How to Use Google Search Console Data to Decide What Content to Write Next
By Ghost Writr · · 8 min read
Most content teams guess. They brainstorm topics, check what competitors rank for, then write and hope. There’s a better way — the data is already in your account.
Google Search Console shows exactly what your site is discovered for. Queries, impressions, clicks, positions — a direct map of where demand exists and where your content falls short. Every one of those signals tells you something different about what to write, what to fix, and what to leave alone.
Start: Pull the Right Report
Open Search Console. Go to Performance → Search results. Set the date range to at least 90 days — shorter windows add noise and obscure slow-building trends.
Enable all four columns: Total clicks, Total impressions, Average CTR, Average position. Every decision here flows from these four numbers together, not any single one in isolation. A query with great position but zero impressions is different from one with great impressions but terrible CTR. Each has a different fix.
Export to a spreadsheet. Sorting and filtering in the GSC interface is limited — working in a sheet lets you apply your own logic.
Signal-one: High Impressions, Low CTR
The most valuable — and most ignored — signal in the dataset.
A page with 4,000 monthly impressions and a 1% CTR isn’t a failure. It’s a specific problem: search intent exists, Google surfaced you, users didn’t click. The gap lives in the title, meta description, or format — not demand. You already have the ranking. You’re leaving the traffic on the table.
To understand the stakes: research on organic CTR by position consistently shows position 1 earns around 28–39% of clicks, position 3 earns roughly 10%, and by positions 8–10 you’re looking at 1–3%. A high-impression page stuck at position 6 with a weak title is perpetually bleeding clicks it could be capturing.
Filter by impressions descending. Find pages with CTR below 3% on positions 1–10. These are your first targets. Rewrite the title to match the specific query — use the actual search term your audience typed, not your preferred phrasing. Update the meta description with a concrete benefit, not a vague summary.
What this looks like in practice: You have a post on “content marketing strategy” ranking at position 7 with 5,200 impressions but 1.1% CTR. The title reads “Our Approach to Content Marketing.” Rewriting it to “Content Marketing Strategy: A Practical Framework for 2025” directly targets the searcher’s intent. You don’t need a new article. You need a better label on the one you have.
Signal-two: Striking Distance Keywords
Sort by average position. Filter to positions 5–20.
Your site already proved relevance for these queries. Google decided your content belongs near the top. A modest improvement yields disproportionate traffic because the CTR curve is steep at the top of page one. Moving from position 8 to position 3 on a high-impression query means jumping from roughly 2% CTR to roughly 10% — a 5× traffic gain on the same URL without publishing anything new.
For each query in this range, ask two questions:
- Does a dedicated page exist — or is the topic buried inside a broader article?
- Does the page directly answer the query in the H1, introduction, and subheadings?
If either answer is no, you have a clear task: build a focused article, or pull the buried section out into a standalone piece and interlink both.
What this looks like in practice: You run a site about project management. Search Console shows the query “kanban vs scrum for small teams” landing at position 14, with 1,900 monthly impressions. Your site has a general post comparing agile methodologies that mentions both — briefly. A dedicated article targeting that exact phrase, with a proper comparison table, is a clear content decision. You already know the demand exists. You just need the page.
Signal-three: Query Diversity by Page
Click a specific URL in the Pages view. Switch to the Queries tab.
You’ll see every search term that URL ranked for. High-performing pages often surface 20, 50, or 100+ associated queries — each is effectively a subtopic map. Individual clusters point to standalone articles your site doesn’t have yet.
This builds topical authority systematically. The queries your existing content attracts tell you what adjacent topics your audience is searching for. Write those articles. Link them back.
What this looks like in practice: Your post on “email marketing for e-commerce” ranks for 60+ queries. Scrolling through, you spot a cluster around “abandoned cart email subject lines,” another around “post-purchase email sequence,” another around “email list segmentation for Shopify.” Each of those is a standalone content opportunity with confirmed demand — the impressions are already there. None of them required you to guess at a topic.
This is one of the most underused signals in Search Console. Most teams look at the aggregate performance of a page. The query-level breakdown tells you which branches of a topic are worth growing.
Signal-four: Trending Queries Over Time
In the date range selector, use the Compare option. Set this period against the same period last year (or the prior quarter, depending on your site’s age and traffic volume). Sort the query list by impressions delta — largest increase first.
Queries climbing fast are worth publishing before they peak. Two things this surfaces:
Seasonal patterns. If a topic spiked in impressions last October, it will likely spike again. The goal is to publish — or publish an updated version — before the surge, so your content has 4–8 weeks to index, accumulate internal links, and build signal before the peak traffic window opens. A post that goes live at the top of a spike starts earning traffic immediately. One published during the decline catches the tail.
Emerging topics. Some queries don’t have seasonal patterns — they just start growing because the topic is new or newly relevant. Early coverage here compounds. Being an early result on a climbing query means earning links and engagement before competition arrives.
What this looks like in practice: You run a tax and accounting blog. Year-over-year comparison in February shows “self-employed tax deductions 2025” climbing 340% in impressions vs. the same window last year. That’s not a coincidence — it’s a signal to publish or refresh that specific article by early March, well before the April filing rush.
Signal-five: Zero-Click and Featured Snippet Opportunities
Some queries show high impressions, position 1, and almost no clicks. That’s often a sign Google is answering the question directly — through a featured snippet, a knowledge panel, or an AI overview — before the user reaches your result.
The decision here is different: it’s not about writing more, it’s about structuring your content to own the snippet position. Format your answer as a concise definition (40–50 words), a numbered list, or a comparison table. Structured answers are more likely to be pulled into snippets, which can either win the zero-click impression or drive the click when the snippet builds enough trust.
Conversely, a query where you’re generating the snippet but someone else is getting the traffic (position 0 with low CTR) suggests the snippet itself — the preview text Google is pulling — needs rewriting to create curiosity rather than complete the answer.
Signal-six: Country and Device Filters
Apply the country filter. Significant impressions from a market your current content doesn’t specifically address — with different terminology, pricing context, or local intent — point to an underserved audience.
Same for devices. Strong desktop rankings but weak mobile CTR suggests a mobile formatting problem. Users on mobile abandon slower, harder-to-scan pages. That’s a content retrieval and readability problem, not just a design issue — shorter paragraphs, cleaner subheadings, better table rendering.
These rarely drive the first content decisions for small sites. For sites with multiple properties, international traffic, or scaling operations, they surface gaps that aggregate view hides.
Putting-it-together: A Weekly GSC Workflow
Run this review weekly, not monthly. Thirty minutes is enough if you keep the export in a live sheet.
- High-impression, low-CTR pages → queue for title and meta rewrites
- Striking-distance queries (positions 5–20) → assign to new articles or existing pages needing depth
- Query diversity by top URLs → add identified subtopic clusters to the content calendar
- Date-range comparison → identify trending and seasonal topics, schedule publication ahead of the peak window
- Featured snippet candidates → restructure answer formatting on relevant pages
This isn’t a one-time audit. It’s a loop. The site changes, rankings shift, new queries emerge — the data updates continuously. The bottleneck isn’t strategy. It’s consistent execution: doing this regularly, publishing the output, keeping the loop running.
When the Workflow Itself Becomes the Problem
The process above works. The friction is doing it repeatedly, week after week, then acting on the output fast enough to matter. Most teams run the audit, build a list, and watch it age in a spreadsheet.
Ghost Writr is built to run this loop autonomously. It reads your Search Console data continuously — flagging high-impression, low-CTR pages, identifying striking-distance queries, surfacing query clusters from top-performing URLs — and drafts the corresponding content or rewrites. Articles, meta updates, and refreshes are queued for your approval, then published directly to your site. The system runs in the background; you approve what matters.
One plan, per site — see pricing.
The data exists. The question is whether you act on it consistently enough for it to compound.