SEO Forecasting: Predicting Traffic and Revenue

Short answer

An SEO forecast estimates future organic traffic and revenue from three inputs: your own history (level, trend and seasonality), the planned work and the opportunity it targets, and your conversion data. It's useful for planning only when it's presented as a range with its assumptions written next to it, because the largest drivers of organic performance, such as Google's updates, competitors and demand, are outside your control. A forecast that gives one number without assumptions can't be checked, and one that can't be checked can't be improved.

Sources

Ask next

Starting From Your History

Collect monthly organic data from GA4 (sessions, key events, revenue) and Search Console (clicks, impressions), for as many full years as you have. Search Console keeps data for the last 16 months, per Google’s Analytics integration help, so save exports if you want a longer series.
Mark the breaks: migrations, redesigns, tracking changes, consent banner changes, Google updates, and any change in how Google serves or counts results. A break in measurement isn’t a trend, and a projection built across it carries the break forward.
Read seasonality by comparing the same months across years. Google Trends can help for demand outside your own data; its data FAQ explains that values are “scaled on a range of 0 to 100”, so Trends shows relative interest, not search counts.
Separate branded and non-branded queries. Brand demand follows your marketing and reputation; non-branded traffic is where to measure the effect of SEO work.

Methods

Method	How it works	When it fits
Trend projection	Last year's months × your measured trend	Stable sites with no planned changes
Opportunity model	Keyword demand × an assumed click rate at a target position	New pages or sections
Scenario model	Low, expected and high versions of the inputs	Forecasts you present to others

Trend projection assumes conditions stay as they were. Use your own year-over-year change, measured on months without breaks.

Opportunity models build up from keywords. Google’s Keyword Planner gives search volume estimates, and its help page warns that “campaign performance depends on a variety of factors” even for ads. The click rate you assume for a target position is your own assumption; write it down, and test it against your Search Console CTR for pages already at similar positions.

Scenarios turn both into ranges. Vary the inputs you’re least sure of (trend, new-page traffic, conversion rate) and keep the rest fixed.

A Worked Example

For example, imagine a store whose organic sessions last March were 12,000. Its measured non-branded trend has ranged from flat to +15% a year, it plans new category pages, and organic sessions convert at 2% with an $80 average order.

Scenario	Trend	Sessions from trend	New pages	Total sessions	Orders at 2%	Revenue at $80
Low	0%	12,000	0	12,000	240	$19,200
Expected	+8%	12,960	600	13,560	271	$21,696
High	+15%	13,800	1,200	15,000	300	$24,000

The forecast for next March is 12,000 to 15,000 organic sessions and $19,200 to $24,000 in revenue, with the expected case near the middle. The written assumptions are the trend range, the traffic from new pages and a fixed conversion rate; if any of them changes, the forecast is recalculated rather than defended. Expected orders are 271.2 before rounding, and the revenue figure uses the unrounded number.

From Traffic to Revenue

Use your own conversion rates by segment where you have them: commercial landing pages convert differently from informational ones.
Value conversions by average order value for ecommerce, or by lead-to-customer rate × customer value for lead generation.
Name the attribution model behind the conversion data, since the same traffic produces different organic revenue under different models; see attribution modeling for SEO.
Use profit when the forecast supports a budget decision, as in SEO ROI calculation.

Risks to Write Into the Forecast

Google updates. Core updates and spam updates can shift rankings. The helpful content system is no longer a separate update to watch: Google’s ranking systems guide says “In March 2024, it evolved and became part of our core ranking systems”. How to read an update’s effect is in core updates.
Results-page changes. AI Overviews and other features can change how many searchers click for the same rankings; history from before a feature appeared doesn’t show its effect.
Competitors entering or investing in your keywords.
Demand shifts from the economy, your category or your own pricing.
Ceilings. A page already at the top of the results has little ranking left to gain; growth there comes from new queries, not from projecting the same rate forward.

Monitoring for these between forecasts is covered in SERP volatility.

Checking and Improving Forecasts

Compare forecast with actual each month and record the gap with its likely cause: an update, a launch that slipped, a tracking change.
Look for a consistent bias. If actuals land below the expected case month after month, an assumption is too optimistic; find which one.
Keep horizons short where you can. Uncertainty grows with the horizon, so show wider ranges for later months.
Refresh the forecast when an input changes, and keep old versions for comparison.

Presenting It

Show the range, the expected case, the assumptions, the main risks and the work the forecast depends on. A sentence such as “12,000 to 15,000 sessions next March if the category pages launch by January” ties the number to a decision. Budget conversations that follow are covered in SEO budgeting.

When actuals land below the expected case month after month, stop explaining each month separately: test the three written assumptions in turn, the trend, the traffic from new pages and the conversion rate, replace the one that’s off with what the data now shows, and recalculate the range.