How to Forecast Revenue in Google Sheets with ChatGPT

Building a revenue forecast in Google Sheets often feels like you need a degree in statistics to get it right. But what if you could create a solid projection simply by asking for one in plain English? By connecting ChatGPT to Google Sheets, you can turn your historical data into an actionable forecast for the months ahead. This article will walk you through exactly how to set up and use this powerful duo for everything from simple trend analysis to complex scenario planning.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

Why Forecast with ChatGPT in Google Sheets?

Traditionally, revenue forecasting involves wrestling with complex formulas like FORECAST.LINEAR or setting up statistical models that can be intimidating. While those methods are powerful, they aren't always approachable for startup founders, marketers, or small business owners who just need a quick, reasonably accurate projection to guide their decisions.

Here’s why using ChatGPT for this task is a game-changer:

• It's Conversational: You don't need to know statistical syntax. If you can describe what you want — like "project our sales for the next quarter considering last year's holiday spike" — you can create a forecast.
• It Understands Context: Unlike a simple trend line formula, ChatGPT can interpret written context. You can tell it about an upcoming product launch or a planned marketing campaign and ask it to factor that into the forecast, making your projections far more realistic.
• It’s Great for Scenario Planning: Quickly model different futures. Ask for a "best case" forecast that assumes 20% growth and a "worst case" forecast assuming a 5% downturn. This takes seconds, whereas building it manually in a spreadsheet would take much longer.

Think of it as having a junior data analyst you can talk to directly inside your spreadsheet. It’s perfect for getting quick answers and exploring possibilities without getting bogged down in technical details.

Step 1: Get Your Historical Data Ready

Before you can forecast the future, you need a clear picture of the past. ChatGPT's predictions are only as good as the data you provide, so preparing a clean data set is the most important step.

You need historical revenue data, typically on a monthly basis. Ideally, you should have at least two to three years of data to give the model enough information to identify trends, patterns, and seasonality.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

How to Format Your Data

The structure should be incredibly simple. Create a new Google Sheet and set it up with two columns:

• Column A: Date (e.g., Month, Quarter)
• Column B: Revenue (e.g., Sales, MRR)

Your sheet should look something like this:

Tips for Clean Data:

• Be Consistent: Make sure your date format is the same for every entry (e.g., YYYY-MM-DD). Use either the end of the month or the beginning, but stick to one.
• No Gaps: Ensure you have an entry for every period. If your business had zero revenue one month, enter a 0, don't just leave the cell blank. Missing data can confuse the analysis.
• Plain Numbers: Format your revenue column as a number, not text. Leave out the dollar signs and commas within the cell itself, use Google Sheets' formatting options to make it look like currency.

Step 2: Install and Configure the GPT Add-on for Sheets

To let ChatGPT access your Google Sheet, you need to install a third-party add-on. The most popular one is called "GPT for Sheets and Docs." It's a reliable bridge between your spreadsheet and OpenAI's AI models.

Installation Guide:

1. In your Google Sheet menu, navigate to Extensions > Add-ons > Get add-ons.
2. In the Google Workspace Marketplace search bar, type "GPT for Sheets" and press Enter.
3. Find the "GPT for Sheets and Docs" add-on and click on it.
4. Click the Install button and follow the prompts to grant the necessary permissions. You’ll need to allow it to access your Google account to function.

Once installed, you'll find the add-on's options back under the Extensions menu.

Connecting Your OpenAI API Key

The add-on needs an API key to communicate with OpenAI. This key is your unique password that proves you have an OpenAI account.

1. Navigate to the OpenAI API key page: https://platform.openai.com/api-keys. You will need to log in or create an account. (Note: Accessing the API is a paid service from OpenAI, but it is extremely inexpensive for this kind of use case).
2. Click on "Create new secret key." Give it a descriptive name like "Google Sheets Forecast" so you remember what it's for.
3. A new key will be generated. Click the copy icon to copy it to your clipboard. Important: OpenAI will not show you this key again, so be sure to copy and paste it into a safe place or directly into Google Sheets in the next step.
4. Back in your Google Sheet, go to Extensions > GPT for Sheets and Docs > Set API Key.
5. Paste the secret key you just copied into the text box and click "Save API Key."

You're all set! Now you can start using =GPT() functions directly in your sheet.

Step 3: Creating Your First Revenue Forecast

With everything set up, a simple formula is all you need to start forecasting. We'll start with a basic linear forecast and then add nuance with more detailed prompts.

Let's assume your monthly revenue data is in cells B2:B25.

Basic Linear Forecast

A linear forecast is just a straight-line projection based on your past growth. It answers the question, "If we keep growing at our current average rate, where will we be in six months?"

Click on an empty cell and enter the following formula:

=GPT("Given this historical monthly revenue data from B2:B25, forecast the revenue for the next 6 months. Provide only a comma-separated list of numbers.")

Hit Enter. After a few seconds, ChatGPT will process the request and place the forecasted values for the next six months in the cells to the right. It’s that simple.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

Adding Seasonality and Context

A straight line is a good start, but most businesses aren't that simple. They have busy seasons and slow seasons. A good forecast accounts for this seasonality. You can guide ChatGPT by giving it more context in your prompt.

Let’s refine the prompt to account for a typical end-of-year sales spike:

=GPT("Given the monthly revenue in B2:B25, forecast revenue for the next 6 months. The data shows strong seasonality with peaks every Q4. Project the following months considering both the overall growth trend and this seasonality. Give me a comma-separated list of 6 numbers without currency symbols.")

This more detailed prompt instructs the model to look for a specific pattern (Q4 peaks) and apply it to the forecast. The results will be more accurate because they mirror your actual business cycle, not just an abstract growth line.

Using ChatGPT for Scenario Planning

This is where the combination truly shines. You can model different outcomes without rebuilding your entire spreadsheet. Simply create different prompts for different scenarios.

Realistic Case (based on current trend):

Use the seasonality prompt we just created above.

Best-Case Scenario:

Maybe you're about to launch a marketing blitz. You can ask ChatGPT to model that.

=GPT("Forecast our next 6 months of revenue based on the data in B2:B25. For this forecast, assume our upcoming marketing campaign increases month-over-month growth by an additional 10% on top of the existing trend. Provide a comma-separated list of 6 numbers.")

Worst-Case Scenario:

What if the market takes a turn? A conservative forecast helps you prepare for the worst.

=GPT("Forecast our next 6 months of revenue based on the data in B2:B25. Be very conservative and assume a slight economic downturn that flattens our growth to almost 0% for the next two quarters. Provide a comma-separated list of 6 numbers.")

By running these three simple functions, you can instantly see three different potential futures for your company, which helps tremendously in budget planning and strategic decision-making.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

Tips for Better and More Accurate Forecasts

To get the most out of your AI-powered forecasts, keep these best practices in mind:

• Be Hyper-Specific: The more detailed your prompt, the better the output. Instead of "forecast sales," say "forecast monthly revenue for the next 12 months based on historical data from A2:B37, considering a 15% increase in Q3 due to a product expansion."
• Iterate on Your Prompts: Your first prompt may not give you the perfect result. If the output seems off, adjust your prompt by adding more context or rephrasing your instructions. Treat it like a conversation.
• Tell It the Output Format: Request your desired output format directly in the prompt, like "provide the output as a comma-separated list of numbers" or "return only the final number." This saves you from having to clean up the output later.
• Use AI as a Co-pilot, Not on Autopilot: AI-generated forecasts are powerful guides, but they shouldn't replace your own business expertise. The AI doesn't know about that new competitor who just entered the market or the key employee who is leaving. Always sanity-check the numbers against your own industry knowledge. The best forecasts blend AI-driven analysis with human insight.

Final Thoughts

Connecting ChatGPT to Google Sheets makes powerful revenue forecasting accessible to anyone, regardless of their technical know-how. By turning historical numbers into clear, context-aware projections, you can stop guessing and start making data-informed decisions for your business with nothing more than a simple prompt.

As you get more comfortable analyzing data this way, you might find that the spreadsheet itself becomes the bottleneck. Building custom reports across different platforms like Google Analytics, Shopify, and your ad accounts still involves a lot of manual data gathering. We built Graphed to solve this specifically. It securely connects to all your business tools and allows you to build real-time dashboards and reports just by describing what you want to see — no spreadsheets, no add-ons, and no API keys required. It's the next logical step after you’ve realized the power of analyzing data through conversation.