Work emails may reveal more than an employee survey

Analyzing the language used in an employee’s everyday workplace communications (such as emails, Slack messages and voice calls) can tell a lot about how they feel about their work. But it’s not what they write, but the language they use, according to a Canadian tech firm called Receptiviti.

Its method has its origins in natural language processing (part of machine learning) but comes at it from a different angle. The technology looks at the passive parts of speech—the bits we use without thinking. This holds the key to how happy we are, claims Receptiviti.

Instead of looking just at an individual employee in isolation, the technology analyses it according to traits and teams. So the data may be sliced to look at gender and department – revealing for example that women in the accounts team are less happy than the men.

People who are less happy at work tend to use personal pronouns such as “I” and “me” more than average. So be careful what you write.