The Beginner's Guide to Search Console

Most SEO tools guess. They crawl your site the way they imagine Google might, estimate your rankings from sampled data, and charge you monthly for the privilege. Google Search Console is different: it's Google itself telling you what it crawled, what it indexed, which searches your pages appeared in, and what people clicked. It's free, it's first-party data, and it's the single most underused tool among small site owners.

The catch is that Search Console is built by engineers for an audience it never quite defined. It buries the three reports you need under a dozen you don't, and it phrases routine non-problems in language that sounds alarming. This guide covers setup, the reports worth your time, the warnings you can safely ignore, and a simple monthly routine that takes fifteen minutes.

Setting up: domain property vs. URL prefix

When you add a site to Search Console, you're asked to choose between two property types, and the choice matters more than the interface suggests.

  • Domain property covers every variant of your site: http and https, www and non-www, and all subdomains. You verify it with a DNS record.
  • URL-prefix property covers exactly one variant — https://www.example.com/ and nothing else. You can verify it with an HTML file, a meta tag, or your analytics account.

Choose the domain property if you can. It's the only way to see your whole site in one place, and it survives you later changing your www preference or adding a subdomain for a shop or blog. The only reason to pick URL prefix is if you genuinely cannot touch DNS — for example, you're a marketer without access to the domain registrar. In that case, add the URL-prefix property now and ask whoever controls DNS to add the TXT record when they get a chance.

Verifying on WordPress

For a domain property, verification is a single TXT record at your DNS host — Search Console shows you the exact value to paste. DNS changes can take a few hours to propagate, so if verification fails immediately, wait and retry rather than adding the record twice.

For URL-prefix verification on WordPress, the meta tag method is the least fragile: most SEO plugins have a field where you paste just the content value of Google's verification tag. Avoid the HTML file upload method on WordPress — theme changes and migrations tend to eat loose files in the web root.

One important note: verification does not affect rankings. Adding your site to Search Console doesn't make Google crawl it more or rank it higher. It just turns the lights on.

The Performance report: the one that pays rent

If you only ever open one report, make it Performance → Search results. It shows four metrics for your pages in Google Search over the last 16 months:

  • Clicks — someone searched, saw your page, and clicked it.
  • Impressions — your page appeared in results for a search (even at position 47 where nobody scrolled).
  • CTR — clicks divided by impressions.
  • Average position — your typical ranking for the queries where you appeared.

The default view is a chart of clicks over time, which is pleasant but not actionable. The value is in the tabs underneath: Queries, Pages, Countries, Devices. Here are three specific things to do with them.

Find your almost-ranking queries

Filter to the last three months, open the Queries tab, and sort by impressions. Now look for queries with lots of impressions, few clicks, and an average position between roughly 8 and 20. These are searches where Google already thinks you're relevant but hasn't committed. As a rule of thumb, moving a page from position 12 to position 5 often triples or quadruples its clicks, because almost all clicks go to the first handful of results. Improving an existing page that ranks eleventh is nearly always cheaper than writing a new one from scratch.

What "improving" means in practice: cover the subtopics competitors on page one cover and you don't, update stale sections, add the query's exact phrasing to a heading if it fits naturally, and add internal links to that page from your other relevant pages.

Spot title problems with CTR

Open the Pages tab and compare CTR for pages at similar positions. If two pages both average position 6 but one gets a healthy CTR and the other a far lower one, the second one has a title or snippet problem. Common culprits: a title that's pure keyword-stuffing, a title that doesn't match what the searcher wanted, or a missing meta description that lets Google pull a random sentence. Rewrite the title to promise the specific thing the query asks for, then check back in three or four weeks.

Watch for query drift

Every quarter or so, look at which queries actually drive traffic to a page versus what you wrote it for. Pages drift: an article you wrote about invoicing software ends up ranking for "invoice template free." That's information. Either lean into the drift (add a template) or accept the mismatch, but know that visitors arriving on that query will bounce if the page doesn't deliver it.

Indexing: what "not indexed" actually means

The Pages report (under Indexing) is where Search Console causes the most unnecessary panic. It lists your indexed pages and then a table of "Why pages aren't indexed," full of phrases like "Crawled — currently not indexed" and "Duplicate without user-selected canonical." Most of these are fine. Some are not. Here's how to triage.

StatusUsually meansAction
Excluded by 'noindex' tagIntentional — or a plugin misfireCheck the listed URLs. If a real page is here, find what set the tag.
Alternate page with proper canonical tagWorking as intendedNone.
Page with redirectWorking as intendedNone.
Crawled — currently not indexedGoogle saw it and wasn't impressedImprove or consolidate thin pages; ignore for tag/archive URLs.
Discovered — currently not indexedGoogle knows the URL, hasn't crawled it yetUsually patience. If persistent site-wide, an internal linking or quality signal problem.
Not found (404)A link points to a dead URLFix or redirect if the URL ever had traffic or backlinks; otherwise fine.
Duplicate without user-selected canonicalGoogle picked a different URL as the "real" oneCheck its choice is sensible; add canonical tags if not.

The pattern to actually worry about is a sudden change: a hundred pages dropping out of "Indexed" in a week, or a real content page appearing under noindex. Gradual accumulation of excluded tag archives, paginated URLs, and old redirects is the normal sediment of a living site.

A useful sanity number: on a typical WordPress site, having a large share of known URLs excluded from the index is unremarkable, because WordPress generates lots of low-value URL variants (tags, authors, feeds, paginated archives). Indexed pages should roughly match the number of pages and posts you'd actually want a stranger to land on.

Sitemaps and the URL Inspection tool

Submit your XML sitemap once, under Indexing → Sitemaps. Any WordPress SEO plugin generates one — the URL is usually /sitemap.xml or /sitemap_index.xml. This is a one-time task; the plugin keeps the sitemap updated, and Google re-fetches it on its own schedule. Submitting a sitemap does not force indexing; it just tells Google which URLs you consider canonical and when they changed.

The URL Inspection tool (the search bar at the top of Search Console) answers a different question: what does Google know about this specific URL? It shows whether the page is indexed, when it was last crawled, which canonical Google chose, and whether it's mobile-friendly. Two practical uses:

  1. After publishing something time-sensitive, inspect the URL and click "Request indexing." This puts the page in a priority crawl queue. It usually shortens the wait from days to hours, though there are no guarantees and hammering the button doesn't help — one request is one request.
  2. When a page has mysteriously vanished from search, inspection tells you whether it's a technical problem (noindex, blocked by robots.txt, redirect) or a ranking problem (indexed fine, just not ranking). These have completely different fixes, and guessing wrong wastes weeks.

The reports you can mostly ignore

Being honest about the rest of the sidebar saves beginners a lot of anxiety:

  • Core Web Vitals — worth a glance quarterly. If it's all green or all "not enough data," move on. If it's red, that's a performance project, not a Search Console project.
  • HTTPS report — should show essentially all pages on HTTPS. On any modern host this takes care of itself.
  • Links report — a rough sample of who links to you. Interesting, rarely actionable, and much less complete than it looks. Do not use it to "disavow" anything; disavowing links is a tool for penalty recovery, not hygiene, and misusing it can hurt.
  • Enhancement reports (breadcrumbs, FAQ, etc.) — these appear only if you use structured data. Errors here mean a rich result might not show; your ordinary ranking is unaffected.
  • Security & Manual Actions — check that both say you're clean, then never think about them unless Google emails you. A manual action email is the one Search Console message that genuinely means drop everything.

A 15-minute monthly routine

Search Console rewards a boring cadence far more than heroic occasional deep-dives. Once a month:

  1. Performance, last 3 months vs. previous 3 months (2 min). Look at the comparison. Clicks drifting down over two consecutive checks is a trend worth investigating; a single noisy week is not.
  2. Queries sorted by impressions (5 min). Pick one position-8-to-20 query and write down one concrete improvement to its page. One. You'll actually do one.
  3. Indexing → Pages (3 min). Scan for step changes in the indexed count or new noindex entries on real pages.
  4. Inspect anything new (2 min). If you published important pages this month, confirm they're indexed.
  5. Manual actions and security (30 sec). Confirm both are clean.

That's it. Set a recurring calendar reminder, because the failure mode isn't misreading Search Console — it's forgetting it exists until traffic has already been sliding for four months and the 16-month data window has swallowed your baseline.

Common beginner mistakes

  • Treating average position as a ranking tracker. It's averaged across every query and every searcher's location and personalization. A page can "rank #4" for its main query while showing average position 23 because it also gets impressions for a hundred fringe queries. Filter to a specific query before reading position.
  • Requesting indexing as a fix for ranking. If a page is already indexed, requesting indexing does nothing. Indexed-but-invisible is a content and links problem.
  • Panicking over excluded pages. Re-read the table above. Exclusion categories describing intentional behavior (redirects, canonicals) are Google confirming your site works.
  • Only looking at clicks. Impressions are the leading indicator. A new page earning impressions at position 30 is a page Google is testing; that's progress you'll miss if you only watch clicks.
  • Letting the property lapse. If you migrate domains or switch from www to non-www, add the new property immediately and use the change-of-address tool. Historical data doesn't transfer, so screenshot or export anything you'll want later.

The honest summary

Google Search Console won't optimize anything for you. What it does is replace guessing with knowing: which queries you're close on, whether Google can see your pages at all, and whether a traffic drop is technical or competitive. For a small site, the queries-by-impressions report alone — checked monthly, acted on one page at a time — beats most paid SEO tooling, because it's the only dataset that comes from the referee rather than a spectator. Set up the domain property this week, submit the sitemap, put the monthly reminder in your calendar, and let the data accumulate. Future-you, staring at a traffic dip, will be very glad the lights were already on.

Photo: Jan van der Wolf on Pexels