits a month. At a 2% conversion rate that is two sales, and at a £400 order value with a 40% margin, about £320 of gross profit a month, or £3,840 a year. If the work to win that cluster costs £8,000, it will not pay back in year one on that cluster alone. That is a useful answer to have before you start, and it tells you to look for bigger clusters, higher-value queries or a cheaper approach.
Build three versions: cautious, expected and optimistic. If the cautious case still breaks even within eighteen months, the investment is defensible. If only the optimistic case works, it is a gamble.
How to attribute revenue to organic search without fooling yourself
Attribution is where most SEO measurement quietly falls apart. It helps to know exactly what Google Analytics 4 does. According to Google's own attribution documentation, GA4 now offers three models: data-driven, paid and organic last click, and Google paid channels last click. The first click, linear, time decay and position-based models were removed in November 2023. All models exclude direct visits from credit unless the whole path was direct.
Four practical steps get you close enough to make decisions on.
Split branded from non-branded organic. In Search Console, use a custom regex filter to separate queries containing your company name and its misspellings. Branded search largely reflects demand created elsewhere: your adverts, your reputation, word of mouth. Non-branded search is the demand SEO went out and captured. Judge the work on the second number.
Check your attribution settings. In GA4's attribution settings, the default key event lookback window is 90 days (30 days for first visits), and it can be shortened to 30 or 60. Considered purchases and B2B enquiries routinely take weeks between the first organic visit and the sale, so make sure nobody has shortened it, and note which reporting model the property uses before you compare any figures across periods.
Look at the path, not just the last step. Data-driven attribution gives organic search partial credit when it started a journey that ended on a paid click. Compare organic's conversions under data-driven and last-click models. A big gap tells you organic search is introducing customers that other channels then close.
Close the loop with your CRM for lead generation. Analytics can tell you a form was submitted. It cannot tell you the enquiry was unqualified, or that it became a £40,000 contract. If you sell services, tracking organic leads through to revenue in the CRM is the difference between a real ROI figure and a guess. It usually reveals that organic leads close at a different rate to paid ones, which changes what you should spend where.
None of this needs to be perfect. It needs to be consistent, so that comparisons between periods and between channels mean something.
Measuring SEO ROI when AI Overviews take the click
AI-generated answers have changed the relationship between ranking and traffic. A Pew Research Center analysis of the browsing data of 900 US adults in March 2025 found that people who saw an AI summary clicked a traditional search result in 8% of visits, against 15% when no summary appeared, and clicked a link inside the summary itself in just 1%. That is US data, but AI Overviews run in UK search too, so the same pressure applies to informational queries here.
For ROI measurement, that means three adjustments:
- Expect impressions and clicks to diverge. Rising impressions with flat clicks is not necessarily failure. It may mean you are visible in searches that the summary answers. Judge the commercial queries, which are less affected, separately from the informational ones.
- Count AI referrals as organic-adjacent. Visits from ChatGPT, Perplexity, Gemini and Copilot show up as referral traffic in GA4, not as organic search. Our guide to tracking AI search traffic in GA4 shows how to group them, and AI citation tracking covers measuring mentions that never produce a click.
- Ask new customers where they found you. A "how did you hear about us" field on enquiry forms catches people who saw your brand in an AI answer and then searched for you by name, which otherwise looks like branded search with no cause.
If a growing share of your demand is being answered by AI, it may be time to treat AI visibility as its own workstream. That is what generative engine optimisation covers, and our comparison of AEO and SEO explains where the two overlap.
Leading indicators to track before ROI turns positive
Revenue is the final measure, but it lags the work by months. These SEO KPIs tell you whether the work is heading the right way long before the ROI figure turns positive:
- Months one to three: technical fixes shipped, pages indexed in Search Console, and impressions on your target non-branded queries starting to appear.
- Months three to six: average position on commercial queries improving, non-branded clicks rising, and organic landing pages converting at a rate similar to your other channels.
- Months six to twelve: organic leads or sales growing, cost per organic acquisition falling, and the rolling ROI moving towards break-even.
If the leading indicators are flat at six months, that is the time to ask hard questions, not month eighteen. Our data-driven marketing guide covers how to set up reporting that makes these trends visible.
How long before SEO ROI turns positive
Honestly: usually somewhere between six and twelve months, and longer in competitive sectors or on a new domain.
The shape of the return matters more than the timeline. SEO costs are largely front-loaded (technical fixes, content, site structure) while returns accumulate slowly and then persist. Measured monthly, the first half-year almost always looks like a loss. Measured over three years, the same work often outperforms other channels, because the traffic keeps arriving after the spending slows.
That asymmetry is also the risk. If your business needs leads this quarter to survive, SEO is the wrong first investment regardless of its eventual return, and paid search is the honest recommendation instead.
What a good SEO ROI looks like
Be sceptical of universal benchmarks. The figures circulated online vary wildly, are rarely stated on a profit basis, and almost never disclose their sample. Your own numbers are the only meaningful comparison, and they should be judged against three things:
- Your other channels. If paid search returns more per pound at your current scale, spend there first. SEO earns its place when it beats the alternatives, not because it is fashionable.
- Your own trend. ROI improving quarter on quarter matters more than the absolute figure, because the cost base is front-loaded and should fall as a share of return over time.
- The cost of doing nothing. If competitors are capturing non-branded demand in your category and you are not, that gap widens with time and gets more expensive to close.
For context on the investment side in this market, our guide to SEO costs in the UK sets out realistic pricing.
Is SEO still worth it in 2026?
For most businesses whose customers search for what they sell, yes, but the bar is higher than it was. AI summaries take clicks from simple informational queries, so the value has concentrated in commercial and local searches, in questions too specific for a summary to settle, and in being the source the summary cites. Equally, there are businesses we tell not to bother, and it is worth being direct about which:
- Nobody is searching for what you sell. Genuinely new categories have no existing demand to capture. You need awareness first, and that is a paid social or PR problem.
- You need revenue within three months. SEO will not arrive in time. Paid search will.
- Your margins cannot carry it. If gross profit per customer is small and volumes are low, the maths may never work. The break-even figure will tell you quickly.
- Your demand is mostly answered on the results page. If your queries are definitions and quick facts, the summary may take most of the clicks you would have earned.
- Your website cannot convert. Sending more traffic to a site that converts at 0.2% multiplies a problem instead of solving one. Fix the conversion rate first.
- You cannot commit past six months. Stopping at month five means paying almost the entire cost and collecting almost none of the return. That is the worst possible outcome, and it is common.
Local service businesses are a partial exception on the timeline. Competition is thinner and results arrive faster, which our local SEO playbook covers in detail. The same is often true for smaller operators, addressed in our guide to SEO for small businesses.
SEO ROI compared with paid search ROI
These are different financial instruments, and comparing them on a single month's figures is misleading.
Paid search is rented traffic. Returns are immediate, measurable and stop the day you stop paying. The cost per acquisition tends to rise as you scale, because you buy progressively less relevant clicks.
SEO is an owned asset. Returns are slow, harder to attribute and continue after spending pauses. The cost per acquisition tends to fall as you scale, because a page that ranks costs the same whether it brings 100 or 10,000 visits.
In practice most businesses should run both, with the balance set by cash flow and time horizon rather than ideology. Paid search funds the present and produces the keyword data that tells you which terms actually convert. SEO compounds and lowers your blended acquisition cost over time. The same discipline applies to every paid channel, as our Microsoft Ads ROI guide shows. If you want the paid side handled alongside SEO, that is our Google Ads management.
Common mistakes when calculating SEO ROI
- Measuring revenue instead of gross profit. Inflates every figure, sometimes by a factor of three or more.
- Trusting a calculator you cannot inspect. Most online SEO ROI calculators assume revenue, a generic CTR curve and no internal costs.
- Counting branded search as an SEO win. You did not earn traffic from people already looking for you by name.
- Omitting internal time and developer costs. The retainer is rarely the full cost.
- Comparing periods measured under different attribution settings. A changed model or lookback window can move organic's numbers without anything real changing.
- Judging at three months. You are measuring the investment phase and calling it a result.
- Using rankings or traffic as the outcome. Neither pays wages. Track them as leading indicators, not as returns.
- Ignoring lifetime value. Repeat-purchase businesses that measure first orders only will systematically understate what organic search delivered.
Underneath all of this sits a simpler point that Google makes in its own SEO starter guide: search engines reward pages that genuinely serve the person searching. Work that chases rankings without serving demand tends to produce traffic that never converts, and traffic that never converts has no return to measure.
Making the number work for you
Calculate it properly once and SEO stops being a matter of faith. You will know what it costs, what it returns, how that compares with paid, and whether the trend justifies continuing. That is a decision you can defend to a board or to yourself.
If you want that measurement built in from the start rather than reconstructed a year later, our SEO services include reporting tied to revenue rather than rankings. If you are still weighing up whether organic search deserves a place in your plan at all, our guide to why SEO matters for business is the better starting point.
SEO ROI FAQs
What is a realistic SEO ROI?
It varies far too much by sector, margin and competition for a single benchmark to be useful, and most published figures are calculated on revenue rather than profit, which inflates them. Measure your own on a gross-profit basis over a rolling twelve months, work out your break-even organic revenue (total cost divided by margin), and compare the result against your other channels rather than against industry claims.
How do you calculate SEO ROI?
Subtract your total SEO cost from the gross profit generated by organic search, divide by the total SEO cost, then multiply by 100. The difficulty is not the formula but the inputs: use gross profit rather than revenue, include internal and developer time in the cost, exclude branded search you did not earn, and measure over at least twelve months.
Is SEO still worth it in 2026?
For most businesses whose customers search for what they sell, yes, but the value has shifted. AI summaries take clicks from simple informational queries, so the return now sits mainly in commercial and local searches, in detailed questions a summary cannot settle, and in being a source that AI answers cite. It is not worth it if nobody searches for your product, you need revenue within three months, or your website cannot convert the traffic.
How long does SEO take to pay for itself?
Typically six to twelve months, longer on a new domain or in a competitive sector. Costs are concentrated early and returns accumulate later, so monthly measurement will show a loss well before the work has failed. Track leading indicators such as non-branded impressions, positions and clicks in the meantime, and judge ROI on a rolling twelve-month view.
Why does Google Analytics undercount organic search?
Mostly because of the attribution model. Under last-click attribution, a sale goes to whichever channel was clicked last, and organic search is often the first touch rather than the last: someone finds you through a search, then returns through a paid advert and buys. GA4's data-driven model shares credit more fairly, so compare the two models. Leads that convert offline also go uncounted unless you connect your CRM, and visits from AI assistants appear as referrals rather than organic search.
How do AI Overviews affect SEO ROI?
They reduce clicks on informational searches. Pew Research Center found US users clicked a traditional result in 8% of visits when an AI summary appeared, against 15% when it did not. Rising impressions with flat clicks may therefore reflect visibility rather than failure. Judge commercial queries separately, track AI assistant referrals in GA4, and ask new customers where they heard about you.
