Forecast Data in Connected Sheets with BigQuery ML

Google has introduced a powerful new capability in Connected Sheets for BigQuery: users can now generate forecasts directly inside Google Sheets using BigQuery ML (BQML) and Google’s advanced TimesFM forecasting model

This update removes the need for SQL, Python, or custom model training making machine learning forecasting accessible to business users directly in Google Workspace.

If your organization uses BigQuery and Google Sheets, this changes how forecasting can be done at scale.

What Is the New Forecasting Feature in Connected Sheets?

Google Sheets users connected to BigQuery can now:

• Generate time series forecasts
• Predict future sales or demand
• Model revenue projections
• Forecast operational metrics

All from within the Sheets interface.

The feature leverages:

• BigQuery ML (BQML)
• TimesFM, Google’s foundation forecasting model

TimesFM is pre-trained on billions of real-world data points, allowing organizations to generate sophisticated predictions instantly without building or training their own models.  This is a major shift toward no-code machine learning in Google Workspace.

Why This Matters for Google Workspace and BigQuery Users

Traditionally, forecasting required:

• SQL-based model building
• Python notebooks
• Data science expertise
• ML training pipelines

Now, forecasting can be initiated directly in Google Sheets using a configuration panel dramatically lowering the barrier to entry

This enables:

• Finance teams to forecast revenue
• Operations teams to model demand
• Sales teams to predict pipeline trends
• Product teams to analyze usage growth

Without relying on dedicated data scientists for every projection.

Key Features of BigQuery ML Forecasting in Sheets

According to the official release, the new feature includes

1. Simple Configuration

Forecasts can be created from:

• Any existing BigQuery dataset
• A custom BigQuery query

All through a user-friendly configuration pane in Sheets. No SQL required.

2. Customizable Forecast Parameters

Users can define:

• Prediction horizon (how far into the future)
• Confidence intervals
• Default model options

This provides flexibility without complexity.

3. Granular Breakouts

The feature supports forecasting across multiple dimensions.

For example:

• Sales by region
• Revenue by product category
• Demand by SKU

Multiple time series forecasts can be generated simultaneously

4. Automatic Visualizations

For single time series forecasts, Sheets automatically generates a chart showing:

• Historical data
• Forecasted trend
• Visual comparison

This accelerates insight delivery for business stakeholders

How to Create a Forecast in Connected Sheets

To use the feature:

1. Create a Connected Sheet linked to BigQuery
2. Go to the Preview view
3. Click Advanced Analytics
4. Select Create a Forecast
   

There are no admin controls required for this feature

Availability:

• Rapid Release and Scheduled Release domains
• Available to all Google Workspace customers and personal Google accounts
  Google Workspace Updates_ Forec…

Rollout began February 16, 2026.

What Is TimesFM?

TimesFM is Google’s foundation model for time series forecasting.

Unlike traditional ML workflows, TimesFM:

• Is pre-trained
• Requires no custom model training
• Works immediately on structured time series data

This makes forecasting dramatically faster and more accessible to non-technical users. It also signals Google’s broader strategy of embedding foundation models directly into Workspace and Cloud workflows.

Use Cases for Enterprise Teams

This feature is particularly powerful for:

Finance

• Revenue forecasting
• Budget projections
• Cost trend analysis

Retail & E-commerce

• Demand forecasting
• Seasonal sales predictions
• Inventory planning

SaaS Companies

• MRR projections
• Churn modeling (time-series inputs)
• Usage growth forecasts

Operations

• Supply chain demand
• Resource planning
• Production capacity modeling

Because it runs directly against BigQuery, it scales to enterprise-sized datasets without manually importing data into Sheets.

Strategic Implications for IT and Data Teams

This update reduces friction between:

Data teams → Business teams

Instead of:
Business asking Data Science to build models

Business users can:
Self-serve forecasts using governed BigQuery datasets

This improves:

• Agility
• Decision speed
• Executive reporting
• Data democratization

While still keeping data centralized in BigQuery.

Is Your BigQuery + Workspace Environment Optimized?

Many organizations use:

• BigQuery
• Connected Sheets
• Google Workspace

But fail to fully leverage advanced analytics capabilities.

Suitebriar helps organizations:

• Optimize BigQuery + Workspace architecture
• Design governed Connected Sheets environments
• Enable forecasting workflows
• Align ML capabilities with business units
• Secure and scale enterprise analytics

If you want to activate no-code forecasting inside Google Sheets or evaluate how BigQuery ML fits your environment, schedule a data strategy session with Suitebriar.