Measuring SEO return without fantasy maths
SEO can be measured, but not with false precision. Here's how to track what it brings in and judge it fairly over time.
Why SEO return is tricky to measure
Paid ads have a clear cost per click and a trackable path to a sale. SEO is messier:
- Customers often visit several times before buying, through different channels.
- Phone calls, walk-ins and word of mouth don't show in analytics.
- Some visitors decline analytics cookies.
- Results build over months, not days.
- Some value, such as branded search and reputation, is hard to put a figure on.
That doesn't mean you can't measure it. It means you should be honest about the uncertainty and look at trends over quarters.
Step 1: Track enquiries and sales from search
For lead generation businesses
- Set up key events in GA4 for form submissions, phone taps and bookings. See our GA4 guide.
- Ask every new enquiry "How did you find us?" and log the answer.
- Record which enquiries become customers, and their value.
For online shops
- Make sure ecommerce tracking records purchases and revenue.
- Look at revenue by channel in GA4, including Organic Search.
Step 2: Value an enquiry
If you know roughly:
- What share of enquiries become customers, and
- What a typical customer is worth,
you can estimate the value of an enquiry. For example, if a quarter of enquiries become customers, each enquiry is worth roughly a quarter of a typical customer's profit.
Use profit, not revenue, where you can. A busy month of low-margin work isn't the same as a few high-margin jobs.
Step 3: Compare with what you spend
Add up what SEO costs: provider fees, tools and your own time. Then compare with the estimated value of enquiries or sales from organic search over the same period.
Do this over quarters, not months. SEO work in January may bring enquiries in April.
Step 4: Be honest about attribution
Analytics tools assign credit to channels using models. GA4 uses data-driven attribution by default for many reports, spreading credit across touchpoints. No model is perfect. Some things to keep in mind:
- Someone who found you through a Google guide, then came back by typing your address, may be counted as "Direct".
- Someone who saw your Business Profile and called without visiting your site won't appear in GA4 at all. Check your Business Profile's performance figures for calls.
- Branded searches often follow other marketing, so not all organic value comes from SEO work alone.
Use several sources together: analytics, Search Console, Business Profile figures and your own enquiry log.
Step 5: Watch leading indicators
Because revenue lags, track signs that SEO is heading the right way:
- Pages indexed.
- Impressions and clicks for non-branded searches.
- Average position for important queries.
- Organic landing page sessions.
- Business Profile actions: calls, website visits, direction requests.
If these improve, revenue usually follows, though not always in a straight line.
Keep a simple log
A spreadsheet is enough. Each month, record organic enquiries, how many became customers, their value and what you spent on SEO. Add a column for notes, such as a new page launched, a Google update or a busy season. After a year, you'll have a record that shows the real pattern far better than any single month's report.
Fantasy maths to avoid
- "Traffic value" from tools. Some tools estimate what your organic traffic would cost as ads. It's an interesting comparison, but not money you made.
- Ranking-based revenue forecasts. Projections that assume a position will produce a set click rate and conversion rate stack guesses on guesses.
- Taking credit for everything. If a TV appearance doubled branded searches, SEO didn't cause that.
- Ignoring costs. Your time counts, too.
A worked example
A small solicitor's practice in Gloucester wanted to know whether its SEO spend was worthwhile. It had GA4 installed but no key events.
It set up key events for contact form submissions and phone taps, added a "How did you find us?" question to its intake process, and recorded which enquiries became instructions and their fees. After two quarters, it compared the fees from clients who said they'd found the firm through Google with its SEO costs. It also checked Business Profile calls separately, as many clients called straight from the map.
The figures weren't exact, and the partners accepted that. But they showed a clear pattern: conveyancing enquiries from search had grown and converted well, while family law enquiries from search were few. The firm shifted the next quarter's SEO focus to its family law pages.
Questions
What's a good return on SEO?
It varies by business, margin and market. Compare SEO with your other ways of winning customers rather than with a universal figure.
How long before I can judge SEO return?
Give it at least six months for meaningful data, longer in competitive markets. Track leading indicators in the meantime.
Start with a free SEO audit
Your whole site, audited free. No card and no obligation.