Google brings no-code machine learning to Sheets with SimpleML

A CGI owl holds a conical flask containing green liquid stands next to the words 'Simple ML for Sheets'
(Image credit: Google)

Google has released Simple ML, a tool to help apply machine learning to data prediction and sorting tasks within Google Sheets.

Available as an add-on, the tool is intended to be stress-free to install and use, while powerful in its potential applications. Using a sheet with at least two columns, users can quickly train and refine a machine learning model to their own needs, opening up possibilities for automation across a company’s existing Google Workspace.


Flexible IT models drive efficiency and innovation

A modern approach to infrastructure management


The tool is intended to enable enhanced activity in Sheets for both beginners and advanced users. It can be used for tasks like predicting how long maintenance could take based on past data, or spotting potential fraud when analysing unusual data in a client's finances.

By default, the tool is shipped with two functions: ‘predict missing values’ and ‘spot abnormal values’. For more complex use, it can even predict values within large data sets - Google gave the example of researchers using Simple ML to predict a person’s age based on their DNA in a molecular ageing study.

Simple ML was made by the same developers as TensorFlow, and is pre-trained to be as effective at processing data as possible. Despite the complicated nature of ML algorithms, and to combat the intense learning curve traditionally associated with training and deploying a machine learning (ML) model, the tool can be used on a fully no-code basis.

Nevertheless, advanced users that are already familiar with training machine learning models still seek to benefit from Simple ML, which allows for detailed evaluation of models to understand the quality of models that they have trained using the tool.

Models are broken down into a number of metrics, depending on the way the model was originally trained, and demonstrated through data and graphical representations.

Google recommends using Simple ML on sheets which contain columns with at least 20 values, though 100 is the recommended number for a model to really start training with accuracy.

Depending on the complexity of the task, far more may be required, and when a task cannot be completed, a detailed error message indicates what has prevented the model from running as intended.

A key benefit of the service, as cited by Google, is the fact that it runs entirely in a browser, so all sensitive data that could be included on a Sheet remains secure.

In addition, ML models are saved directly to Google Drive in their own folder, enabling them to be shared between a team in a business environment or downloaded for backup should they be crucial to new business processes.

In a blog post on the release, Google outlined the ease of use by listing the three steps necessary: opening the data; selecting the task that suits the use case, such as identifying anomalous data; running the model and seeing the results, which are weighted by probability.

On a technical level, Simple ML utilises decision forests - a group of randomly-generated decision trees governed by a machine learning algorithm to output predictive results.

Google announced the update at the annual Women in ML Symposium, an online event held by the firm to promote an inclusive community within the field. Simple ML is available right now in beta.

Rory Bathgate
Features and Multimedia Editor

Rory Bathgate is Features and Multimedia Editor at ITPro, overseeing all in-depth content and case studies. He can also be found co-hosting the ITPro Podcast with Jane McCallion, swapping a keyboard for a microphone to discuss the latest learnings with thought leaders from across the tech sector.

In his free time, Rory enjoys photography, video editing, and good science fiction. After graduating from the University of Kent with a BA in English and American Literature, Rory undertook an MA in Eighteenth-Century Studies at King’s College London. He joined ITPro in 2022 as a graduate, following four years in student journalism. You can contact Rory at or on LinkedIn.