SEO ROI

SEO ROI answers one question: did the money I put into SEO bring back more than I spent? The formula is dead simple: ((revenue - cost) / cost) x 100. Spend £5,000, generate £20,000 in organic revenue, and your ROI is 300%. But the hard part isn't the maths — it's what you count as revenue and what you count as cost. I've seen people report a 500% ROI by ignoring their own salary, the designer's time, and half the content spend. Then the finance team runs the real numbers and the SEO budget gets halved. Don't be that person.

On the revenue side, you need to decide which conversions count. An ecommerce order is easy. A newsletter sign-up or a whitepaper download? Assign a value. I use average lifetime value for leads that later convert. That's imperfect but better than ignoring them. The attribution model matters more than most people admit. Last-click from Google Analytics is the default, but organic search often assists rather than closes. If you use last-click, you'll under-report SEO's contribution by 30% or more. I've switched to a custom model that gives 40% weight to the first touch, 20% to middle touches, and 40% to the last touch. It's not perfect, but it aligns better with how my prospects actually research. You can build this in [Google Analytics](/google-analytics/) or using a tool like [Looker Studio](/looker-studio/) to blend data from Search Console.

Forecasting ROI is a separate skill. You estimate organic traffic from rank improvements, multiply by expected CTR from your [rank tracking](/rank-tracking/) data, then by conversion rate and average value. I've done this for content projects and the forecast was within 15% of actuals twice out of four times. The rest were humbling. The biggest variable is conversion rate — a small change there swings ROI massively.

SEO ROI matters most when you are asking for money or defending what you already have. If you work in-house, the CFO does not care about rankings. She cares about whether the £20,000 monthly SEO spend generates more than £20,000 in profit. If you work at an agency, clients will eventually ask: 'What am I getting for my money?' A ranking report won't cut it. You need to show a return.

I've used SEO ROI to justify hiring an additional content writer. I showed that the last three articles each produced £8,000 in attributed revenue, and a new writer at £30,000 per year would pay for themselves within four months. The budget was approved. I've also seen ROI used to kill a project. An expensive link-building campaign was generating £1,200 in revenue per month against a £2,500 monthly cost. Negative ROI. We stopped it. Without that number, the campaign would have run for another year.

The metric is also critical for [SEO reporting](/seo-reporting/) to stakeholders. A quarterly report that shows ROI trending from 150% to 220% is far more persuasive than one showing traffic up 12%. Combine it with [SEO KPIs](/seo-kpis/) like conversion rate and cost per acquisition. One tradeoff: time horizon. SEO takes months to compound, so a 90-day ROI can look weak. I always run a 12-month rolling ROI alongside the quarterly number. The longer view smooths out the seasonality and the slow ramp of new content. If you only report short-term, you'll underfund the work that pays off later.

Attribution is the biggest landmine. If your ecommerce site uses last-click, SEO may show a 50% ROI when it's actually 200%. I've seen this happen. Fix it before you present numbers. Use a [Conversion Rate Optimisation](/conversion-rate-optimization/) lens to audit your attribution setup. It's a boring hour of work that saves you from looking foolish in a board meeting.

First time I calculated SEO ROI, I only counted tool subscriptions and a freelancer. Ignored my own salary, the content writer's time, and the designer who made the infographics. My ROI looked amazing — 800%. Then my boss asked why the number didn't match the P&L. Embarrassing. I now track every hour I spend on SEO, including the technical audit that took three days. That alone changed the ROI from positive to barely break-even. It hurt, but it was honest.

I also treated all organic traffic as SEO-driven. If a page ranked for 'best coffee machines' and got 10,000 visits, I assumed SEO was responsible for the full 10,000. But half of those visits came from branded queries that would have happened anyway, or from direct traffic that GA misattributed. I now isolate non-branded organic traffic and use Search Console to filter out clicks from impressions that aren't from actual search results. The difference is often 30-40%. I wrote about this more in my [SEO analysis](/seo-analysis/) notes.

Another mistake: I used the same ROI formula for every type of SEO work. A technical improvement that fixes crawl errors has a different payback period than a content campaign. Combining them into one number hides which tactic is working. I now calculate ROI separately for technical SEO, content, and link building. That granularity helped me shift budget from link building (negative ROI) to content (strong positive). Finally, I assumed SEO ROI should be measured monthly. That's fine for mature sites, but for a new project, three months of data is noise. I now use a 12-month rolling average and report quarterly. The numbers are still ugly some quarters, but the trend tells the real story.