How can I increase my organic traffic?

“Increase organic traffic” is a goal, not a plan. Turning it into one starts with recognising that organic traffic is an equation, and only four things in it can move.

The equation

Organic traffic is the number of queries you appear for, multiplied by how often you are shown for each, multiplied by how often people click.

Traffic = queries you rank for × impressions per query × click-through rate

That gives four levers, because impressions per query is really two things — your position, and the underlying demand.

Most teams start at lever three — publish more — because it feels like doing something. It has the second-longest payback of the four and it is frequently not the binding constraint.

Lever 1: click-through rate

The fastest lever, operating entirely on traffic you have already earned.

Export Search Console, sort by impressions descending and CTR ascending. The top of that list is pages shown to thousands of people who consistently choose someone else. The ranking work is done; what is failing is the twenty seconds of writing that converts a listing into a visit.

What moves it: titles using the searcher’s phrasing rather than your internal vocabulary; specificity over cleverness; a visible reason this result answers the question; numbers and dates where honest; descriptions written as a promise rather than a summary.

What does not: keyword stuffing, brand-first titles, and clickbait — which lifts clicks briefly and returns visitors to Google faster, which is visible and counterproductive.

A realistic expectation is a two to four point CTR improvement on the pages you touch. On a page with 40,000 monthly impressions, that is over a thousand additional visits for an afternoon of writing.

Lever 2: position

The second fastest, and the one with the best return per unit of effort — provided you work the right band.

The click-through curve is steepest between roughly positions ten and four. Moving from 15 to 11 changes very little; moving from 11 to 6 can double a page’s traffic. That is why the 8–20 band is where position work belongs, and why moving a page from 60 to 40 is usually a waste of a week.

The work at this band is depth: cover the subtopics the ranking pages cover and you do not, answer the core question earlier, refresh anything that has aged, and route internal links from your strongest related pages. Our guide to ranking higher by band covers this in detail.

Lever 3: query coverage

Ranking for things you do not currently rank for at all. This is where new content belongs — and only here.

The discipline that makes it work is targeting genuine gaps rather than adjacent restatements of what you already have. A new page on a topic you already cover does not add coverage; it splits it, and both pages rank worse than the original did.

Find real gaps through competitor comparison, question mining, your support inbox and internal site search. Then cluster them — two hundred missing keywords are usually eight to twelve pages, not two hundred. Our guide to finding keyword opportunities covers the sources.

Lever 4: demand

Making more people search for what you do. This is brand and category work — PR, community, product, word of mouth — and it operates on a horizon of years rather than quarters.

It is out of scope for most SEO programmes, with one exception worth naming: branded search is the one form of demand you can directly create, and it is the most valuable traffic on your site. Anything that makes people remember your name — a genuinely useful free tool, original research people cite, a visible presence where your audience already gathers — converts into searches that have no competition at all.

This lever has also become more important for an indirect reason. Brand strength and consistent third-party description are among the strongest inputs into whether an AI assistant names you when someone asks for a recommendation. Demand work now pays into a second channel that did not exist three years ago.

Which lever is constraining you

Twenty minutes in Search Console answers this. Export twelve months of query data and check four things.

1. Count queries with impressions. Flat or falling over a year means coverage is your constraint — lever three.
2. Check the position distribution. A large cluster in the 8–20 band means position is your constraint — lever two, and it is your best opportunity.
3. Check CTR against typical rates for your positions. Materially below par means lever one, and you can start this afternoon.
4. Check impressions against position. Good positions with falling impressions means demand is moving — lever four, or the pie itself is shrinking.

Most sites find their answer is lever one or two, and most sites are spending their budget on lever three. That mismatch is the single most common reason a content programme runs for a year without moving the number it was funded to move.

When the pie is shrinking

One scenario breaks the equation, and it is increasingly common: your position holds, your CTR holds, and traffic falls anyway.

That is AI Overviews absorbing the click. Pew Research, tracking roughly 68,000 real queries, found users clicked a result in 8% of visits where an AI summary appeared against 15% where one did not. Ahrefs measured a 58% click-through decline for top-ranking pages in February 2026, up from 34.5% in April 2025.

When that is what is happening, pulling harder on levers one to three produces diminishing returns, because you are competing for a shrinking share of the same page. The response is different: be the source cited inside the answer rather than the result listed beneath it, and start measuring visibility on surfaces where no impression is ever recorded. Our AI Search 2026 report covers the data, and the combined growth guide covers running both motions together.

A worked example

A B2B site wants to double organic traffic in twelve months. Current state: 18,000 monthly clicks, 900,000 impressions, 2% overall CTR.

The diagnosis. A 2% site-wide CTR against 900,000 impressions is the first signal — that is low, and it means a great deal of visibility is not converting. The position distribution shows 340 queries in the 8–20 band with over 100 impressions each. Query count has grown 8% year over year, so coverage is expanding slowly but is not the binding constraint.

The plan that follows. Lever one first: rewrite titles and descriptions for the forty highest-impression, lowest-CTR pages. Even a two-point CTR improvement across a third of total impressions adds roughly 6,000 monthly clicks — a third of the way to the goal from a fortnight of writing.

Lever two second: take the top sixty striking-distance queries, cluster them onto the twenty-five pages that own them, and deepen each against the subtopics the ranking competitors cover. Moving even half of those from the 8–20 band into the top five roughly doubles their traffic individually.

Lever three third, and only for genuine gaps — perhaps eight new pages across the year, each targeting a cluster nobody on the site currently addresses.

Lever four continuously, but measured as branded search growth rather than as an SEO deliverable.

What this plan deliberately does not include: a weekly publishing cadence, a site-wide speed project, or a schema rollout. None of those is the constraint, and each would consume the capacity the first two levers need.

Where the equation breaks down

One assumption in the equation is doing quiet work: that an impression is worth roughly what it used to be. On queries where an AI Overview appears, it is not. The same position now yields materially fewer clicks than the equivalent position did two years ago, which means a plan built on historical CTR benchmarks will overshoot its forecast even if every piece of work lands.

Adjust for it explicitly. Check AI Overview presence across a sample of your priority queries, and apply a discount to expected CTR on the affected subset rather than modelling the whole site at one rate. It makes forecasts less flattering and considerably more accurate.

The most common planning error

Setting a traffic target without checking whether the demand exists to reach it. If every query you rank for, at position one with a perfect CTR, would still only produce 40,000 clicks, then a 60,000-click target requires lever three or four by definition — and pretending otherwise guarantees the plan fails.

Calculate your ceiling before committing to a number. Total impressions multiplied by a realistic best-case CTR gives you the honest upper bound of what the current query set can produce. If the target exceeds it, the plan needs new coverage or new demand, and knowing that in month one is worth a great deal more than discovering it in month nine.

Sources

• Pew Research Center — approximately 68,000 real search queries; clicks fell from 15% to 8% of visits when an AI summary was present.
• Ahrefs — CTR impact of AI Overviews; 34.5% decline for position one (April 2025), approximately 58% (February 2026).
• Publisher traffic reporting, H1 2025 — median publisher down ~10% year over year; non-news content sites down ~14%.

Frequently asked questions