How to find low hanging fruit keywords in Google Search Console

You open Search Console, you see a table of a few thousand queries, and you close it again. It is a reasonable reaction. The report shows everything and suggests nothing.

A small part of that table is worth your attention, and the filters that find it are already in the interface. These are the low hanging fruit keywords: queries where Google already shows your site, just below the place where people start clicking. This guide gives you the filters, a rule for which rows to ignore, and a decision for each row that survives. At the end you will have a short list you can act on.

What "low hanging fruit" actually means

A low hanging fruit keyword is a search where your site already appears in Google's results, close to but below the positions that get clicks. Google has already decided your page is relevant. The work left is smaller than starting a new page, because you are improving something that half works.

You may have seen the same idea called "striking distance keywords" in other articles. It is the same thing. Three signs that a query belongs in this group:

• Your site already appears for it. It has impressions, so it is being shown.
• The average position is close to the first page, but not near the top of it. Usually somewhere between 5 and 20.
• It is a query you would genuinely want traffic from. Someone searching it could plausibly become a reader, a subscriber or a customer.

If any of those three is missing, the row is not low hanging fruit, whatever the numbers look like.

Why position 5 to 20 is the zone worth your time

Position in Search Console is the average place your site appeared for that query, counting the topmost result whenever it was shown (Search Console Help, checked 9 September 2026). Different positions usually mean different problems, so the band tells you what kind of work is in front of you.

Position band	What it usually means	Worth touching
1 to 4	The page is already working for this query	Only if the clicks look low for the impressions
5 to 10	Google is showing you on page one, lower down	Yes. Often the title and description do the most work here
11 to 20	Just off page one, so almost nobody sees it	Yes. This usually needs the content improved, not the snippet
21 and below	A long way back, for many possible reasons	Usually more expensive than writing a new page

The bands are a rule of thumb, not a law. Published click-through-rate curves by position, such as the free monthly study from Advanced Web Ranking (checked 9 September 2026), are orientation only: your own numbers will differ, and that study notes that when an AI Overview appears, the top position typically sees a markedly lower click-through rate than on a standard results page. Nothing here predicts that a change will move your position.

Set up the report step by step

Seven steps, each one a single action in the interface. Nothing here needs a paid tool.

1. Open the Performance report. In Search Console, choose your site, then Performance, then Search results.
2. Set the date range to Last 6 months. The default is three months, and 28 days is far too short: a short window hides queries that trickle steadily and over-weights whatever happened last week. Search Console keeps data for the last 16 months (Google, checked 9 September 2026), so six or twelve months is available to you.
3. Turn on all four metrics. Click Total clicks, Total impressions, Average CTR and Average position at the top of the chart, so all four appear as columns in the table below. The last two are off by default and they are the two you need.
4. Switch to the Queries tab. The table underneath the chart has tabs. Queries is the first one.
5. Filter by position. Use the filter above the table, choose Position, then Greater than 5. Look at that list, then run the same report again with Smaller than 20. The overlap of the two passes is your zone.
6. Sort by impressions, highest first. Click the Impressions column heading, then draw a line: on a site with steady traffic, ignore anything under about 100 impressions in six months, because there is not enough data there to judge.
7. Take the top twenty rows. Export them, or just work down the list on screen. Twenty is enough for a first pass.

If your site is small. A site with 300 impressions a month will return nothing at a 100 impression threshold. Do not change the method, change the numbers: set the date range to 12 months and lower the impression threshold to 20 or 30. The logic is identical, and the list will simply be shorter.

What to do when the list is empty

If both filters together return nothing, that is useful information rather than a failure. It means your site does not yet have positions in that band, and the next question is a different one: is Google finding and indexing your pages at all, and is anything on the site getting in the way. That is a whole-site question, and what an SEO audit checks walks through the areas to look at, in order.

High impressions, low clicks: two different problems

A row with many impressions and almost no clicks is not one problem. It is one of three, and they need completely different work. Read the position column before you decide anything.

What you see	What it usually is
Position 5 to 10, clicks low for the impressions	The result is seen and not chosen. A snippet problem
Position 11 to 20, clicks near zero	A normal outcome for that position. A content problem
High impressions, near zero clicks, query unrelated to your business	Not an opportunity at all. Noise

Branch A, a snippet problem. People see your result and click something else. The title and the description do the persuading here, and both are editable in a few minutes. Make the title answer the exact question in the query, in words a person would use.

Branch B, a position problem. At position 14 almost nobody scrolls that far, so a low click-through rate is the expected result rather than a symptom. Rewriting the title will not help much. What usually helps is the page answering the query more completely than it does now.

Branch C, noise. A page sometimes picks up impressions for a query unrelated to what you do. Nothing to fix. Skip the row.

Decide what to do with each row

Four questions, in this order. Stop at the first one that gives you an action.

1. Is there already a page Google shows for this query? Click the query in the table, then open the Pages tab. Search Console will show which URL was served. If there is one, you are editing, not writing.
2. Does that page actually answer this query, or a neighbouring one? Read the page as if you had typed the query. If the page is about a related but different question, the honest answer is a new page for this query, and an internal link between the two.
3. Is there a direct answer in the first hundred words? Many pages answer the question in paragraph nine. Moving a two sentence answer to the top is often the single cheapest change available, and it also makes the page easier to quote.
4. Are two of your own pages competing for the same query? If the Pages tab shows the served URL flipping between two of yours, they are splitting the signal. Usually the better move is to merge them into one page and redirect the weaker URL.

The four actions this produces, in rough order of effort: rewrite the snippet, add a missing section, write a new page, merge two pages into one. If you draft any of that with a language model, check every number, date and name against a source before it goes live, because the page will carry your name and not the model's.

Rows you should skip

A stop list saves more time than the filters do. Skip a row if it is:

• Your own brand name. Those people were coming anyway, and the position is not the interesting part.
• Somebody else's brand name. You are unlikely to outrank a company for its own name, and it is not traffic that converts.
• A question Google answers on the results page itself. Conversions, dates, definitions of one word. The click often never happens for anyone.
• A query about something you do not do. Interesting, occasionally flattering, still not your business.
• A single spike rather than steady impressions. Look at the trend, not the total. One news week is not demand.
• A results page filled entirely with marketplaces and aggregators. If the whole first page is large platforms, a small site is not the missing tenth result.

Check whether it worked

Three steps, and the third one is the part almost everyone skips.

1. Write down the date you changed the page, and what you changed. A note in a text file is enough.
2. Wait four to eight weeks. Google has to recrawl, and the report lags by a couple of days.
3. Open the same report with the same filters and turn on Compare. In the date range picker, choose Compare, and set the period after your change against the period before it. Read impressions, clicks and average position side by side for that query.

Sometimes the position moves, sometimes it does not, and sometimes it moves for reasons that have nothing to do with you. The point is that you can see it either way, on your own data, rather than taking anyone's word for it. If your list came back with fifty rows and you cannot tell which of them to start with, it can help to look at the whole site first: the Kitto audit puts its findings in priority order, and you can run a free SEO audit without a card or an account.

A quick recap

1. Open Performance, Search results, Last 6 months, all four metrics on, Queries tab.
2. Filter Position greater than 5, then smaller than 20, and sort by impressions.
3. Drop the rows on the stop list before you look at anything else.
4. For each remaining row, read the position to decide whether it is a snippet problem or a content problem, then work through the four questions to pick one action.
5. Note the date, wait four to eight weeks, and compare the same report against itself.

This is the same sort of review that AI SEO content software runs programmatically, and it is worth doing by hand at least once so that you know what such a tool should be looking at. Kitto builds its plan from exactly these numbers, and the method above stays free and yours whether or not you ever use it.