SEO Forecasting: How to Predict Organic Traffic (Without Overpromising)

Published: 19 October 2026 | Last updated: 19 October 2026 | 20 min read

By Graeme Whiles

Somebody has just asked you how much traffic SEO will bring in next year.

Usually it's the finance director. Sometimes it's the founder. Either way, they want a number, and they want it by Friday.

So you open a spreadsheet, multiply search volume by a click-through rate you found in an old blog post, and send over a confident monthly total.

However, that number was wrong before you pressed send.

Not because SEO forecasting doesn't work.

Because a single number pretends you know things you don't, and in 2026 the click-through rates most templates rely on no longer describe the results page.

An honest SEO forecast is a range, with its assumptions written down where everyone can see them.

In this guide, I'll explain what SEO forecasting actually is, walk you through the two methods worth using, show you how to adjust for AI Overviews, and turn the result into revenue, ROI and a range your finance team will believe.

Short on time? Here are the key takeaways

• Forecast a range, not a number: conservative, expected and aggressive scenarios, with the assumptions written next to each one.
• Use two methods and compare them: a historical forecast for the traffic you already have, and a keyword-based forecast for the pages you haven't built yet.
• Adjust for AI Overviews: Ahrefs found they cut the top result's click-through rate by 58%, so a pre-2024 CTR curve will overstate your clicks.
• Strip out branded traffic first: it grows with your brand, not your SEO, and Search Console now separates it for you.
• Check it every month: compare forecast against actual, and re-forecast after every Google core update.

What is SEO forecasting?

First, let me explain what I mean by 'SEO forecasting', because it gets muddled with reporting all the time.

SEO forecasting is the process of estimating how much organic traffic a website will earn from search over a set period, and what that traffic is worth in leads, sales or revenue. It combines search volume, keyword rankings, click-through rate (CTR) and historical data, and the output should always be a range rather than a single figure.

Reporting tells you what happened.

Forecasting tells you what's likely to happen if you do the work you're planning, and what it's worth if you do.

A useful forecast answers three questions.

• How much traffic? Organic clicks or sessions by month, as a range.
• What will it be worth? Leads, sales or revenue, using your own conversion rate.
• What has to be true? The rankings, pages and links the numbers depend on.

That last question is the one most forecasts leave out.

It's also the one that saves you six months later, when someone asks why the numbers didn't land.

Why bother at all?

Because SEO is a budget line competing with paid search, and paid search comes with a forecast built in.

Google Ads will tell you roughly what £10,000 buys before you spend a penny.

SEO has to show its own working.

Why most SEO forecasts are wrong

Most SEO forecasts fail for the same five reasons, and none of them is bad maths.

• A single number: "48,000 visits a month" reads as a promise. The moment you miss it, the whole SEO programme looks like it failed, even if the work was right.
• A borrowed CTR curve: older click-through studies put position one at anywhere from 27.6% (Backlinko) to 39.8% (First Page Sage) of clicks. Ahrefs' aggregated Search Console data put position-one CTR on informational keywords at 3.9% in December 2025.
• Branded traffic left in: people searching for your company name grow with your brand, your PR and your ad spend. Leave them in and your SEO forecast takes credit for things SEO didn't do.
• No seasonality: a straight line drawn through a retailer's November peak will overshoot every month of the following spring.
• No allowance for AI Overviews: the answer at the top of the page now takes clicks that used to go to the first organic result.

The fix?

Build the forecast from your own data wherever you can, and only borrow other people's numbers where you have none of your own.

First-party vs third-party data: what to forecast from

Every SEO forecast runs on one of two kinds of data, and knowing which one you're using matters more than which tool you're using.

First-party data: Google Search Console and GA4

First-party data is what Google and your own analytics measured on your site.

Google Search Console gives you clicks, impressions, CTR and average position by query, page, country and device.

Google Analytics 4 (GA4) gives you what happened after the click: sessions, conversions, conversion rate and revenue.

It's the most accurate data you'll ever have, because it isn't an estimate.

The catch?

Search Console only keeps 16 months of performance data. If you want a longer history, export it every month or set up the bulk data export to BigQuery, before the oldest months disappear.

The other thing worth doing before you forecast anything is splitting out your brand. Google added a branded queries filter to the Performance report in November 2025 and opened it to all eligible sites in March 2026. It uses an AI-assisted classifier rather than a keyword list, so it also catches misspellings and product names.

Forecast the non-branded line. Report the branded line separately.

Third-party data: Ahrefs, Semrush and SE Ranking

Third-party data comes from SEO tools that estimate search volume, keyword difficulty and competitor traffic from their own panels and crawls.

It's the only option for keywords you don't rank for yet, and the only way to see what a competitor's traffic looks like.

However, it's an estimate, and different tools will give you different volumes for the same keyword.

My rule is simple: first-party data for anything you already rank for, third-party data for anything you don't. When the two disagree, Search Console wins.

Method 1: keyword-based SEO forecasting

Keyword-based forecasting builds the forecast from the bottom up, one keyword at a time.

Use it for new content, new categories, or a site without much history to project from.

The formula is short.

Forecast clicks = monthly search volume × your CTR at the target position × the probability you actually reach that position

Most templates skip the last part.

That's how you end up forecasting position one for every keyword on the list.

1. Group your target keywords into clusters

Forecast pages, not keywords.

One page ranks for a whole keyword cluster (a group of searches with the same intent), so group your target keywords by the page that will answer them and add up the search volume for each cluster. My guide to content cluster strategy covers how to group them.

Ahrefs' Traffic Potential metric is a useful sense check here, because it shows what the current top page earns across every keyword it ranks for.

2. Set a realistic target position

Look at who ranks now.

Check keyword difficulty, the strength of the sites in the top ten and whether the SERP features (AI Overviews, People Also Ask boxes, video carousels) leave much room for organic results. Then pick the position you can honestly reach in twelve months, not the one you'd like.

3. Use your own CTR curve

A CTR curve is the average click-through rate at each ranking position.

Build yours from Search Console: export your non-branded queries with their average position, round the positions, and take the median CTR for each one. Split it by whether an AI Overview appears if you have enough data, because those keywords behave very differently.

Only fall back on a published curve if your site is too new to have one.

4. Apply a ranking probability and a ramp

Not every page reaches its target, and none of them gets there on day one.

I use a ranking probability of 40 to 60% for most new content, then ramp it in: a small share of the forecast clicks in months one to three, around half by month six, and the full figure from about month nine to twelve.

5. Add it up by month

Here's what that looks like for a single new page targeting an "expense report template" cluster. The numbers are illustrative.

That's 231 clicks a month once the page is fully ramped.

Now assume every keyword hits position one at a 30% CTR, roughly what those older click-through studies suggested. The same cluster "forecasts" 2,730 clicks a month.

That's nearly 12 times too high, from one page.

Multiply that across a content plan of fifty pages and you can see how SEO earns its reputation for overpromising.

Method 2: forecasting from historical traffic

Historical forecasting projects your existing organic traffic forward using its own trend and seasonality.

Use it for everything you already rank for.

It's the more reliable of the two methods, because it's built on what actually happened rather than what a tool estimates.

1. Export at least 16 months of non-branded data

Pull monthly non-branded clicks from Search Console, and sessions and conversions from GA4 if you'll forecast revenue later.

Search Console's weekly and monthly views, added in December 2025, make the trend far easier to read than the daily chart.

Before you go further, mark anything that wasn't SEO: a site migration, a tracking break, a viral PR spike. Either remove those months or note them, because a time-series model will happily treat a one-off as a trend.

2. Find the trend and the seasonality

Look at year-over-year (YoY) growth first, because it cancels out seasonality on its own: this October against last October tells you more than this October against September.

Then work out a seasonal index for each month. Divide each month's clicks by your average month. A December index of 0.7 means December runs 30% below a normal month, and a September index of 1.2 means it runs 20% above.

A three-month moving average smooths out the noise if your traffic jumps around.

3. Project it forward in Google Sheets or Excel

You don't need special software for this part.

Google Sheets' FORECAST function fits a linear regression (a straight line of best fit) through your seasonally adjusted history and extends it forward. Multiply each future month by its seasonal index and you have a seasonal forecast.

If you work in Excel, FORECAST.ETS uses exponential triple smoothing and will detect the seasonality for you.

4. Go further with Prophet (optional)

If you have two or more years of data and strong seasonality, Prophet is worth an afternoon.

It's open-source forecasting software from Meta's data science team, it runs in Python or R, and you can use it in a free Google Colab notebook without installing anything. Ahrefs has published free Colab scripts that run it on SEO traffic data, which is the quickest way to try it.

The part I like most? It returns an uncertainty interval by default, so you get a range rather than a single line.

5. Layer your planned work on top

Here's what people miss.

A historical forecast is your baseline.

It tells you what happens if you keep doing exactly what you're doing now.

Add the keyword-based forecast for new pages on top of it. Then subtract a little for content decay, because older pages lose traffic if nobody updates them. My free content decay detector shows you which ones are already slipping.

Baseline, plus new work, minus decay. That's the whole forecast.

How AI Overviews change your SEO forecast

Most SEO forecasting guides still skip this part, and it's the one that matters most in 2026.

When Google shows an AI Overview, fewer people click the organic results underneath it.

How many fewer depends on whose data you read.

• Ahrefs: its February 2026 update found position-one CTR on keywords with an AI Overview fell from 7.3% to 1.6% between December 2023 and December 2025, about 58% lower than it would have been without one.
• Seer Interactive: its April 2026 study of 5.47 million queries found organic CTR on queries with an AI Overview recovering from 1.31% in December 2025 to 2.36% in February 2026, against 3.82% on queries without one.
• Pew Research Center: tracking 900 US adults, it found people clicked a traditional result in 8% of visits when Google showed an AI summary, against 15% when it didn't.
• SparkToro: its 2026 analysis found 68.01% of Google searches in the first four months of 2026 ended without a click.

Different methods, different datasets, same direction.

Zero-click search isn't a trend you can ignore in a forecast any more. It's the majority of searches.

1. Flag the keywords that trigger an AI Overview

Ahrefs, Semrush and SE Ranking all show which SERP features appear for a keyword. Add a column to your keyword forecast that marks whether an AI Overview shows up.

2. Apply an AI Overview discount

For those keywords, reduce the expected CTR.

If you have enough data, measure the gap on your own site. If you don't, the published figures put it somewhere between roughly 40% (Seer's latest numbers) and 58% (Ahrefs). I use 50% as the expected case, 60% in the conservative scenario and 40% in the aggressive one.

It's worth knowing that being cited inside the AI Overview helps. Across 2025, Seer found informational queries where the brand was cited averaged 2.07% organic CTR, against 0.94% where it wasn't.

3. Forecast AI visibility as a separate line

Clicks aren't the only thing worth forecasting now.

Search Console's generative AI performance report shows your impressions in AI Overviews and AI Mode by page, country, date and device. Google rolled it out to all sites worldwide on 31 August 2026. It doesn't give you clicks.

So track it as its own line, next to the traffic forecast rather than inside it. I cover the KPIs worth tracking in AEO vs GEO vs SEO, and how to measure your baseline in what is AI visibility.

How to turn a traffic forecast into revenue and ROI

Traffic is the middle of the forecast, not the end of it.

Nobody in finance signs off a budget for clicks. They sign off for revenue, so take the traffic range and run it through your own conversion data.

• Ecommerce: forecast clicks × conversion rate × average order value (AOV) = forecast revenue.
• Lead generation and B2B: forecast clicks × lead conversion rate × close rate × average deal value = forecast revenue.
• Subscription and SaaS: swap average deal value for customer lifetime value (CLV), because a customer is worth more than their first invoice.
• ROI: (forecast revenue minus the cost of the SEO work) ÷ the cost of the SEO work.

Take the expense report page from earlier. At 231 clicks a month, a 2% trial sign-up rate, a 20% trial-to-paid rate and a customer lifetime value of £1,800, it's worth roughly £1,660 a month once it's fully ramped.

That's about £20,000 a year from one page.

Use your own conversion rates from GA4 and your CRM, not industry averages.

And if you want a number finance will actually sign off, use gross margin rather than revenue when you work out ROI.

My free SEO ROI calculator does this maths for you, and my guide to content marketing ROI covers attribution in more depth.

How to present an SEO forecast (scenarios, not promises)

This is where most forecasts are won or lost.

Present three scenarios, each with its assumptions written in plain English underneath.

The numbers are illustrative. The structure is the point.

A range like this is a rough confidence interval: the band you'd expect the real number to fall inside most of the time. Prophet's default is an 80% interval, which is a sensible level to present.

Then write down what has to be true for each scenario.

• Output: how many pages get published, updated or merged, and by when.
• Rankings: the target position for each cluster, and the ranking probability you've assumed.
• Market: the AI Overview discount, seasonality and any known changes in demand.
• Dependencies: developer time, approvals, budget for links or digital PR.

When the forecast misses, those assumptions tell you why.

If the content shipped three months late, the forecast didn't fail. The plan did. That's a very different conversation, and having it in writing protects you.

The forecast should sit inside your wider SEO roadmap, so every number is tied to a piece of work someone has agreed to do.

How to check your SEO forecast was right

A forecast you never check is just a guess with a nicer font.

Backtest before you share it

Backtesting means forecasting a period you already have data for, then comparing the forecast with what actually happened.

Hold back your last six months, build the forecast from the months before, and measure the gap. The simplest error measure is mean absolute percentage error (MAPE), which is just the average of how far off each month was, as a percentage.

If your backtest misses by 25%, your range needs to be at least that wide.

Track forecast against actual every month

Put the expected line and the actual line on the same chart, with the range shaded behind them.

That chart belongs in your monthly report, next to the SEO KPIs the forecast is meant to move.

A month or two outside the range is noise. Three in a row means one of your assumptions is wrong, and you should find out which.

Re-forecast after core updates

Google ran two core updates in the first half of 2026 alone, from 27 March to 8 April and from 21 May to 2 June.

Each one can move rankings enough to break a forecast.

Check the Google Search Status Dashboard, wait for the rollout to finish, then re-forecast from the new baseline rather than arguing with the old one.

I re-forecast every quarter regardless.

When SEO forecasting doesn't work

Forecasting isn't the right tool for every situation, and pretending otherwise is how forecasts lose credibility.

• Brand-new websites: with no history and no rankings, you're forecasting from competitor benchmarks alone. Give a very wide range, or forecast the output (pages published, links earned) for the first six months instead of traffic.
• Site migrations and redesigns: the old baseline stops applying the day you launch. Forecast a dip and a recovery, and read my guide to Shopify migration SEO before you move anything.
• Tiny keyword volumes: in niche B2B, where your best keywords get 50 searches a month, volume estimates are too rough to forecast from. Forecast pipeline and enquiries instead.
• News and trend-driven topics: demand that spikes and disappears doesn't follow a curve you can project.

If someone hands you a traffic forecast for a brand-new domain that's accurate to the nearest hundred visits, they're guessing.

SEO forecasting tools worth using

You don't need a paid forecasting tool to build a good forecast. Most of the work happens in a spreadsheet.

• Google Search Console: free, first-party, and the foundation of every historical forecast. Use the branded queries filter and the monthly view.
• Google Analytics 4: free, and where your conversion rate, AOV and revenue come from.
• Google Sheets or Excel: FORECAST, TREND and FORECAST.ETS cover most of what you need.
• Ahrefs, Semrush or SE Ranking: third-party search volume, keyword difficulty, SERP features and competitor traffic for the keyword-based method.
• SEOmonitor: built for agencies that forecast as part of a pitch, with scenario modelling built in.
• Prophet in Google Colab: free, and the best option for long histories with strong seasonality.

The tool matters far less than the assumptions you feed it.

Final Thoughts

SEO forecasting isn't about being right.

It's about being honest about how wrong you might be, and showing your working so everyone can see why.

Build a baseline from your own Search Console data. Add a keyword-based forecast for the work you're planning. Discount the keywords where an AI Overview sits above you, and present the whole thing as a range.

The forecasts that win budget aren't the ones with the biggest numbers. They're the ones whose assumptions survive the first question from finance.

If you'd like a forecast built on your own data, with the scenarios, the assumptions and a roadmap to match, that's part of my SEO consulting work.

You can also run your numbers through my free SEO ROI calculator in a couple of minutes.

Get in touch, or email me directly.

Author Bio

Graeme Whiles is an independent SEO and AEO consultant at GWContent. He has worked with enterprise and SaaS brands, including Originality.ai, Connecteam, 6sense, Practice Better, and Peppr, growing organic traffic and AI search visibility across some of the most competitive categories in B2B. He also built Three Putt Golf Clothing from a blank domain as a live proof of concept for his methodology.

SEO Forecasting Frequently Asked Questions

What is SEO forecasting?

SEO forecasting is the process of estimating how much organic traffic a website will earn from search over a set period, and what that traffic is worth in leads, sales or revenue. It uses search volume, rankings, click-through rate and historical data, and the result should be presented as a range of scenarios rather than a single number.

How accurate are SEO forecasts?

A historical forecast for an established site with stable traffic is far more accurate than a keyword-based forecast for new content, because it is built on real data rather than estimates. The only way to know your own accuracy is to backtest: forecast a period you already have data for and measure the gap. Make your range at least that wide.

How far ahead should you forecast SEO?

Twelve months, broken down by month, is the most useful horizon. It covers a full year of seasonality and the time most new content needs to reach its ranking potential. Re-forecast every quarter and after every Google core update, rather than defending a forecast built on rankings that no longer exist.

What is the best SEO forecasting method?

Use both main methods together. A historical forecast built from Search Console data gives you a baseline for the traffic you already have. A keyword-based forecast, using search volume, your own CTR curve and a ranking probability, covers new pages. Add the two together, then subtract an allowance for content decay.

Should I include branded traffic in an SEO forecast?

No. Branded searches grow with your brand, PR and advertising, not your SEO work, so leaving them in inflates the forecast and gives SEO credit it hasn't earned. Use the branded queries filter in Search Console to separate them, forecast the non-branded traffic, and report branded traffic as its own line.

How do AI Overviews affect SEO forecasts?

AI Overviews reduce clicks to the organic results below them. Ahrefs found they cut position-one click-through rate by 58% by December 2025, and Seer Interactive's 2026 data puts the gap closer to 40%. Flag keywords that trigger an AI Overview and apply a lower CTR to them in your forecast.

Can you forecast SEO for a new website?

Only roughly. A new website has no history and no rankings, so any forecast relies on competitor benchmarks and third-party estimates. Give a very wide range, or forecast the work itself for the first six months, such as pages published and links earned, then switch to a traffic forecast once you have data.

How do you forecast SEO ROI?

Multiply forecast clicks by your conversion rate, then by average order value for ecommerce, or by close rate and deal value for lead generation. Subtract the cost of the SEO work and divide by that cost. Use your own GA4 and CRM data, and gross margin rather than revenue if finance will review it.