Google Search Console is the one place where Google tells a site owner, in its own words, what it knows about the site. It is free, it takes an afternoon to set up, and I have done the setup myself. Most of what follows comes from Google’s own help pages, which I have linked as I go, because the pages are specific and my memory of menu paths is not something worth trusting.

Intro to Google Search Console – Search Console Training, Google Search Central on YouTube

What the tool is for

Pencil line graph drawn on graph paper beside a ruler

Analytics counts the visitors; Search Console counts the doors Google found

A shop owner who has read about analytics tends to expect Search Console to be more of the same. It is a different instrument. Analytics describes what visitors did once they arrived. Search Console describes what happened before that, on Google’s side of the door: whether Google found the pages, whether it kept them, and how often the site showed up in results. If analytics is the till tape, Search Console is the delivery log. I wrote about the till-tape side in reading Google Analytics, and the two reports answer separate questions.

The reports are only worth opening once the site is proven to be yours. That is the first job.

Proving the site is yours

Six ways in, and only one covers every address

Google will not show a stranger the search data for a site, so Search Console asks for proof of ownership first. According to Google’s help page on verifying site ownership, you add a property, pick a verification method, and follow its instructions. A verified owner has the “highest degree of permissions” in the tool.

The methods on that page are these:

  • Uploading an HTML file Search Console gives you to the root directory of the site.
  • Placing a meta tag in the head of the home page.
  • Using an existing Google Analytics tracking snippet, if you can edit that Analytics account.
  • Using a Google Tag Manager container snippet, if you have publish or admin permission on it.
  • Signing in with the same account that manages the site, for Google Sites and Blogger.
  • Adding a TXT or CNAME record at the domain name provider.

The last one is the only method Google lists for a Domain property, which covers every version of the address in one go. That matters for a small shop that has a bare domain, a www version and perhaps a leftover address from an old host. A Domain property catches all of them. The other methods cover a single address.

I would pick the method by asking who controls what. A person who built the site on a hosting platform and never touched the domain settings will find the meta tag or the Analytics route easier. A person who bought the domain and still holds the login for the registrar will find the DNS record cleaner, because it does not live inside the site and cannot be deleted by a redesign. Google’s page makes the same point about the HTML file: remove it and the verification goes with it. A redesign that clears out the root directory would take the file with it.

The page also says you can add several methods at once in case one fails. That is cheap insurance.

Telling Google where the pages are

Blank index cards pinned in a tree layout on a cork board

A sitemap is a list, and a list can be wrong

Once the property is verified, the next job is a sitemap. Google defines it as “a file on your site that tells Google which pages on your site we should know about.” The same page admits that small sites with well-linked pages may not need one, and that many hosting services produce one automatically. So the first step is to find out whether the site already has one rather than build one.

Submitting it takes five moves according to that help page. Check that you have owner permissions on the property, post the sitemap on the site, confirm Google can reach it with a live URL inspection, paste its address into the “Add a new sitemap” box in the Sitemaps report, and then check the status. If the status says anything other than “Success,” clicking the row shows the error details.

A sitemap is a list, and a list can be wrong. If it names pages that no longer exist, or leaves out the pages that bring in orders, Search Console will report the mismatch faithfully and nobody will be happier for it. The cheap check is to open the sitemap file in a browser and read it like an inventory sheet: does every page the shop depends on appear, and does anything appear that should not?

Clicks, impressions and the Performance report

Being shown and being chosen are two different events

The report people open first is Performance, and its columns are easy to misread. Google’s definitions are short. Clicks are “the number of times a user clicked your site from Google Search results.” Impressions are “how many times your site appeared in Search results.” CTR is “the click count divided by the impression count.” Average position is “the average position of the topmost result from your site.”

The distinction that matters is between the first two. An impression is the site showing up on a results page. It does not mean anyone saw it, read it, or cared. A page can pile up impressions for months and collect no clicks, and that tells you something specific: the page is findable and the listing is not persuasive. A page with few impressions and a good click rate tells you the opposite. The two problems need different repairs, and the Performance report is how you tell which one you have.

The report lets you slice the numbers. Google lists the filters: date range, search type (web, image, video or news), and dimensions that group the table by queries, pages, countries, devices, search appearance or dates. The queries view is the one to sit with. It lists the actual phrases people typed before Google offered the site. For a shop selling something handmade, those phrases are rarely the ones the owner would have guessed, and they are usually plainer. People type what the thing is, not what it is called on the label.

The average position figure needs one caution that Google builds into its own definition. In the chart it covers the whole site. In the table it reflects the specific URL or grouping in each row. A site-wide average of 14 can hide one page sitting at 3 and another at 40. Read the table, not the headline.

The tool has a limit, and I would be lying to leave it out: a small site will not produce enough data to settle arguments. A page with eleven impressions this month has no trend to speak of, and a query needs a few weeks behind it before a movement means anything.

Whether a page made it into the index

Magnifying glass over a stack of paper pages on a desk

Not indexed is not always a fault

The second report worth opening is Page indexing. Google describes it as showing how many of the site’s URLs it has crawled and indexed, and why other URLs were not indexed. The help text says sites under 500 pages probably do not need it, which is worth hearing for a shop with twenty pages. For a small site the tool to reach for is the single-page one.

The URL Inspection tool, documented here, gives a verdict of “URL is on Google” or “URL is not on Google,” and warns that the first does not guarantee the page appears in results. Its “Test live URL” button fetches the page in real time, which is how you find out whether Google can crawl and read a page you changed this morning. A passing live test does not ensure indexing, because the test does not check sitemaps, referring pages, canonical selection or manual actions.

There is also a “Request indexing” button. Google says plainly that it asks for a crawl and guarantees nothing, that there is a daily limit, and that for many new or updated pages a submitted sitemap is the better route. I read that as permission to stop pressing the button.

Not every page that is out of the index is a problem. Google says that “not indexed” is “not necessarily bad,” and the reasons table can list legitimate causes such as a duplicate or a robots.txt block. The report also tells you whether the source of the problem is Google or the website, and generally only the website ones are yours to fix. The goal Google states is simple: the important canonical pages should be indexed, and duplicate or alternate pages generally should not be.

A routine that fits a small shop

One index card, once a month

Nothing in Search Console needs daily attention, and checking it daily mostly produces anxiety. A monthly pass is enough for most small sites, and it can fit on one index card:

  • Open Performance and look at the queries for the last three months, not the last three days.
  • Note any page with plenty of impressions and few clicks, and reread its title and description.
  • Open Page indexing and look for the website-sourced reasons on pages that matter to the business.
  • Inspect any page that was rewritten or added since the last pass.
  • Check that the sitemap status still says “Success.”

The titles are the part within the owner’s reach. Google’s guidance on title links asks for titles that are descriptive, unique and concise, with the most important words first. A vague title is one plausible reason a page has impressions and no clicks, and rewriting it costs nothing but ten minutes at a desk.

Verification, a sitemap, one report for what people search and one for what Google kept: that is the whole working kit. The rest of the tool is there for sites with thousands of pages and staff to read the results.

For informational purposes only. Menus and report names inside Search Console change over time, so check Google’s current help pages before following any step.

Sources