Low CTR at a good position: title, or AI Overview?

A page can rank on the first page of Google and still barely get clicked. It shows up for the search, sits at position four or five, collects impressions, and yet the clicks only trickle in. That gap between where you rank and how often you get chosen is the cheapest traffic in your Search Console account, because you have already done the expensive part: you earned the ranking. Winning a larger share of the clicks already in front of you is a title-and-description edit, not a months-long fight to climb the results.

The catch is that a low click rate at a good position has two very different causes, and Search Console shows you the same low number for both. Sometimes the title genuinely undersells the page and a rewrite wins the clicks back. Sometimes the page is fine and an AI Overview is answering the query above it, so the click never reaches anyone. Telling those two apart, in your own data, is the whole difference between a quick win and wasted effort.

TL;DR:

• A CTR quick win is a page that already ranks well (roughly positions 3 to 10) but earns fewer clicks than that position should, so a better title and description can capture clicks you are already being shown for.
• Judge the click rate against your own site's pages at the same position, not a generic CTR table. Read it on page-level data, the honest source for volume.
• A low click rate at a strong position is often not a weak title. It is frequently an AI Overview taking the click. The fingerprint is high impressions, a strong position, and a click rate near zero.
• When the title is the problem, rewrite it to match the query and be specific, then measure the lift over the next few comparable weeks.
• When an AI Overview is the problem, a rewrite does nothing. The move is to become the source it cites: a crisp answer, a specific stat, a clearly sourced fact.

What is a CTR quick win?

A page's click-through rate is the share of the people who saw it in the results who actually clicked. A quick win is a page where that share is low while the ranking is already good. The whole advantage is in that second half: ranking is the part that costs months of work. A page stuck at position thirty needs real effort to move. A page sitting at position four with a title that undersells it needs one edit to the title and description, and it is competing for clicks it is already being shown for.

That is why this is the first move to make on an existing site with traffic. You are not chasing new rankings or new demand. You are collecting clicks that already land on the results page next to your listing and choosing someone else. If you want the ground under the two numbers first, what CTR and position honestly measure covers both before you act on them.

Which pages are the quick wins?

Three conditions have to hold together, or the signal is not real:

• A good but not top position, roughly 3 to 10. The page is on page one, or close to it, and already earns impressions. The room to gain is in winning more of the click, not in moving up. A page sitting just off page one is a different job: there the move is to climb onto page one, not to win a bigger share of clicks you already get.
• Enough impressions that the click rate means something. A page shown fifteen times with one click has a click rate you cannot trust. Set a floor and ignore the pages below it, the same way you would ignore a percentage swing on a page with three clicks.
• A click rate below what your other pages earn at that position. This is the actual signal, and the next section is about how to judge it honestly.

Rank the candidates by the clicks a reasonable title could realistically win, which is roughly the impressions multiplied by the gap between the current click rate and a fair one. That sorts the pages with real upside to the top and keeps a page with a slightly soft click rate but tiny impressions from crowding them out.

What counts as a "good" CTR? Your own baseline beats a table

Every CTR-by-position table you find online is a rough shape, not a ruler. First Page Sage's 2026 numbers put position one near 39.8%, position two near 18.7%, and position three near 10.2% on a clean results page, and the curve keeps falling into low single digits by the bottom of page one. That is a useful gut check for the order of magnitude, and nothing more, because two assumptions inside it rarely hold for your pages.

The first is that all sites and queries behave alike. They do not. A branded query, a how-to query, and a product query at the same position earn wildly different click rates, so your real baseline is your own site: what do your pages at position four typically earn? That is the number a soft page should be measured against, and it is why page-level data is the honest source here. The per-query view undercounts, because Google hides rare queries while still counting their clicks in the page total, so judge the click rate at the page level or you will measure against a number that is already wrong. The difference between page-level and query-level data is the whole reason to start at the page.

The second assumption is that the results page is clean. In 2026, the most valuable queries rarely are.

The honest catch: a low CTR is often the AI Overview, not your title

Here is the trap that sends people rewriting titles that were never the problem. When an AI Overview sits at the top of the results and answers the query, the click often does not happen at all, no matter how sharp your title is. Ahrefs measured the top organic result losing 58% of its clicks when an AI Overview is present, across 300,000 keywords. Pew Research found people click a normal result on about 8% of searches that carry an AI summary, against 15% without one. Your title did not get worse. The results page changed shape above you.

The fingerprint to look for is specific: high impressions, a strong position, and a click rate near zero. A page at position three with thousands of impressions and a click rate of a fraction of a percent is almost never a title problem. It is a page whose click is being intercepted before it reaches anyone. This is the same honest read that separates real decay from a SERP feature on the content-decay side: when clicks fall while impressions and position hold, the page is not the thing that broke. Rewriting the title here spends an afternoon where it cannot help.

So the first thing to do with a low-CTR page is not to open the title tag. It is to look at the impressions and the position, and ask which of the two problems you actually have.

When it is the title: how to win the click

Once you have ruled out the AI Overview (the position is good, the impressions are healthy, and the click rate is merely below your own baseline rather than near zero), the title and description are the lever, and writing the title from the queries you already rank for is the whole job. What actually moves the click:

• Match the words of the query. If the page ranks for "how to do X" and your title is something clever that never says X, you lose the scan. Say the thing the searcher typed; a reader deciding between ten links picks the one that visibly answers them.
• Be specific over broad. A number, a year, a concrete outcome reads as more relevant than a vague promise sitting in a list of blue links. Specificity is what makes one listing look like the answer and the rest look like maybes.
• Write the description as the argument for the click, not a summary. Google rewrites descriptions often, but when it does show yours, a description that gives the reason to click still beats a restatement of the title.

Then measure. A title edit is a hypothesis, and the honest test is the click rate over the next few comparable weeks, read on page-level data so the hidden-query gap does not distort the before-and-after. Change one page's title at a time and note the date, or you will not know which edit moved the number.

When it is the AI Overview: cite, don't rewrite

If the page is holding its rank while an AI Overview caps the click, the move flips. You are no longer trying to write a better ad for the click. You are trying to become the source the AI Overview quotes, because a cited page earns the click back. Seer Interactive, across 53 brands and millions of queries, found a page cited inside the AI Overview earns roughly 120% more clicks per impression than one that is not.

What makes a page citable is not a keyword but a clean, quotable answer near the top of the page, a specific stat, and a clearly sourced fact. That is a different job from writing a title, so it gets its own walk-through: how to get cited in an AI Overview when a rewrite can't win covers finding those pages in your data and the honest ceiling on what citation recovers. The short version is the honest one: you can see the click being lost, but whether becoming the citation wins it back is something you confirm by measuring afterward, not something the rewrite guarantees. It is a better bet than a title rewrite on a page the title was never hurting, and it is still a bet.

The limits worth stating

• Judge CTR at the page level. The per-query click rate undercounts by the hidden-query share, so a page's true click rate lives in the page view. Use the query view to understand what the page ranks for, not to total its clicks.
• Position is an average, not a coordinate. The single "position 6" on a page blends your topmost placement across every query and every searcher who saw you, hiding the queries where you were second and the ones where you were fifteenth. Treat it as a rough band when you decide whether a page is "ranking well enough" to be a quick win.
• Winning the click is not winning the customer. A better title earns the visit. Whether that visit converts is a different measurement, after the click, in your analytics or your product. QueryScope reads your search data and stops at the click; it can show you the click you are leaving on the table, not what the click is worth once you win it.

Finding your quick wins without reading every page by hand

Done by hand, this is patient work: pull every page's clicks, impressions, and position, work out a fair baseline for each position on your own site, rank the pages by the clicks a reasonable title could realistically win, and on every one of them check whether an AI Overview is quietly taking the click before you spend an afternoon on a title that was never the problem. It is exactly the weekly discipline that slips first under a deadline.

That is what QueryScope does in your terminal. It reads your real Search Console data, flags the pages ranking well but under-earning the click, and separates a title you can fix from an AI Overview you cannot, so the pages actually worth a rewrite surface instead of hiding inside a healthy-looking total. If you want the one-line definition behind click rate or position first, the Search Console glossary covers each metric with the caveat that comes with it.

Sources

• First Page Sage, Google Click-Through Rates by Ranking Position (2026) (clean-SERP curve: position 1 ~39.8%, position 2 ~18.7%, position 3 ~10.2%, falling into low single digits by the bottom of page one).
• Ahrefs, Update: AI Overviews Reduce Clicks by 58% (December 2025 analysis of 300,000 keywords: the top organic result loses 58% of its clicks when an AI Overview is present, so a page can hold its rank and still lose the click).
• Seer Interactive, AIO Impact on Google CTR: 2026 Update (across 53 brands and millions of queries: a page cited inside the AI Overview earns roughly 120% more clicks per impression than one that is not).
• Pew Research Center, Google users are less likely to click on links when an AI summary appears (users clicked a traditional result on 8% of visits with an AI summary, against 15% without).
• Google Search Central, A deep dive into Search Console performance data filtering and limits (rare queries are dropped from the query table while their clicks stay counted in the page total, so page-level data is the honest source for volume).