Your search traffic doubled. Your enquiries did not. Was the SEO work a success?
Maybe you reached more potential customers who are not ready to buy yet. Maybe you reached the wrong people. Maybe the contact form stopped working. A traffic chart cannot tell those stories apart.
To measure SEO results, connect three questions: can the right people find you, do they visit, and do those visits lead to something useful? Here is a small review you can actually maintain.
1. Choose one result that matters to the business
Before opening a report, finish this sentence:
SEO should help more suitable people ______.
For a software business, the answer might be “complete a trial sign-up.” For a consultant, it might be “send an enquiry about a service we offer.” For a shop, it might be “place an order.”
Choose the completed action, not just the button before it. A click on “Contact us” is a useful clue, but it is not a submitted enquiry. A trial sign-up is not a paying customer either; keep those stages separate.
Then name a page that should help that action happen. Start with a product or service page, or a guide that leads naturally to it. You do not need a measurement plan for every article on day one.
2. Check that the action is being recorded
Google Analytics 4, usually shortened to GA4, records interactions as events. You can mark an important event as a key event so it appears in the relevant reports.
If your website does not already record the action, ask whoever maintains it to set up an event for the successful completion. Avoid using a button click as a substitute for success unless that is genuinely the outcome you want to measure.
Test the setup yourself:
- Complete the action with an approved test account or test submission.
- Confirm the expected event appears in GA4's Realtime or DebugView report.
- Try an unsuccessful attempt and check that it does not record a success.
- Refresh or revisit the confirmation screen and check for accidental repeat events.
- Note when tracking started and how test activity is handled.
Google explains event setup and checking key events in its documentation. DebugView requires debug mode; it is not a second report where all ordinary visits automatically appear.
Keep names, email addresses, and form answers out of analytics event data. Google's privacy guidance for Analytics covers information that must not be sent.
If you only begin tracking today, say so in your baseline. Last month's missing data is not evidence of zero enquiries.
3. Use Search Console to check who can find the page
Google Search Console tells you about your appearance in Google Search. It does not track what someone does after arriving on your site.
Filter the search Performance report to the page you chose. Compare similar periods and read the queries, not just the total at the top.
For example, a booking app might receive more impressions for “free printable calendar.” That could increase visibility without reaching anyone looking for online appointment booking.
Record the questions you actually want to answer. Look for whether the page is appearing for those questions and receiving useful clicks. An impression is a reported appearance in Search; a click is a visit initiated from the result. Neither is a sale.
Average position can help you investigate a change, but it is an average across the included searches, not one permanent rank. A changing mix of queries can change that average. Our guide to using Search Console for blog ideas explains how to read the queries without treating every phrase as a new article opportunity.
4. Use GA4 to check visits and completed actions
In GA4, open the Traffic acquisition report. Use a session-based source or channel dimension so you are looking at visits, including visits from returning people.
For unpaid search traffic, inspect Organic Search with the Session default channel group dimension. That includes supported search engines, not only Google. To focus on Google, use Session source / medium and inspect google / organic.
Review sessions alongside the particular key event you chose. If you use Session key event rate, select the relevant event rather than mixing sign-ups with every other marked action. This rate describes the share of sessions containing the selected key event; the event count itself is not a count of unique customers. Google's Traffic acquisition report guide explains these dimensions and metrics.
Keep Search Console clicks and GA4 sessions in separate columns. They come from different reporting systems and should not be forced to match. Consent choices, tracking availability, and what each system counts can affect the comparison. A gap is something to investigate, not automatically missing traffic.
Also distinguish reporting credit from proof. Someone can discover you through a guide, return later, and then sign up. One acquisition report does not describe every interaction in that person's decision.
5. Put the numbers in a small monthly review
Use a consistent date range, the same filters, and the same definition of success. Compare equal-length periods with a similar mix of weekdays; for a seasonal business, also consider the corresponding season last year.
Here is a made-up example for a software website, not an SEO Dispatcher customer result:
| Measure | Earlier period | Later period |
|---|---|---|
| Google Search clicks to the site | 250 | 480 |
| GA4 organic-search sessions | 200 | 400 |
| Organic sessions containing a completed trial sign-up | 8 | 9 |
| Share of organic sessions containing that sign-up | 4% | 2.25% |
Visits increased, but the number of sessions containing a sign-up barely changed. That is a reason to investigate the new audience, landing pages, offer, and form—not proof that any one of them caused the result.
Add a short note next to the numbers: what changed, when it went live, whether tracking changed, and whether anything unusual happened. A promotion, a broken form, or a new consent setting can matter more than a headline rewrite.
Check lead quality separately. Nine enquiries that your business cannot serve are not equivalent to nine suitable enquiries. Your sales records can tell you things an analytics event cannot.
6. Use the result to choose the next job
Different patterns call for different investigations:
- Little relevant visibility: check indexing, the page's purpose, and whether it answers a real customer question.
- Relevant impressions but few clicks: investigate the queries, positions, title, and competing results together.
- Visits but few completed actions: test the form, clarify the offer, and check whether visitors can see a sensible next step.
- More completed actions but weak customer fit: review the promises on the page and the audience it attracts.
- Very little data: keep the raw counts, extend the observation period, and avoid treating a change from one to two enquiries as a reliable trend.
These are starting points, not automatic explanations. Change one identifiable problem and keep a record of the work.
Where SEO Dispatcher fits into this review
In SEO Dispatcher, connect your website's Search Console property and GA4 property. Search Insights brings in the available search reports; Analytics shows traffic, sources, and page activity from GA4.
Use those reports to choose a page, then run an SEO audit to investigate specific issues with that page. If a useful content change is needed, prepare a draft, review it, and send the approved Markdown or MDX file through GitHub.
SEO Dispatcher does not install your sign-up or enquiry tracking. Its current Analytics views focus on traffic reports, not a complete key-event or revenue dashboard. Use GA4 directly for the configured key event and your business records for qualified enquiries and paying customers.
That boundary matters. The app helps you connect evidence to a change; your measurement setup tells you whether the change helped the business.
Your next monthly review can fit in one sentence: “We changed this page for this reason, saw this result, and will investigate this next.” If you can write that honestly, the numbers are doing their job.