Emojis have come a long way since their introduction in 1998. In fact, in 2020 the Unicode Consortium officially added 117 new emojis bringing the total number of approved emojis to 3,304. Emojis (also known as emoticons, digital icons) have become a significant element of business communications thanks to instant chat services like Slack, Skype and Teams, yet their usage in the workplace has not been rigorously studied.
My team and I recently worked on a project to anonymously analyze how leaders use emojis while communicating with their team members at work. We parsed anonymized emoji reactions from Slack messages at four different enterprises over a 180-day period in 2020. The total data set included over 83,000 messages with over 101,000 emojis of 466 different types (custom Slack emojis were excluded). Our team also extracted metadata on the messages the emoji were responding to and the sentiment of the emoji themselves. This gave us the unique ability to not only look at summary distributions of emoji usage, but to also analyze the utility of emojis in directing conversation.
It’s worth noting that (surprisingly) there was very little change in the data from Q1 2020 to Q2 2020, so the increase in remote work thanks to the coronavirus pandemic did not appear to have a significant impact on emoji usage.
Now on to some of the key findings:
Do people stick to using the same set of emoji? In short, yes, they do. Some 71.2% of users used fewer than 10 emojis in the last 180 days and 50.7% of users used fewer than five emojis in that time. It’s not clear from our data if this is due to personal preference, or if the “Frequently Used” and “Handy Reactions” menus in Slack encourages people to stick to the same set (likely some of both).
Do people use certain types of emoji in response to specific types of messages? There are significant differences in emoji usage based on the intent of the message that the emoji is applied to. We calculated the difference between the global usage rate for each emoji, and the usage rates of that emoji in response to messages of a specific intent (we've fine-tuned BERT to classify messages into a number of conversational intents, such as "giving praise", "scheduling a meeting", "requesting information", etc.).
For example, an ad-hoc scheduling request will receive the Thumbs Up emoji 16% more often than general posts will (they are also likely to receive the Coffee Cup). Posts that give recognition receive Clapping Hands 6% more and the Noisemaker 4% more. Posts that inform that work has been completed get Thumbs Up 15% more (and are also more likely to receive the Rocket Ship) and posts that express doubt also get the Thumbs Up 9% more.
Do managers and employees use emojis uniquely? The top five emojis used by managers are entirely different than the top five used by their team members. In general, manager emojis tend to show positivity and appreciation (like the Clapping Hands and the Noisemaker) and team member emojis tend to show acknowledgement (like Checkmark and Eyes). Our team is pleased to see so many positive and encouraging emojis used regularly by managers!
Do different companies use emojis differently? Our team found overlap in how different companies used Slack emojis, but also significant differences. Companies do seem to develop their own similar but distinct emoji “vernacular.” This could be caused by a variety of elements including company culture, differences in team structures, etc.
Despite all these differences, which emoji made the leaderboard overall? The most commonly used emojis overall were (in order of usage) Thumbs Up at 30%, Heart at 8%, Laughing Face at 6%, Fire at 5%, and the Noisemaker at 4%. As you can see, Thumbs Up is the most popular by a wide margin and all of the top five emojis have a positive sentiment. Negative emojis are not used often.
All in all, emoji use is more customized that you might think and depends on an individual’s company and position. Next time you include an emoji in your chat service of choice, consider why you chose that particular one -- it just might help improve your digital communications with your team.
Andy Horng is Co-Founder and Head of AI at digital leadership platform Cultivate. He has a background in data science and has worked as a software engineer building machine learning tools for legal document analysis and medical research. He holds a BS in Electrical Engineering/Computer Science and a BA in Cognitive Science, both from the University of California Berkeley.