Content Writing
How to Use Google Search Console to Improve Blog Content
Five Search Console reports that turn directly into edits: near-miss queries, titles that do not earn clicks, off-topic queries, posts competing with each other, and decay.
Use Search Console to find what each post is already close to ranking for, then edit the post to answer those searches better. In practice that means five checks: queries where you sit on page two, queries with impressions but few clicks, queries that show Google misreads the page, posts competing with each other, and posts in decline. Each one turns into a specific edit.
Everything below happens in the Performance report (Search results), with the search type left on Web. A few facts about the data before you start, because they change how you read it:
- Average position is the topmost position your page reached for each query, averaged across queries (Google's definition). Always read it per query, never per site.
- The table shows 1,000 rows at most, and very rare queries are withheld for privacy, so totals by query will not match totals by page. Both points are explained in Google's deep dive on performance data.
- You get about 16 months of history, which Google confirms in its Analytics help on Search Console data. Export anything you want to keep longer.
- AI Overviews and AI Mode appearances are counted inside the Web search type (Google Search Central), not reported separately.
Start With One Post at a Time
Click + Add filter, choose Page, and paste the post URL. Set the date range to the last three months, then open the Queries tab and switch on all four metrics: clicks, impressions, CTR and position. This list is the most honest description you will ever get of what Google thinks the post is about.
Read it before you look at anything else. If the post is titled "How to choose a CRM for a small law firm" and most impressions come from "what is a CRM", you have learned something no keyword tool would have told you.
Check 1: Near Misses on Page Two
Sort by impressions and look for relevant queries with an average position between roughly 8 and 20. These are searches where Google already considers the post a candidate but ranks something else above it.
The edit: search the query yourself and see what the top results cover that yours does not. Usually it is a missing subsection, a missing definition, or a question the post skirts. Add a section with a heading that uses the reader's phrasing, answer it in the first two sentences, and link to it from a related post. Do not paste the query into the text five times; that is the opposite of what works (see how to avoid keyword stuffing).
Check 2: Impressions Without Clicks
Still on the Queries tab, look for queries with high impressions and a CTR well below your other queries at a similar position. Position matters: a low CTR at position 9 is expected, while the same CTR at position 2 means searchers are seeing your result and choosing another.
The edit: rewrite the title tag and meta description so they match the query's wording and promise the specific answer. The method is in how to write meta titles and descriptions for blog posts. A quick example:
| Title tag | |
|---|---|
| Before | Blog Content Tips | Company Name |
| After | How to Use Search Console to Fix Blog Posts (5 Checks) |
Google does not always use your title, so check the live result a week later. If it rewrote it, the rewrite usually borrows from your H1 or first heading, which tells you where to make the change.
Check 3: Queries That Show a Topic Mismatch
If a large share of impressions come from queries the post was never meant for, the post is sending a mixed signal. Two common causes: the introduction talks around the topic for several paragraphs before stating it, or one long section on a side issue is clearer than the main argument.
The edit: make the opening state the topic and answer it, tighten or cut the side section, and check that the H2s describe the intended subject. If the off-topic query is valuable in its own right, that is a separate post, not a section bolted onto this one.
Check 4: Posts Competing for the Same Query
Remove the page filter, add a Query filter for an important search term, and open the Pages tab. If two or more of your posts collect impressions for it, and positions swap between them over time, they are splitting the signal.
The edit: decide which one should own the query, then either differentiate the other (give it a clearly different angle and title) or combine them. The decision rules and the redirect mechanics are in should you update, merge, or delete an old blog post.
Check 5: Decay
Open the date filter, choose Compare, and compare the last three months with the same three months last year. On the Pages tab, sort by the difference in clicks. Posts with large drops are your refresh queue.
Click into each and check the Queries tab with the same comparison. If impressions held but clicks fell, the result page probably changed (new competitors, AI features, a different title shown). If impressions fell too, the post lost rankings and needs a substantial update. The full process is in how to refresh outdated blog content.
A Regex Filter Worth Saving
Search Console accepts regular expressions in the query and page filters via Custom (regex), a feature Google added to the Performance report in 2021. This one lists every question people asked that brought your blog an impression:
^(who|what|why|how|when|where|which|can|does|do|is|are|should)\b
Apply it as a Query filter with the page filter set to your /blog/ folder. Questions with impressions and no matching heading on the page are ready-made subsections, and questions no post answers are topic ideas.
A Worked Example
Here is what the process looks like on a single hypothetical post, "How often should you post on a business blog", over a three-month window:
| Finding | Data | Edit made |
|---|---|---|
| Near miss | "how many blog posts per month for seo", position 14, 1,900 impressions | Added an H2 answering it directly with a monthly range by business size |
| Low CTR | Main query at position 4, CTR well below other position-4 queries | Rewrote title tag to include the answer range |
| Mismatch | Impressions for "blog posting schedule template" | Linked to the existing content calendar post rather than adding a template |
| Competition | A second post on blog frequency also gets impressions for the main query | Retitled the second post around a different angle and cross-linked |
Four edits, one afternoon, every one justified by data. Record the date and leave it alone for at least four weeks before comparing.
Do not react to a single week. Positions on low-volume queries move several places for no reason you can act on. Compare periods of at least 28 days, and judge an edit on the queries it targeted rather than on the whole page.
Turning This Into a Monthly Routine
- Week 1: run checks 1 and 2 across your ten highest-impression posts. Make the edits.
- Week 2: run check 5 (decay) and add the losers to the refresh list.
- Week 3: run check 4 on your five most important queries.
- Week 4: run the regex filter, feed new questions into the content plan, and log every edit with its date.
If the routine keeps slipping behind everything else, you can hand the blog over: our blog management service is run by an in-house team, with no lock-in on standard plans.
Questions
Questions about this article
Still unsure about something? Talk to a specialist — no script, no sales pitch.
Give it at least four weeks, and compare the same length of period before and after the change date. Google has to recrawl the page, and rankings settle over several weeks. Use the compare option in the date filter, keep the page filter on, and note the edit date somewhere you will find it, because Search Console does not record your changes for you.
They measure different things. Search Console counts clicks on Google results, before the visitor reaches your site. Analytics counts sessions that loaded its tag, which excludes people who block cookies or leave before the page loads, and it attributes traffic by its own rules. Expect Search Console clicks and Analytics organic sessions to differ and use each for what it measures.
It depends almost entirely on position and on what else appears on the results page, so a single benchmark is misleading. Compare like with like: sort your own queries into position bands (1 to 3, 4 to 7, 8 to 10) and look for pages well below the median of their band. Those are the titles and descriptions worth rewriting first.
Yes. Paste the URL into the URL Inspection tool and choose Request Indexing. It asks Google to recrawl the page sooner; it does not guarantee faster ranking changes. For routine edits it is optional, because Google recrawls pages that already get traffic, but it is worth doing after a substantial rewrite or a URL change.
Google withholds very rare queries to protect searcher privacy, so the query table does not add up to the page totals. The interface also shows at most 1,000 rows. For a small blog the anonymised queries are mostly one-off long-tail searches; you will still see the patterns that matter.