SEO
Free SEO traffic forecaster from keyword volume and position
By Charles Summers · Updated · Free, no signup
Short answer
Paste keywords and monthly search volumes, choose the ranking position you hold or are aiming for, and this applies an explicit position-to-CTR curve to produce projected monthly clicks per keyword and in total. The curve is printed in the output so you can argue with it. Treat the result as a planning band, not a forecast: published CTR curves disagree with each other by a factor of two at position one, mostly because SERP features change the answer more than rank does.
Use the seo traffic forecaster
What does this tool actually do?
Paste keywords and monthly search volumes, choose the ranking position you hold or are aiming for, and this applies an explicit position-to-CTR curve to produce projected monthly clicks per keyword and in total. The curve is printed in the output so you can argue with it.
It runs entirely in your browser. Nothing you type is sent to a server, no account is required, and there is no usage limit, because there is no cost per run to control.
Every published CTR curve disagrees with every other one, for a reason
Look up the click-through rate for position one and you will find estimates spread from around a fifth of searches to around two fifths, depending on whose dataset you read. The spread is not sloppiness. Each study is measuring a different population: some pool Search Console data across thousands of consenting sites, some use clickstream panels, some restrict to desktop, some blend branded and non-branded queries. Branded queries alone shift the average enormously, because a search for a company name sends most of its clicks to that company regardless of layout.
The shape is more reliable than the level. Across essentially every dataset, the drop from position one to position three is somewhere near a factor of two and a half to three, positions four to ten flatten into a long shallow tail, and the difference between position eight and position ten is small enough to be noise. That means a forecast that moves you from ten to five is far more sensitive to curve choice than one that moves you from three to one, and it means chasing the last two places on page one is usually the worst-value work available.
This tool uses a single blended curve, printed in the output, sitting at 27% for position one and 1.9% by position ten. That is deliberately mid-range rather than optimistic. If you have twelve months of Search Console data for the site in question, your own average CTR by position is strictly better than any published curve, because it already accounts for your brand strength, your SERP layouts and your snippet quality. Use it. This tool exists for the case where you do not have the site yet.
SERP features move the curve more than position does
Position is a coordinate on a page whose contents change per query, and the contents matter more than the coordinate. Ranking first below an AI-generated answer, a featured snippet you do not own, four shopping ads and a local pack is a different commercial position from ranking first on a clean ten-blue-links page, even though both are "position one" in your rank tracker. Any forecast built from rank alone is silently assuming a SERP layout it never looked at.
The direction of the effect is consistent even where the magnitude is contested: features that answer the query in place suppress clicks to everything below, and features that qualify intent (shopping units, local packs) redirect clicks to a different kind of result rather than merely reducing them. Zero-click behaviour is real and growing, but the headline percentages quoted for it vary wildly with definition, since counting navigational searches, refinements and in-SERP answers differently produces different answers to the same question. The honest position is that the direction is certain and the size is dataset-specific.
Before you trust any number this tool gives you, spend twenty minutes actually looking at the results pages for your top ten keywords by volume, and adjust:
- An AI answer or featured snippet occupying the top of the page. Expect meaningful suppression of organic clicks below it. If you own the snippet the picture is different and often better, but you own it at the pleasure of the algorithm.
- Ads above the fold. On commercial queries the first organic result can sit below several paid units, which pushes the effective position of everything organic down the page regardless of rank.
- A local pack or map unit. If the query has local intent and you are not in the pack, your organic click share is being competed for by a block most users treat as the answer.
- Sitelinks under a competitor. A result with sitelinks occupies several times the vertical space of a plain one, which changes what "below it" means for everyone else.
- A forum or video block. Discussion and video units absorb a share of clicks that never touch the standard results, and they appear disproportionately on exactly the informational queries most content plans are built around.
The volume number is an estimate too, and the errors multiply
Keyword tools do not have your search volume, they have a model of it. Google Keyword Planner reports rounded, bucketed monthly averages built for advertisers and frequently groups near-identical variants into one figure. Third-party tools model volume from clickstream panels calibrated against that same bucketed data. Comparing either against Search Console impressions for a query you already rank for routinely shows discrepancies of a factor of two in both directions, and the direction is not consistent enough to correct for with a single fudge factor.
Then there is annual averaging. A keyword with a monthly average of 5,000 might be a flat 5,000 every month or it might be 40,000 in November and 700 in June, and a content plan built on the average will either miss the window or drastically overbuild for the trough. Check the trend line for anything you are betting real budget on, and plan publication dates from the shape rather than the mean.
The forecast this tool produces multiplies two estimates together, so their errors compound rather than cancel. If your volume figure is 40% high and your CTR curve is 30% optimistic, your click forecast is nearly double what you will get, and nothing in the arithmetic will warn you. This is why the output is worth reading as an order of magnitude and a ranking of opportunities, and why the most useful thing on the page is often not the total but the comparison between what the same keyword list is worth at position three versus position seven.
Making a forecast that survives contact with a stakeholder
Present a band, not a number. Take the output here, halve it for the pessimistic case and add roughly half again for the optimistic one, and put all three on the slide. The band is not hedging; it is the actual precision of the inputs, and stating it honestly at the start is considerably better than defending a single figure that was never going to be right. A forecast presented as a point estimate will be quoted back to you as a commitment, and you will spend the following two quarters explaining a variance that the method could never have avoided.
Attach a timeline, because clicks do not arrive when the page publishes. New content on an established domain typically takes several months to reach a stable position, and longer on a young site or in a competitive category. A forecast with no ramp is wrong in month one by roughly one hundred percent, which is the version of the error people notice first and remember longest.
Then discount for the part of the traffic you cannot use. Not every click is a prospect: informational queries at the top of the funnel convert at a fraction of the rate of commercial ones, and a plan that hits its click target entirely through definitional queries has succeeded at the metric and failed at the job. Split the keyword list by intent before you forecast, run the numbers separately, and let the two totals be reported separately. When someone asks what the SEO programme is worth, the answer they want is in pipeline, and getting there requires exactly one more multiplication than most forecasts bother with.
Finally, write the assumptions down next to the number: which curve, which position, which volume source, which date. In six months somebody will compare actual traffic against this forecast, and the only way that conversation produces learning rather than blame is if the assumptions are recoverable. A forecast whose inputs are lost cannot be wrong in any useful way, which sounds like safety and is actually the reason most teams never improve their estimates.
Numbers worth knowing
| Metric | Typical | What it means |
|---|---|---|
| Position 1 CTR across published studies | roughly 20% to 40% | The spread comes from different datasets, devices and branded-query handling, not from one study being wrong. This tool uses 27% as a mid-range default. |
| Drop from position 1 to position 3 | about 2.5x to 3x | The most consistent finding across every dataset. It is why the first three places carry most of the value and positions 6 to 10 differ from each other very little. |
| Page two visibility | under 1% per position | Traffic beyond position 20 is close to a rounding error. Ranking 14th is not "nearly there", it is a different traffic regime from ranking 4th. |
| Third-party volume vs Search Console impressions | often off by 2x either way | Keyword tools model volume from panels calibrated on bucketed advertiser data. Where you already rank, trust your own impression data over any tool. |
| Time for new content to reach a stable position | several months | Faster on an established domain with topical authority, slower on a new site or in a competitive category. A forecast without a ramp is wrong in month one by about 100%. |
Mistakes that quietly cost you results
- Forecasting from rank without looking at the actual results page
- Position one under an AI answer, a snippet you do not own and four shopping ads is not the same commercial position as position one on a plain page. Open the top ten queries by volume, note the features, and discount accordingly before anyone sees the total.
- Presenting a single number instead of a band
- You are multiplying a modelled volume by a modelled CTR, so the error compounds. Show a pessimistic, expected and optimistic case, roughly half to one and a half times the central figure, and say why the range is that wide.
- Treating an annual average volume as a monthly reality
- A 5,000-a-month keyword can be 40,000 in November and 700 in June. Check the trend before committing budget, and set publication dates from the seasonal shape rather than the mean, or you will build the page a month after the window closed.
- Forecasting clicks and reporting them as business value
- Split the list by intent first. Informational and commercial queries convert at very different rates, so a plan that hits its click target entirely on definitional queries has passed the metric and failed the objective. Run the two totals separately.
- Using a published CTR curve when you have Search Console data
- Your own average CTR by position already accounts for your brand strength, your SERP layouts and your snippet quality. Any generic curve, including the one in this tool, is a substitute for data you may already have sitting in an export.
What does the output look like?
This is the exact output the tool produces from the example inputs. It is generated by the same code that runs when you click the button, so what you see here is what you get.
Frequently asked questions
Where does the CTR curve come from?
It is declared explicitly in the tool and printed in the output: 27% at position one, 15% at two, 10% at three, easing to 1.9% by position ten and under 1% beyond position eleven. It is a deliberately mid-range blend rather than a copy of any single published study, because the published studies disagree with each other by roughly a factor of two at the top of page one. You can see every value the tool used and substitute your own judgement where you know better.
What do the three site profiles actually change?
They multiply the entire curve. Brand-strong applies 1.35x, on the basis that a recognised name earns clicks a generic domain in the same position does not, particularly on queries where users are choosing between familiar and unfamiliar sources. Weak applies 0.65x for unknown domains, thin titles and feature-heavy results pages. Typical applies 1.0x and leaves the curve alone. The result is capped at 55%, because no realistic organic position collects more than about half of all clicks for a non-branded query.
How should I format the keyword list?
One keyword per line, then a comma, then the monthly search volume: "project management software, 18100". Volumes containing thousands separators are handled, and a tab or a pipe works as the separator instead of a comma. Any line the parser cannot read is reported back to you rather than silently dropped, so a total that looks low is always traceable to specific lines.
Why does the output show positions 1, 3, 5 and 10 as well as the one I chose?
Because the interesting question is almost never "what is this worth where I am", it is "what is the movement worth". Seeing the same keyword list priced at four positions turns a forecast into a prioritisation: if going from seven to five adds very little but going from three to one nearly doubles the total, that tells you where to spend, and it tells you in a form that survives being wrong about the absolute numbers.
Can I use this to forecast revenue?
Only with two further multiplications you should do deliberately. Clicks times conversion rate gives leads, and that conversion rate differs by an order of magnitude between an informational query and a commercial one, so run the two lists separately or the average will be meaningless. Then apply your close rate and deal value. Do that on the pessimistic end of the click band, not the central estimate.
Related free tools
- SEO Meta Tag Generator SEO
- Schema Markup Generator SEO
- Open Graph Tag Generator SEO
- Competitor Ad Deconstructor Competitive intel
- CAC Payback Period Calculator Analytics
Some links on this site are affiliate links, which means Hacking Demand may earn a commission if you buy through them at no extra cost to you. This does not influence which tools are listed. The tools on this page are free and have no affiliate relationship of any kind.