Most stakeholders have seen SEO forecasts that looked polished in a slide deck and never matched what happened. Traffic was overestimated, lead quality was vague, and revenue numbers rested on assumptions nobody could explain.
If you want your forecast to hold up in front of leadership, finance and sales, build it on commercial logic: traffic, then qualified leads, then revenue in £, with every assumption visible.
Start with business outcomes
Rankings matter, but they are not the end goal.
If you lead with ‘we think this keyword can move from position 8 to position 3’, that may sound like progress to an SEO team, but it does not automatically mean much to a stakeholder. What they want to know is what that movement could mean in practice.
A stronger starting point is to ask:
- How much relevant organic traffic could this work generate?
- How many leads could that traffic produce?
- What could those leads be worth in £?
That is a more useful conversation because it puts SEO in the same commercial frame as the rest of your channel mix.
Keep your traffic estimates realistic
The quickest way to lose trust is to forecast as though every important keyword will rank at the top of page 1.
That is not how SEO works. Progress is uneven. Some pages improve faster than expected. Others stall because competition is stronger, intent is mixed, or the page simply is not good enough yet.
A more believable model uses scenarios instead of a single heroic number. In practice, that usually means a conservative case, a base case, and an upside case. This makes your forecast easier to defend because you are showing a range of likely outcomes rather than pretending the future is fixed.
You should also build those estimates around the current state of the site. A brand with strong foundations, useful content, and a healthy technical setup has a very different growth profile from a site that still has crawl, indexation, or page quality problems.
Forecast by page type
A lot of forecasts fall apart because they are built from a keyword list rather than a real website.
That usually leads to inflated numbers and weak commercial logic. A better approach is to forecast by page type, because different types of pages do different jobs.
Your core service pages may attract lower traffic than broader informational content, but they often bring stronger intent. Insight content may drive more visibility and help people discover you earlier in the journey. Supporting hub content may strengthen relevance and improve internal pathways across the site.
When you forecast this way, the model is easier to explain. People can see how each area of the website contributes to the result, and you stop treating all clicks as equal.
Treat lead forecasting with more care than traffic forecasting
Traffic is where forecasts usually start. Leads are where trust is won or lost.
Many SEO models apply one sitewide conversion rate to all projected visits. That may be convenient, but it is rarely accurate. Someone landing on a commercial service page is not behaving the same way as someone reading a top-of-funnel article. If you apply the same conversion rate to both, your revenue model becomes shaky very quickly.
A better approach is to split your assumptions by page type, search intent, or funnel stage. That will not make the model perfect, but it will make it much more defensible.
Measurement matters here too. If your tracking setup is cluttered, inconsistent, or disconnected from the way your business defines a lead, your forecast will inherit that weakness.
Define a quality lead before you forecast leads
A form submission is not a lead. Before you forecast leads, agree with sales what counts as a qualified one, and write it down. A useful definition covers:
- Company size: the size range you can serve profitably
- Sector: industries where you win and keep work, and those where you do not
- Role or seniority: who the enquiry needs to come from, or reach, to become a deal
- Intent stage: ready to commission now, or months away from a decision
- Geography: where you can deliver
Then forecast qualified leads, not form fills. If 40% of last year’s organic enquiries met the definition, apply that rate to the projected enquiries, and use a different rate for each page type if the data supports it.
The minimum tracking you need
- A GA4 conversion event for each form type
- Consistent source tracking, including UTM parameters on links you control
- A CRM field that records the original marketing source of each lead
- Lead status (qualified, unqualified, won, lost) recorded in the CRM so it can be matched back to organic landing pages
With that in place, you can see which pages and topics produce qualified enquiries rather than volume. At low volumes, a monthly conversation with sales about which enquiries were worth pursuing tells you more than the numbers.
Revenue in £ should be simple enough to explain in one meeting
You do not need an overly complicated financial model to create a useful SEO forecast.
In most cases, a practical revenue estimate can be built from four inputs:
Estimated leads × qualification rate × lead-to-sale rate × average revenue per sale
The qualification rate is where your lead quality definition earns its place. Without it, the model counts every enquiry as a potential sale.
A worked example (illustrative figures)
- Base case: 1,200 additional organic visits a month to service pages
- Enquiry rate on service pages: 2%, giving 24 enquiries
- Qualification rate: 50%, giving 12 qualified leads
- Lead-to-sale rate: 25%, giving 3 sales
- Average revenue per sale: £8,000, giving £24,000 a month
Each line is a number someone in the room can challenge, and each comes from your own data. Replace the figures with yours and run the same sum for the conservative and upside cases.
Stakeholders can follow it straight away. They can challenge the close rate, question the average revenue figure, or ask where the lead assumptions came from. Good. Open logic makes a forecast more credible.
If you work in a longer sales cycle, explain that SEO often influences the pipeline before it shows up as closed revenue. That is especially true in B2B and higher consideration services; our guide to B2B SEO for long sales cycles covers how to measure that influence.
Be clear about the assumptions behind the forecast
Visible assumptions earn more trust than big numbers.
That means spelling out what sits underneath the model:
- Your current organic baseline
- The expected uplift in visibility
- The click-through rate range you are using
- The lead conversion assumption
- The qualification rate
- The lead-to-sale rate
- The average revenue per sale in £
- The timeframe for impact
Once those assumptions are visible, the forecast becomes easier to challenge and easier to improve. It also opens the door to better planning, because you can test what happens if conversion rate improves before traffic does, or if authority growth is slower than expected.
Compare the forecast with your competitors’ visibility for the same page types and with what you already pay for those clicks in paid search. You are then looking at search opportunity through a commercial lens rather than in isolation.
Adjust for AI answers and zero click results
Click through rates are not stable. Google AI Overviews and other answer features now sit above the links on many informational queries, and studies from SEO tool vendors report lower click through rates where an AI Overview appears. The size of the effect varies by study and query type, so do not import one figure into your model.
Work from your own data instead:
- Mark which forecast queries currently show an AI Overview, by checking the results pages or using a rank tracker that records search features
- Apply a lower click through rate range to those queries in the conservative case
- Check Search Console for queries where impressions rise while clicks fall, which is a common sign
- Treat inclusion in AI answers as a separate outcome to track, because it shows up in branded search and direct enquiries rather than in organic clicks
Our guide to generative engine optimisation covers how to track whether AI answers include your brand.
Keep the forecast as a working model
Treat the forecast as a model you revisit.
Once activity starts, compare forecasts against actual performance. Update the assumptions. Tighten the conversion model. Rebalance expectations where necessary. That keeps it useful as well as persuasive.
Every number in it should be explainable without jargon.
If you want a second view on a forecast before it goes to leadership, get in touch with Totally.Digital.