Organic Search Forecasting: A Simple Model

Organic search forecasting estimates how much traffic, how many signups and how much revenue a set of pages will produce if they rank. The useful version for a startup is simple: search volume, times the click-through rate you can expect at your target position, times a ramp for how long ranking takes, times your conversion rates. It will be wrong. It's still the best way to decide which pages are worth building and to set expectations before you spend.

Below is the model, a worked example, how to turn it into an SEO ROI forecast, and the mistakes that make forecasts useless.

Why forecast organic search at all

You forecast so you can make decisions, not to predict the future. A forecast tells you which keyword clusters are worth building first, how long until they pay back, and whether the budget makes sense compared with paid search or other channels.

It also protects the program. SEO is slow, and a founder who expected signups in month two will cut the budget in month four. A forecast with a realistic ramp makes the slow start part of the plan.

The organic search forecasting model

Forecasting organic traffic works page by page, or cluster by cluster, never for the whole site in one number. For each page or cluster:

1. Monthly search volume for the target keyword and the close variants the page will also rank for.
2. Target position, and the click-through rate you can expect there.
3. A ramp: the share of that traffic you'll capture each month as the page climbs.
4. Conversion rate from visit to signup or demo request.
5. Rate from signup to paying customer.
6. Value of a customer, ideally first-year revenue so the numbers stay conservative.

Multiply across and you have a forecast for traffic, signups, customers and revenue by month.

Where the inputs come from

Search volume comes from tools like Semrush or Ahrefs. Their volumes are modeled estimates, so treat them as ranges, not counts.

For click-through rate, use your own data. Google Search Console's performance report shows clicks, impressions and average position for every query, so you can build your own CTR curve by position instead of borrowing an industry average that doesn't match your results pages.

For the ramp, be patient. Ahrefs found that only 1.74% of 1 million new pages from its September 2023 crawl reached Google's top 10 within a year. Low-difficulty keywords move faster, but a model that assumes full traffic in month three is fiction.

Conversion rates come from your analytics: organic visit to signup on similar pages, and signup to paid from your product data.

A worked example

Here's a hypothetical cluster for a SaaS startup, to show the arithmetic. Every input is an assumption you'd replace with your own.

Year one comes to about 2,700 visits, 81 signups, 8 paying customers and roughly $9,700 in first-year revenue. Year two, at full ramp, comes to 7,200 visits, 216 signups, about 22 customers and roughly $25,900.

Notice the shape. Most of the value arrives after the first year, which is why SEO budgets get cut by people who didn't see a forecast.

How to forecast organic traffic in three scenarios

Don't publish one number. Forecast organic traffic as a range with three scenarios:

1. Conservative: lower volume, a lower position and a slower ramp.
2. Expected: your best estimate for each input.
3. Aggressive: the pages rank higher and faster than planned.

Present the expected case and show the range around it. If the plan only works in the aggressive case, it doesn't work.

Turning it into an SEO ROI forecast

An SEO ROI forecast puts cost next to the revenue line. Add up what the pages will cost to research, write, design and maintain, plus the tools and any outside help. Then compare cumulative cost with cumulative revenue by month.

The output you care about is payback: the month when cumulative revenue passes cumulative cost. In the example above, a hypothetical $15,000 investment wouldn't pay back in year one, and would pay back well inside year two. That's normal for SEO, and it's the conversation to have before you start, not after.

To forecast SEO ROI across a whole program, run the model for each cluster, then rank the clusters by payback. Build the fastest-payback clusters first.

Mistakes that break a forecast

These are the errors I see most often:

1. Forecasting the whole site from total keyword volume instead of page by page.
2. Assuming the page reaches position one. Plan for the position you can realistically win.
3. Using an industry CTR curve instead of your own Search Console data.
4. Ignoring AI answers. When an AI summary appears in Google, people click results less often: Pew Research Center found clicks on 8% of searches with an AI summary versus 15% without one. Discount informational keywords accordingly.
5. Counting branded search. Branded traffic grows with everything else you do, so keep it out of an SEO forecast.
6. Never checking the forecast against actuals.

Check the forecast every month

Compare the forecast with what happened each month: impressions, position, clicks, signups and customers for each cluster. When a cluster runs ahead, build more like it. When it lags, find which input was wrong, whether volume, position, CTR or conversion, and fix the page or the plan.

Forecasting and measurement are the same discipline. My post on marketing effectiveness covers comparing channels once the actuals come in, and the forecast itself belongs inside the plan from my guide to SEO for SaaS startups.

Make organic search forecasting part of planning

Organic search forecasting is worth doing even though it's always wrong, because it forces you to state your assumptions and then check them. Forecast by cluster, use your own data, plan for a slow ramp, and judge the program on payback.

If you want a forecast built for your site, it's part of how I scope my SEO and GEO services.

SEO Forecasting FAQs

How accurate is organic search forecasting?

Organic search forecasting is rarely accurate to the number, and it isn't meant to be. Search volumes are estimates, rankings are uncertain and click-through rates change as Google adds AI answers. A good forecast gets the range and the order of priorities right, and it improves as you replace assumptions with your own data.

What data do you need to forecast organic traffic?

To forecast organic traffic, you need search volume for each target keyword, your expected position, a click-through rate for that position, and a ramp for how long ranking takes. Use Search Console for your own CTR by position. To go further and forecast signups and revenue, add conversion rates and customer value from your analytics and product data.

How do you calculate SEO ROI?

You calculate SEO ROI by comparing the revenue organic search produces with what it costs over the same period, including content, tools and outside help. Because SEO compounds, measure ROI over 12 to 24 months and track payback, the month cumulative revenue passes cumulative cost, rather than judging it on a single month.

How long until an SEO forecast shows results?

An SEO forecast usually shows little in the first three months, because new pages take time to rank. Low-difficulty pages often start ranking within a few months, and most of the value arrives in the second year. Build that ramp into the forecast so the slow start doesn't look like failure.