Skip to content

Light version: pages open quickly, without the 3D graphics. You are seeing the light version because the 3D graphics would be slow on this device. The content is the same.

Turn on 3D

Preparing the graphics. This can take a moment on the first visit.

Henry Lõiv

Web design & development

GuideSEO & AI Search 4 min read

How to use Search Console and GA4 with AI to choose website improvements

Use Search Console and GA4 through MCP to choose website improvements. A worked fictional example, priority worksheet and prompts for checking the results.

You have connected the Search Console MCP and Analytics MCP. What should you ask next? This walkthrough turns their separate reports into an evidence-backed shortlist of pages to improve. It does not promise that an assistant can reveal which Google query generated a lead.

Outcome: a spreadsheet with a baseline, an observed problem, a proposed change and a way to verify its effect. You need: read access to the correct GSC and GA4 properties, an MCP-capable client, and roughly 50 minutes. The worked numbers here are fictional, not this site's analytics.

1. Make the periods comparable

Choose two complete 28-day windows, excluding the most recent days if GSC data is not final. Record the date ranges, property, timezone, search type, device/country filters and whether branded searches are included. GA4's property timezone and GSC's reporting timezone can differ; compare trends, not row-by-row identities.

Ask your agent:

Using my read-only Search Console tools, show clicks, impressions, CTR and average position by page for the two specified periods. Include the exact periods, property and filters. Never invent missing rows. Separate complete data from fresh/incomplete data.

GSC may return top rows rather than every query or page; a missing row is not automatically zero. Query-level privacy and aggregation limits also matter.

2. Find candidate pages

Start with pages having meaningful impressions and declining clicks or CTR; also inspect pages with rising impressions but no corresponding clicks. Don't apply an arbitrary universal CTR threshold: position, device and search intent change the expected CTR. For each candidate, request query mix and device split.

Fictional example only:

PageEarlier impressionsRecent impressionsEarlier clicksRecent clicksFirst question
/guides/mcp/2,0002,6007052Did the position or query mix change?
/services/1,2001,1004438Is this a seasonal or technical shift?
/faq/payments/500900813Are new queries less relevant?

In the first row, clicks fell by 18 (about 26%) despite 30% more impressions. That is a reason to investigate, not proof the title is poor.

3. Check what visitors do in GA4

Ask for Google organic sessions and engagement for the relevant landing pages, plus configured key events. Define the metrics explicitly. eventCount counts event occurrences; it is not automatically the number of sessions with that event or the number of qualified leads. A contact-page view is not a form submission; an unconfigured event is unknown, not zero.

From my GA4 property, compare Google organic traffic on these landing pages over the same named periods. Show sessions, engaged sessions, and the configured contact-submit/key event. Clearly label absent instrumentation or unavailable data. Don't join individual search queries to individual GA4 conversions.

Normalize paths, redirects, UTM variants and language prefixes before comparing GSC page URLs with GA4 landing page dimensions. Expect discrepancies: the platforms use different collection and attribution mechanisms.

4. Recommend only testable changes

For each shortlisted page, write Observation → Hypothesis → Change → Verification. Examples: align the title and introduction to queries actually received; add a troubleshooting section answering a real repeated question; improve a relevant call-to-action; fix discoverability on mobile. Do not claim a change will raise rankings.

Use the priority worksheet to record effort, confidence and expected business relevance. A low-traffic service page may matter more commercially than a highly viewed general AI comparison.

5. Run a controlled review

Capture the baseline and publication date. After changes are live, monitor comparable periods, noting seasonal events and site changes. A small sample cannot establish causality. Keep the original report/CSV so someone can reproduce the decision.

Troubleshooting

  • Properties disagree: check property identity, landing-page path, country/device filters, timezone and GA4 tracking coverage.
  • No key events: verify instrumentation instead of reporting zero leads.
  • MCP tools missing: re-open the relevant setup guide and verify the client connection and permissions.
  • Sudden drop: check whether the latest GSC reporting days are incomplete.

Search Console MCP setup · GA4 MCP setup

Official references: Google on using both data sources, GSC Search Analytics API, GA4 API metric definitions.

Downloads and next steps

Download the fictional period-comparison CSV and priority worksheet. The CSV uses the same example numbers above; its paths do not refer to measured pages on this site.

Need help turning the findings into website changes? See website development.

Source references and fictional calculations checked on 12 November 2026. The sample is not a client report and does not identify individual leads.

Turning search visits into useful next steps?

Useful content needs a clear website behind it. I can help improve page structure, performance and the path from an article to your services, products or enquiry form.

See how I build and improve business websites⁠

Written by Henry Lõiv

Related guides

  1. Google Search Console MCP: free setup for Claude and Codex Connect Search Console to Claude or Codex on macOS and Windows. Free local MCP setup, read-only Google access, useful SEO prompts and fixes for common errors. GuideSEO & AI Search 12 min read
  2. Google Analytics MCP: Claude and Codex setup on Mac and Windows Connect GA4 to Claude or Codex on macOS and Windows with a free local MCP server. Read-only OAuth setup, useful prompts and fixes for common errors. GuideArtificial Intelligence 13 min read

Keep reading

All articles⁠
  1. Best AI agents in 2026: Grok Bot vs Dots and EU availability Compare AI agents by the work they can handle and whether you can use them in Estonia or the EU. Artificial Intelligence 5 min read
  2. AI Agent, Chatbot or Automation: What Does Your Business Need? A chatbot is an interface; an agent is one way to handle the work behind it. Follow a fictional customer enquiry to choose a practical system for your business. Artificial Intelligence 8 min read
  3. Claude Code vs Codex for Frontend Development: A Practical Comparison A documentation-based comparison with a runnable frontend exercise and a blank scorecard. No measured winner is claimed. Artificial Intelligence 9 min read

Hi!

I really love the sea, yachts, sunsets and starry skies.

Through this little 3D world, I can share a small part of myself with you.

Thank you for opening this bottle and reading my letter.

Henry

Let’s talk about your project

Tell me about your project

Tell me what you’d like to build or improve. You don’t need a finished brief — we’ll work out the scope and next steps together.

01 How can I help?
02 How many pages do you need?
03 Which extras do you need? Choose all that apply
04 When does it need to be ready?
05 Project details
06 Your contact details

Enquiry received

Thank you!

Your enquiry has been received. I usually reply by email within 2 working days.