Back to the Campfire Blog

Annual planning in SEO

Written by: Adrienne Kmetz Tags: analytics, template, biz of seo

Published: Sep 30, 2025  |  Last Updated: Apr 14, 2026

There are three main ways to develop a forecast or an annual plan in SEO and publishing departments:

• Traffic potential based on target keywords,
• Working backwards from a target and breaking it down month by month,
• Or to use historical trends to infer a trajectory for the future. 

Today we're covering #3 – how to develop a forecast based on the momentum you already have. 

TLDR: 

• A forecast is an extrapolation of a historical pattern, not a prediction, with only "committed and funded" projects having an impact (eg. an app launch). 
• Keep it simple – more granularity means more variables that must all occur in order to hit your goals. 
• Finance usually needs to separate paid media from payroll and project-based investment like agencies. For shorthand, some call this paid and nonpaid.
• Planning is simply a guess, with an investment it's a bet. So a forecasting model is about enabling smart decisions throughout the year, not perfect accuracy. 

Develop your strategy

You are here: Find the current place in the roadmap

Where are you in the context of the business plan? Are you on track to hit your current OKRs or are you significantly ahead or behind? This will help you determine whether you need to make bigger changes now, or if you can focus on what's coming next. 

If your company has an existing strategic plan, roadmap, 3-5 year plan, then you likely already know what long-term investments are being made into revenue driving activities like getting more members, sales, or partners.

A distinct north star (new product launch, platform upgrade, whatever usually development-led investment that is going to change the business in the next FY) helps you align your activities to this goal. 

As a project manager you're likely already very familiar with what needs to be prioritized in the next FY, where the biggest risks and opportunities lie. If not, determine what the necessary audits are in terms of technical, PR/links, and content, so you can determine where to go next. 

What are we trying to accomplish with forecasting?

Our current trajectory, plus changes based on the strategy: Potential increases in organic traffic, conversion rate changes due to investments in CRO and the product, and ROI impact of SEO investments. 

It's just as much about painting a realistic picture that includes historical context, as much as it is about getting folks excited about a future where strategic investments in SEO can lead to compounding results. 

The Secret CFO tells it straight up: 

Typically, a simple model works for 99% of businesses. It should be the thinking that is intricate, not the math. And if your dimension count is too big to be managed in Excel, you probably got Phase 1 (define parameters) wrong.

Your job is to facilitate choices and decisions, not produce the most accurate model you can.

The key functionality you need is to run scenarios and add or kill various building blocks or organizational units (this will be important when testing scenarios).

Plan for volatility, then do what's right for your customer anyway

I asked Tory Gray of Gray Dot Company what the key is to getting 2026 planning right. She encourages us not to get distracted by shiny objects or too many variables: 

"Annual marketing planning today should lean on ranges, not single-number forecasts. Fan charts are especially useful now given the volatility AI is introducing into search and referral traffic.

Citation policies, for example, directly affect discoverability: When platforms reduce outbound links, brands lose referral traffic they once relied on. Similarly, shifts in how often Reddit is cited (in Google & AI/LLM results) can alter both its role in search visibility/traffic and the ROI of direct marketing efforts in the Reddit community.

The most resilient plans balance agility with a focus on customer-centered strategies that remain valuable regardless of algorithmic change.

In other words: plan for volatility, and keep your eyes on your customers – instead of AI shiny objects.”

Forecasting process using historical data for organic 

1. Gather your high level actuals for the last FY

Keep this simple: Total organic traffic every month over 12 months, and total leads. Divide these to learn your actual conversion rate of nonpaid traffic. 

Include all referral traffic here too, especially if its fairly consistent – leave out social and email only if the outliers are too high. This is a picture of all "nonpaid" traffic.

You likely won't be able to fill in the last month until it's over, but you can use an estimate until that time. 

2. Repeat this for paid 

The rows should look similar, except paid will have a row for total cost that hopefully your paid person is able to fill in. 

Add another row for the average cost-per-lead associated with it in addition to CR. Similar to CR, average CPL is the total cost divided by the total leads of all paid media activity for the month. 

but why? In organizations where paid is a major investment, organic is the profit margin on top of a very narrow margin in ads. Showing the two datasets together enables finance to see total traffic and how much it costs. 

3. Determine your historical average MoM growth rate

Insert a row with a formula to show % growth rate month-over-month (MoM). Average this number to determine your annual MoM growth rate.  

Subtotal all rows. Label this entire section "Actuals - 2025 FY" 

5. Create a forecast section

Duplicate the entire Actuals section, and edit the label to Forecast - 2026 FY. Take the last month's data, apply the growth rate, and insert it into the first month of 2026. 

Apply the growth rate MoM until the end of the year.

Repeat for paid. 

Apply subtotals, as well as a total column to see traffic, leads, and costs for the year. 

Now, you can play with the growth coefficient based on what you know about your company's investments, industry, and roadmap. 

That's really it. Come up with something that seems realistic, because your leadership will inevitably want to grow it 10-15% or more before they show it to the board. 

Get a free organic forecasting template

Use this template and pop in your actuals. Please also let me know if you notice that any of the formulas are off. All of the numbers are fake, so they won't really make sense until you replace the Actuals with your data. 

Make a copy of my free forecasting template in Google Sheets.

When researching, I also found these resources to be extremely helpful. Make sure you read a wide range of information and take inventory of known assumptions so that you end up with something useful now and well into the future. 

Ramona Joita's SEO forecasting template blends investment level scenarios with growth rates. Ramona says "Despite these challenges, you can build reliable SEO forecasts—by using your own historical data." 

Maeva Cifuentes' detailed look into three ways to build an SEO forecast is extremely easy to understand. She details how many external forces there are making it harder and harder to predict: "The search landscape has transformed dramatically, introducing variables that make traditional forecasting models increasingly unreliable."

Try to leave out seasonality from this exercise

Don't add seasonality unless it is extremely predictable YoY – like the impact of Black Friday.

And even then, remember that with accrual or cash accounting, planning for spikes that are realized in a different month makes things a bit messy. When Black Friday fell on the last day of November, we had to continually contextualize why we had a very low November compared to the goal, and a very high December compared to the goal – we just were off by a few days. 

Meanwhile, if we just said, we know our target for the whole quarter, we understand some months may be above or below the average, then we would have ended up at the end of the quarter all on the same page, saving this energy for better things. 

Leave out rankings increases too

There's too much variation in the SERPs to be able to truly rely on the accuracy of a "rankings increase" sheet. Has anyone ever said to you, "thank gawd I did that rankings sheet, otherwise..." what?

Total capturable market size should be a relatively quick and rough exercise, and the KW data will change by the time you start the project. I am totally open to new information on this; please share anything I'm missing because it seems like a lot of work for something that should be a SaaS tool add-on by now. 

The main reason I don't use one however is because 12 months is too long to be this specific. Something is going to prompt a change in course throughout the year. It begs too many questions: If you pivot to new words, do you reforecast? Why are or aren't we targeting competitor XYZ?

Your site is in a continuous state of improvement, expansion, and contraction – therefore if the minutiae of the details change, the results shouldn't. 

That said, on a project-by-project basis, I create SEO scenarios that reflect the rough outcomes from a representative sample of what we'll be working on. 

David Heinemeier Hannsen of 37 Signals (the folks behind Basecamp) discusses with cofounder Jason Fried on their podcast ReWork, about how deadlines are arbitrary and self-inflicted pain: 

"You sit down and make these projections up. These are your guesses for the future. And you know what, not only are those guesses, not plans, they’re also subject to change. You can change them. You can simply guess something else. You can come up with something else, and just revise it in a bunch of different directions."

One of the main principles when it comes to OKRs is that the initiatives (projects and their tasks) should change if needed, to hit the key results. It's up to you to recognize when something needs to pause and adjust a few degrees, in order to reach the destination. 

Language lately that has come out of product is placing a "bet". EG: We're betting that the app launch is a better investment than whatever the opportunity cost is of doing something else. The reason why a bet feels more powerful than a guess, is because there's money behind it. 

Piecing it together

Your company's north star + your strategy + your forecast + your known budget and resource constraints = Your pitch

Turn that forecast into a budget

Based on a unit of measurement like "2 pieces per day per staff writer", you can measure what your team can produce in a 13 week quarter. This can back out into Full-Time Equivalent (FTE) needs, which quantifies how much help you need and how much it will cost, to get the projects done, to achieve the forecast. 

The pitch

I've been in situations where the forecast was accepted with, "this looks great – thanks!" and I've also been in situations where you need to present your strategy and budget and hiring plan in a deck. 

Tom Critchlow has put together a tidy list of guides on SEO pitching to leadership that covers this area quite well. 

Andy Strager has been working on adjusting his presentation style to commercial leaders over the past year:

"I am trying to speak marketing to marketers. Before, I'd way overcomplicate things with heavily technical slides that showed our process. Now, I focus on showing where we are working to increase brand visibility on money pages.

I created a super simple correlation graph with GSC impressions and Shopify revenue reporting, dumped into a Google sheet and combined in Looker. This way it has been much easier to talk to non-SEOs about the impact we are creating." 

Here's an anonymized version of that graph so you can see how simple it can be: 

Ongoing data monitoring and improvement

After three years of doing this for one company, you will have 3 tabs of historicals vs actuals and all their growth summaries.

• I can even mark when we had major events to add more context. 
• Each month and year, compare your actuals to what you forecasted to see how far off you were. 
• With all of this new data in hand, it's up to you to monitor it and feed learnings back into the system. 

It's not enough to mark an OKR as "not on track". You can now measure the % complete, set an alert for when an OKR is at risk of not being completed, and calculate the daily average and pacing changes it would take to make up the gap. And if you can't make up the gap in time with the current strategy, you now can stop and pivot earlier, rather than waiting for a monthly report or QBR. The gd light saber is now in your hands!!! 

Now you can focus on data interpretation, rather than collection. 

Reforecasting and H2

Pro tip: If leadership changes any parts of the budget, go back and re-forecast. 

We've heard some horror stories about how that doesn't matter – I can attest that often, your forecast gets locked in, then the budget gets pulled, but the goals don't change. Magically, you're supposed to just make it happen with thin air. 

Somewhere along the way, OKRs turned into both performative and punitive exercises. Leaders need to learn they cannot have it both ways – if you're faced with this kind of situation, I feel you. At least you will know you did your due diligence. 

It may be common to be asked to "reforecast for H2", mostly because finance needs to adjust targets for revenue and costs based on what has actually happened in the 1st half of the year – giving you another chance to adjust expectations, reflect on what is working and what is not. 

If you have any insights to add to this article please contact us.