SEO forecasting is a process that uses historical, present and estimated future data to build a return on investment scenario of SEO results. For optimal results we should not just use historical data, but also include potential traffic, keywords and leads data that a business could get if and when an SEO campaign is implemented – this is because complete data can generate better forecasts than partial data.
This above is a typical SEO forecast or traffic and leads estimator over time. It shows the cumulative increase of all SEO optimisations starting now this moment in time and projected over 12 months.
How do we calculate future performance? If you want to calculate future performance, you should spend some time in getting the correct data which looks like this:
- Historical data:
- Traffic data
- Keywords performance
- clicks to lead conversion rate
- lead to close rate
- average order value
- total revenue
- Future data
- Competitor keywords you don’t yet rank for
- Search volume and CTR by position
- Estimated traffic from future rankings and clicks
- Estimated leads and revenue
Historical data tells you where the business has been. Future data tells you where it can realistically go. Most forecasts fail because they lean entirely on one or the other: either they extrapolate past traffic in a straight line and ignore the untapped market, or they promise a keyword universe the site has no authority to win. A defensible forecast sits between the two.
Once you have the data, the calculation itself is simple. It’s a chain, and every link has to be a real number, not a hopeful one:
Clicks → Leads → Customers → Revenue.
You take the projected clicks for a given month, apply the visit-to-lead conversion rate, then the lead-to-close rate, then the average order value. Multiply them together and you have the monthly revenue that level of traffic is worth:
Monthly revenue = clicks × 30 × visit-to-lead rate × close rate × average order value
So if an SEO campaign is projected to grow organic traffic from 50 to 300 clicks a day, and the business converts 3% of visits into enquiries, closes 30% of those, and earns £150 per job, month-by-month that traffic growth translates directly into a revenue figure the client can put next to your invoice.
That comparison – projected revenue against cost of the campaign – is the entire point of a forecast. It turns “SEO will help you rank better” into “this investment returns X.”
Why forecasts get rejected (and how to avoid it)
Here’s the uncomfortable truth: most SEO forecasts don’t get rejected because the numbers are too low. They get rejected because the person approving the budget doesn’t believe them.
I do a lot of SEO forecasts during my sales calls and I have increase the number of clients after I was able to show them how their SEO budget is spent, and roughly how long it takes to hit their targets.
The forecast is a subtle art of reassuring prospect clients (or a boss or a CMO) that the budget is well spent. How they say “an image is worth a thousand words”.
Budget approvals are a trust problem, not a maths problem, and it’s where the “subtle art” comes in. A forecast that promises a 10x return in three months gets filed under “too good to be true” and quietly killed.
A forecast built on the client’s own historical data, with conservative conversion assumptions and a timeline that admits SEO takes time, gets approved – even when the final number is smaller – because the person signing it off can defend it to their boss.
That being said, I must absolutely make things clear: an SEO forecast will NEVER be 100% realistic, it shouldn’t be taken word by word, which means the numbers will never be fully credible or defendible. While a forecast can get close to reality, it doesn’t eliminate risks or external factors that are outside of our control. But the general consensus is, forecasts built with historical and future data are more closely aligned with real SEO results than forecast based on historical data only.
Understand what you are up against
Google controls the rankings. The client’s dev team controls when fixes ship. Competitors control how hard they fight for the same terms.
Your forecast is a bet placed across all three, and pretending otherwise is how consultants end up defending numbers they can’t hit.
So the first job isn’t calculating, it’s understanding the forces working against a clean prediction. There are a few worth naming.
Google is a moving target. Algorithm updates, SERP layout changes, AI overviews eating clicks from the top of the page – the click-through rate you model today may not hold in six months. The position-one CTR that studies quote at 20-something percent is an average across industries and intents; your client’s actual curve could be very different, especially if the SERP for their terms is crowded with ads, map packs, and featured snippets that push organic results down the page. A forecast that assumes textbook CTR on a SERP full of distractions is already optimistic before you’ve done any other maths.
Rankings don’t move on your timeline. You can fix every technical issue, publish every page, and earn every link, and Google will still take its time to recrawl, reassess, and re-rank. This lag is the single biggest reason time-based forecasts get consultants in trouble. You promise month three; Google delivers month five; the client remembers the promise, not the caveat. This is why anchoring a forecast to ranking states (“when these keywords reach page one”) is safer than anchoring it to dates – you’re forecasting the thing your work influences, not the thing Google decides.
The data you start with is often broken. Search volumes are estimates, not counts. Conversion rates pulled from analytics are muddied by attribution — a lead that touched organic, paid, and email gets credited unevenly depending on the model. If a site has recently lost rankings or pages, its current numbers may reflect a damaged state, not a normal one, and forecasting off a broken baseline compounds the error. Garbage in, confident garbage out. Part of building a forecast is auditing whether the inputs are even trustworthy before you multiply them together.
Competition is not standing still. Your forecast implicitly assumes the SERP stays roughly as it is while your client climbs it. But competitors are optimising too. Winning a competitive commercial keyword isn’t a matter of doing the work once; it’s out-investing rivals who want the same position. The more valuable the term, the harder it’s defended — which is exactly why the high-volume head terms in any forecast are the least certain, not the most.
The person approving it is sceptical by default. Whoever signs off your budget has almost certainly seen an SEO forecast that never materialised. That scar tissue is working against you before you say a word. You’re overcoming a reasonable prior that forecasts are marketing fiction. This is why credibility beats ambition every time: a smaller number your client believes is worth more than a bigger number they don’t.
None of this means forecasting is pointless: it means the goal isn’t precision, it’s defensible direction. A good forecast doesn’t claim to predict the future exactly.
It says: based on real data, realistic assumptions, and an honest read of the obstacles, here’s the return this investment is likely to generate, and here’s the range around it.
Understand the forces stacked against you, build those constraints into the model, and you produce something that survives scrutiny — which, in the end, is the only kind of forecast worth presenting.
So the real question is: how do you keep a forecast credible?
How do you keep a forecast credible?
Anchor everything to real data
If the client already ranks for a set of keywords, use their actual click-through and conversion rates as your baseline. A forecast that starts from measured reality is far harder to argue with than one built on industry averages.
Separate what’s certain from what’s ambitious
Recovering traffic a site used to have is a high-confidence bet. Winning competitive head terms it has never ranked for is a growth play that depends on links and authority.
Show them as different layers, and never present the ambitious layer as a guarantee.
Be honest about timing
SEO compounds; it doesn’t switch on. A forecast that shows a gradual ramp over 12 months is more believable — and more accurate — than one that hits the target in month one. Tie your timeline to the things you actually control, like when the technical fixes ship.
Present a range, not a single number
Give a conservative, likely and optimistic scenario. It signals that you understand forecasting is probabilistic, and it lets the client choose their own comfort level rather than feeling sold to. Present moving targets for what they are, without trying to fix a future SEO return on investment “because you say so”.
Not even the sentence “based on my experience” can help in this case, unless you have several SEO case studies from the same industry, the same country and many more similarities that give you enough confidence to bring your point forward to the client.
So the best case scenario is to present those moving targets, and in this picture below you can see that I intentionally built radio buttons to change the number of clicks generated per month, on a given month.
We can see that clicks go from the initial baseline of 50/day to a very optimistic target (or better called estimate) of 800 clicks a day.
So the real question is moved away from the expected “can we really reach 800 clicks/ day?” This would be very hard to promise and hard to answer and we don’t like hard things, especially during sales calls and budget meetings.
What we are trying to do is to move the discussion to a more manageable field, which is the one of trust, honesty and realistic timelines. We can say things like “in the best case scenario the traffic prediction is 800 clicks a day”.
In this case, you can see the revenue numbers moving up from £17,000/month SEO ROI to £32,400/month, IF, and only IF, things wil go as planned.
The key thing to take away from this is that we should decide not only what the numbers look like based on our predictions and calculations, but also what type of discussions we want to have. A hard, not credible discussion based on one specific future prediction? Or a more realistic and honest view of how the SEO return will look like, given a certain number of variables we cannot control?
Nobody can dispute the fact that uncontrolled variables are changing the future results, so we just need to mention them and plan accordingly.
Uncontrolled, external variables that impact SEO forecasts
We can’t know what the future looks like, and again, this is the general consensus.
Going back to our uncontrolled, external variables, they are already factored in the forecast model above and clients can see the ups and downs that these variables create. You play around with the numbers and explain to them that these are some of the factors influencing our results:
- Google updates can increase or decrease your traffic
- Seasonality
- Competitors’ SEO efforts
- Web developers changing/ updating the website
- Migration / redirection projects
- Human errors
Here is an example of a positive seasonality impact factored into the SEO forecast.
On the other end, if your business suffers from negative seasonality (for example, Christmas or August when normally people do on holiday), you can see the opposite trend with a massive drop of traffic, clicks and revenue:
CMOs and Bosses are playing the devil’s advocate
Bosses and clients like to break down our most positive predictions because they know their business inside out and don’t want to spend on SEO what they normally spend on PPC. SEO builds a permanent, compounding traffic asset through content and site optimisation. Traffic is free, but initial results take £1,500–5,000/month and 3 to 6 months to materialize. PPC pays for immediate, targeted visibility. Costs range from £100–20,000/month, and traffic stops immediately when payments cease.
In other words, PPC is perceived as a necessary evil, while SEO is a nice to have. When things go wrong, traffic drops, inquiries don’t generate, usually a company tends to spend more budget on PPC rather than SEO.
I recently had a call with a company with a massive organic traffic drop that is spending £20,000/month in PPC and it’s surprised that they SEO is bad, after all they are spending £2,000/month.
This enormous difference is explained by the fact that PPC forecasts are easy to make, Google Ads system make them for you efore you even spend a dime, while SEO consultants don’t have an equivalent tool for doing exactly the same forecast with organic traffic.
This image above conveys more trust than any SEO forecast because you spend X and generate Y, you can’t go wrong.
There are no external factors involved, nothing that can bother your predictions, other than the skills and experience of your Google Ads consultant.
All of this for saying that SEO forecasts have a twofold challenge:
- Bosses, clients and CMOs often compare Google Ads ROI vs SEO ROI.
- Given this comparison, they want to be guaranteed the same results
Both of those are legitimate business principles: my marketing budget should work as hard as possible. If I can choose a safe way of generating clicks and leads, I am sold Google Ads over SEO every time.
This above is an example of a company increasing their PPC budget by 93% MoM because the SEO traffic is decreasing by 18% from the previous month. This means that they believe the SEO forecast doesn’t look so good, given past failures.
SEO Forecasting: how to do it right and get the budget signed off
By now the pattern should be clear. Getting a forecast approved has very little to do with the size of the number at the end and almost everything to do with whether the person holding the budget trusts how you got there. So here is the method I use on sales calls, built from everything above – the honest way to forecast, and the way that actually gets signed off.
Start with their data, in front of them
Don’t arrive with a finished PDF full of numbers you generated alone.
Open the client’s own figures live — their current traffic, their ranking keywords, their conversion rate, their average order value — and build the forecast in the room. The moment the baseline is their reality rather than your industry averages, the conversation stops being “do I believe this consultant” and becomes “do I believe my own numbers.” That’s a far easier sell, and it’s why I build the forecast during the call rather than before it.
Use my own forecasting tool to get familiar with the concept
Show the mechanism, not just the outcome
A boss who sees “£16,000/month” with no visible working assumes marketing spin. A boss who watches you go clicks → leads → customers → revenue, one honest multiplication at a time, sees a model they can interrogate.
Let them push on the conversion rate. Let them argue the close rate.
When they can poke the assumptions and the number moves accordingly, the forecast stops being a claim and becomes a shared calculation – and people defend calculations they helped build.
Separate the safe money from the ambitious money
Recovering traffic a site used to have is a high-confidence bet; winning competitive head terms it has never ranked for is a growth play that depends on links and authority.
Present them as distinct layers and label them honestly.
This does two things: it protects you when the ambitious layer takes longer, and it gives the client a choice rather than a single take-it-or-leave-it figure. Cautious buyers approve the safe forecast and treat the rest as upside; ambitious ones fund the whole thing. Either way, you’ve made it approvable.
Forecast ranges and ranking states, never a single dated promise
This is the heart of it. Instead of “you’ll hit 300 clicks a day by month three” – a promise Google, not you, controls – give a conservative, likely and optimistic band, and tie the milestones to ranking states you influence rather than dates you don’t.
The radio-button model I use does exactly this: the client can drag the target from a modest baseline to an optimistic 800 clicks a day and watch the revenue move from £17,000 to £32,400 a month. The question is no longer the un-answerable “can you guarantee 800 clicks?” It’s the manageable “here’s the best case, here’s the likely case, here’s what has to be true for each.” You’ve moved the discussion off certainty – which you can’t offer – and onto honesty and realistic timelines, which you can.
Name the variables you don’t control, before they do
Google updates, seasonality, competitor investment, a dev team that breaks things, a botched migration. Don’t hide these; build them into the model and show the client the ups and downs they cause. Walk them through a seasonality dip so they see the line drop and understand why. Counter-intuitively, naming everything that could go wrong makes you more credible, not less: it proves you’ve thought like a sceptic, so the client doesn’t have to.
A forecast that admits its own uncertainty is the one a CFO trusts.
Meet the PPC comparison head-on
Understand what you’re truly competing against in that room: Google Ads’ budget planner, which promises “spend X, get Y” with no external factors and no waiting. You will never match that certainty, so don’t pretend to. Instead, reframe the trade. PPC rents visibility — the traffic stops the day the payments stop.
SEO builds a compounding, permanent asset, slower and less certain up front, but it keeps returning long after the invoice is paid. When a client is spending £20,000 a month on PPC and £2,000 on SEO and wondering why the organic results lag, that gap is the answer, and the forecast is where you make it visible. You’re not selling a safer bet than PPC; you’re selling a fundamentally different asset, and the forecast is what lets a sceptical buyer see its shape.
Do all of this and the number at the end almost doesn’t matter — because you’ll have built the one thing that actually gets budgets approved: a forecast the person signing it can defend to their boss. That’s the whole subtle art. Not the biggest projection you can justify, but the most honest one they can stand behind.
What not to do when building an SEO forecast
Don’t fake or overinflate the numbers
Tuning inputs until the total says £10k is the fastest way to lose the room. Buyers can smell a number built backwards — let it land where it lands.
Don’t include informational keywords
AI Overviews are eating the clicks on “what is” and “why” style terms. Counting their volume as traffic you’ll capture inflates the forecast on searches that no longer convert.
Don’t forecast on historical data alone
Past traffic tells you where the business has been, not where it can go. Add the future layer — gap keywords, competitor terms, untapped demand — or you’re modelling half the picture.
Don’t drift too conservative or too optimistic
Lowball it and the ROI looks pointless; oversell it and nobody believes you. Credibility lives in the honest middle — the range you can actually defend.
Don’t let commercial goals write the forecast
When a client or boss needs it to say a certain number, the forecast stops being a forecast. Model what the data supports, then have the harder conversation – that’s what protects your name when the results come in.
Conclusion
A good SEO forecast is about:
- Building in front of the prospect, boss or CMO
- Show the mechanism on how and why we get to the ROI, not just the final result
- Preparing for objections and create defensible options
- Presenting a range of possinble results, not a single number
- Naming the varialbes you don’t control
- Your main competitor is PPC Ads and forecast, prepare arguments to defend SEO