How to Measure SEO with Search Console and Analytics

The short version #

Treat SEO as a data review, not a feeling. Search Console covers the search side: impressions, clicks, CTR, queries, pages, indexing. GA4 covers what happens after the click: sources, behavior, events, forms, WhatsApp taps. The two never reconcile if nobody wrote down a baseline on launch day. Don't fixate on rankings in the first quarter. Week-to-week ranking moves are mostly noise. Watch impression trends, query breadth, target queries appearing for the first time, and whether your service pages are actually indexed. Run a monthly review with four sections: what grew, what broke, what content next month, what technical work next month. The template at the bottom is the one we hand to clients.

A client asked us six months after launch: "is our SEO actually working?" Marketing pulled a GA4 chart showing organic search up 45% and the CEO was happy. Three months later, inquiries were below target and the SEO budget got cut. When we dug back through the data, 70% of that 45% was branded search. Buyers who already knew the company name were typing it into Google. The non-branded queries that bring in new inquiries had barely moved.

The data wasn't lying. Nobody knew which numbers to read. This article isn't a Search Console or GA4 feature tour. The official docs cover that. It's the playbook for what to look at every month after an overseas site goes live: which metrics matter, what counts as normal noise, and what's a real problem.

Two tools, two jobs #

A common mistake is treating Search Console and GA4 as interchangeable. They watch different halves of the same funnel.

Search Console covers the search side — what happens before the click:

• Impressions: how many times your pages appeared in Google's results. The first sign that SEO is working is a steady rise in impressions, not clicks.
• Clicks: how many of those impressions turned into a visit.
• CTR: clicks divided by impressions. Low CTR usually means the title or description isn't compelling, not that the ranking is bad.
• Average position: directional only. Don't read it standalone.
• Queries: the actual phrases people typed. This is the most valuable section of any monthly review — more accurate than any keyword tool because it's real search behavior.
• Pages: which URLs are pulling impressions and clicks.
• Index coverage: of the URLs in your sitemap, how many got indexed, how many got excluded, and why.

GA4 covers what happens after the click — once a visitor lands on the site:

• Traffic source: organic search, direct, referral, social, paid, etc.
• Landing pages: where visitors first arrive.
• Engagement (rate and average time): GA4's replacement for bounce rate. Tells you whether content actually gets read.
• Events: form submissions, WhatsApp taps, email clicks, PDF downloads, video plays. These are what tie traffic to pipeline.
• Key Events: in GA4 you have to manually mark form submissions or WhatsApp taps as Key Events. Skip this and the rest of the funnel is invisible.

In one line: Search Console tells you what people searched and whether they saw you. GA4 tells you what they did once they arrived. Most reporting disagreements come from looking at only one side.

Launch-day baseline #

This step is the foundation for every review afterward. Miss it and you wait three months for the next chance.

In launch week:

• GA4: confirm page_view is firing on every page. Walk through the main pages in DebugView and watch the events fire. Mark form submissions, WhatsApp taps, email clicks, and phone clicks as Key Events. Add IP filters for your team, your agency, and your office. Without that, the first two weeks of "data" is just internal traffic.
• Search Console: verify the property at the domain level via DNS TXT or GA4 link. Submit sitemap.xml. Open the Page Indexing report and note the reasons URLs are excluded (redirect, alternate canonical, blocked by robots).
• Bing Webmaster Tools: Bing's share is higher than China-side teams expect, especially for B2B, older Western buyers, and corporate intranet search. Submit the sitemap there too.
• Baseline sheet: a one-page Notion doc or Google Sheet with launch-day impressions (should be near zero), indexed pages, and the prior week's GA4 sessions and users. You add one row to this sheet every month from now on.

If you inherit an already-launched site with no baseline, take this month's numbers as the baseline. Don't try to retroactively claim improvement against a number that was never recorded.

Don't lead with rankings #

The first question on any first-time SEO review is "where are we ranking?" In the first three months, that question barely matters. Three reasons:

1. Rankings drift. Google personalizes by location, login state, device, and recent searches. You see position 8 in the morning, your colleague sees position 14 in the afternoon. The site didn't change. The SERP did.
2. New sites start unranked. Google holds new pages back while it figures out trust. Volatile rankings in months one through three are normal and not a signal to act on.
3. Rankings don't equal inquiries. A factory-tour blog post at position 3 may not generate a single inquiry all year. A service page at position 18 may bring in two or three qualified leads every month.

What to actually watch in the first quarter:

• Impression trend: rising steadily? Which pages are pulling impressions?
• Query breadth: how many distinct queries are surfacing your site? Breadth matters more than any single position.
• Target queries: are non-branded queries that match your business starting to appear? Showing up at position 30 for a real buyer query is a much better signal than position 1 for your own brand name.
• Index coverage: are your service pages and case studies actually indexed?

Rank trackers (Ahrefs, Semrush, Mangools) earn their keep after month three, and even then they supplement Search Console rather than replace it.

Monthly review structure #

Here's the structure we use with clients. Thirty to sixty minutes of work, two pages of output. Four sections.

Section 1 — What grew

Pick the two or three pages or queries with the most obvious gains in impressions, clicks, or engagement. Write the absolute numbers, not just percentages. Write down why it grew: new content, last month's internal linking, an external mention, a refreshed title.

Section 2 — What broke

Pick two or three things that look wrong. Examples:

• A core service page lost clicks two weeks in a row.
• A new batch of 404s appeared in Search Console.
• Bounce rate from one country jumped to 80%.
• Form submissions hit zero while WhatsApp taps stayed normal. Usually a broken form.

Write a one-line "next action" under each problem. Don't list problems without owners.

Section 3 — Content next month

Use this month's query data to pick next month's content. If "WhatsApp Business API price" appeared in queries but you don't have a page on it, that's a blog post. The point is queries drive content choices, not marketing brainstorms.

Section 4 — Technical work next month

Run a small backlog: image compression, missing schema, internal links, speed fixes, indexing exclusions to investigate. Two or three items per month is plenty. Don't list twenty and finish none.

Send the review to the business owner and the CEO. The clients who get the most out of SEO put this monthly review on the executive team's monthly meeting agenda. Business numbers and SEO numbers in the same room.

Tying traffic to pipeline #

Traffic alone is vanity. Inquiries are the goal. This section connects to UTM Tracking for WhatsApp, X, Forms, and Email Leads.

In GA4, confirm:

• Key Events configured: form submissions, WhatsApp taps, email clicks, phone clicks, key PDF downloads.
• Source attribution: under "Acquisition → Traffic acquisition," you can see how many Key Events each source produced.
• UTM consistency: every external link (email signature, social, ads) uses the same UTM naming convention. Once naming gets messy, retroactive cleanup is brutal.
• CRM reconciliation: if you use HubSpot, Pipedrive, or a custom system, run a monthly check of "GA4 form submissions vs CRM new leads." A gap larger than 20% means an event isn't firing or replies are getting flagged as spam.

Skip this section and the entire SEO program collapses to "we looked at a traffic chart."

Must, worth, skip #

Sort monthly work by priority or you'll burn time on busywork.

Must do every month:

• Read the Search Console queries and pages reports.
• Track index coverage; investigate new 404s and excluded URLs.
• Reconcile GA4 Key Events against CRM leads.
• Refresh one or two old articles or service pages each quarter (title, intro, internal links).

Worth doing once you have headroom:

• Connect Looker Studio so Search Console and GA4 sit on one dashboard.
• Pull Search Console data via API for long-term query trend analysis.
• Track branded search volume as a brand-health signal.

Skip for now:

• Running four rank trackers in parallel.
• Building complex attribution models in month one. Last-click done correctly is plenty for the first half year.
• A/B testing pages before you have enough traffic for statistical significance.

For the broader maintenance cadence after launch, see Post-Launch Website Maintenance Checklist, and for the first month specifically, see The First 30 Days After Launching an Overseas Website. For the AI-search side of visibility, which Search Console doesn't fully cover, see How to Monitor Brand Visibility in AI Search.

Monthly review template #

Copy this table into your next monthly review.

Fill one row per month for three consecutive months and your read on the site will outperform any tool on the market.

References: the Google Search Central SEO Starter Guide, Search Console's Page Indexing report, and GA4 documentation on Key Events.

FAQ #

Why don't GA4 organic numbers match Search Console? #

Different counting methods. A 10–30% gap is normal. GA4 counts sessions that actually loaded (ad blockers can drop them); Search Console counts clicks from the SERP (whether the page finished loading or not). If the gap exceeds 50%, check tracking deployment, cross-domain configuration, and whether a redirect is stripping the referrer.

How long until SEO produces real traffic? #

Non-branded organic traffic typically takes three to six months to stabilize, assuming content, internal linking, and technical SEO are all being worked. If the line is still flat at month three, audit content depth and check for technical blocks (noindex, robots disallow, broken canonicals) before assuming you just need more time.

Is Bing Webmaster Tools worth the time? #

Yes, but after Google. In some North American and European verticals (finance, enterprise IT, government procurement), Bing reaches 10–15% share. If those are your buyers, the extra ten percent of search traffic is worth an hour of setup.

Do we need a paid rank tracker? #

Not in year one. Search Console queries are more accurate and free. In year two, an Ahrefs or Semrush subscription is mostly useful for competitor research and backlinks, not for tracking your own rankings.

Get a diagnosis #

If your overseas site is live but the analytics aren't telling a clear story, give us read access to your Search Console and GA4 properties. We'll run this exact monthly review template against your data as a free SEO/GEO measurement diagnosis and tell you which metrics are quietly improving, which numbers point to real problems, and what to fix first.