After publishing a website, the first useful question is not whether one large counter is rising. Ask where people encounter the site, whether they choose to open it and whether the page helps them. Those are different stages, and a sensible report keeps them separate.
Start with search visibility
Google Search Console measures appearances and clicks for the property you verify. Its performance reports can show impressions, clicks, click-through rate and information such as pages, queries, countries and dates. These describe the site's presence in the relevant Google service, rather than every visit from every source.
An impression means an appearance counted under the report's rules. A click records a qualifying click through to the site. Click-through rate is clicks divided by impressions. For a simple arithmetic example, two clicks from ten impressions produce a 20% CTR; those small numbers are not enough on their own to establish a reliable trend.
Do not turn the report into a headcount
Search appearances are not a count of unique people. Google also explains that anonymised queries are omitted when filtering by query, so a branded-search estimate is incomplete. A report of queries containing your site's name should not be described as the total number of people who searched for that name everywhere.
Keep the selected date range, country and search type with any figure you record. A change in the filter can change the result even when nothing about the website changed. Compare like with like before deciding that a headline or page needs replacing.
Keep on-site behaviour separate
A website analytics service can answer questions about visits and activity after arrival, depending on what it is configured to collect. Affiliate reporting answers another question again: whether qualifying purchases were attributed to a referral. An outbound product click is not a confirmed sale.
For a new site, choose a few decisions you want the data to support before installing a large collection of tracking tools. Review the data collection and privacy setup for the actual services selected. A private preview with indexing disabled is not a useful test of public search demand.
Make a short weekly review
- Record the same seven-day window and comparison period.
- Note the pages and queries bringing useful search clicks.
- Inspect pages with impressions but few clicks for relevance and clarity.
- Check whether a visitor can find the next useful guide or tool.
- Write down one change to test and when to review it.
Use the results to improve a specific page or topic, then give the change time to gather evidence. Publishing more material is useful when it answers real questions; a bigger article count alone does not guarantee traffic.
Sources & further reading
Checked 17 Nov 2026. Source-based reporting and guidance; individual test results are identified where available.
Tell us the article title, the detail and a supporting source.
Corrections & contact →