Do your employees laugh at your jokes? ?

How do you track if your employees think you're funny? And should you even be tracking that?

“A sense of humor is part of the art of leadership, of getting along with people, of getting things done.” – Dwight D. Eisenhower

Humor can help you build a more vibrant, creative, inspiring workplace — it’s both a driver of success and reflects success at work. It’s core to attracting and keeping great employees. It’s a stress-buster and is proven to help improve wellbeing. A sense of humor is also a strong trust-builder and trust is core to work relationships across the board. Humor is also a powerful catalyst for creative thinking – therefore can help drive a creative workplace. Research on humor as a moderator of leadership effectiveness found that the use of humor in constructive ways helps shape the nature of the culture we want to create. 1


So how can you track humor at work? 

According to Yew et al (2011) 2, it’s possible to build a relatively accurate Naive Bayes classifier that uses social action data, without the use of content or media-specific metadata, to understand if something is ‘funny’. On this basis, it could be possible to build up a picture of whether an email, presentation, meeting was perceived to be funny based on how someone interacts with it. Sisense, a platform that enables developers, data engineers, and business analysts to simplify complex data and transform it into powerful analytic apps, developed a tool that scrapes text for words that are most commonly used in popular jokes to gauge if words or subjects that are commonly perceived as funny are being used. 3


Should you track how funny things are at work? 

I would argue that this could be a healthy practice. Having a high-level understanding of whether key all-hands, email announcements, etc. are being perceived in a positive light and are contributing to a positive culture, or not, can be incredibly helpful. At Google, googliness is a core trait that’s regularly measured across the organization. It is meant to measure the prominence of the signature positive culture that Google embodies – and a sense of humor is characterized by the elusive concept of googliness. Googliness is one of the core criteria in recruiting, but also is held as a standard for leaders and is regularly pointed out as one that is very hard to measure – perhaps a more robust understanding of humor in communications could be indicative of googliness.  

I would however push that we shouldn’t be optimizing around humor data. I would hypothesize that over-fitting for humor has diminishing returns. When you try too hard to be funny, it falls flat and this can have negative consequences on culture. 

Furthermore, as was mentioned in the aforementioned article in The Academy of Management Journal, it’s important to be aware of the potentially negative impact of humor, especially when audience expectations and group composition is not considered. This could be captured by cohorting the group and understanding individual sentiment, however this could require identifying information.

Hence, privacy violations are a key concern. This is especially the case with the method that requires tracking of commentary on the shared content as well as the sharing patterns. Therefore, I’d recommend that only high-level reactions to the degree and type of humor be tracked and if more granular results are needed that they only be used if the sample size is large enough and can be effectively anonymized.


In short, humor is a serious matter.


 1 Avolio, Bruce J., et al. “A Funny Thing Happened on the Way to the Bottom Line: Humor as a Moderator of Leadership Style Effects.” The Academy of Management Journal, vol. 42, no. 2, 1999, pp. 219–227. JSTOR, Accessed 15 Apr. 2020.
2 Jude Yew, David A. Shamma, and Elizabeth F. Churchill. 2011. Knowing funny: genre perception and categorization in social video sharing. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’11). Association for Computing Machinery, New York, NY, USA, 297–306. DOI:
3 Roth, E. Analyzing What Makes Us Laugh. Sisense. 2014.


A Trail of “Bread-Crumbs” Leading to Organizational Insights.


Closing the gap between People Analytics and Workforce Development

Student comments on Do your employees laugh at your jokes? ?

  1. As a lover of and devotee to all things comedy, I agree- humor IS a serious matter! It brings people together in times of stress or division, it provides people with hope and relief, and it humanizes those who use it. Even during times of crisis, when used effectively, humor sets a helpful tone and focus for what we need to be worried about versus what we can not take so seriously (like ourselves). I also agree with your urging of caution in how to use this data. If the presence of this data were to push people in the workplace to deliberately try to be funnier, I suspect this would backfire. I think what is more pertinent is the evidence surrounding the fact that simply the presence of humor in the workplace makes for better culture and more effective leadership. Instead of taking this information and aiming to hit a certain joke quota in emails and presentations, I urge organizations to make room for moments of comedic relief (employee contests, meme of the week, talent shows, etc) and to celebrate humor when it arises organically.

  2. Great article! I agree with you and Jade that humor is valuable in many situations, but we should be very cautious about what we measure and how it might influence behavior (I have had many coworkers tell terrible jokes in the past, and definitely would not want to encourage them to try harder!). I also suspect that the effectiveness of humor is heavily dependent upon the existing culture and leadership styles of those involved, making this difficult to generalize. Finally, it should be pointed out that humor is subjective, and what is funny for some may be hurtful / offensive for others. If humor were to be evaluated more systematically in the workplace I would want to be extremely mindful of employees’ varying perspectives and how certain attempts at humor may impact them.

  3. Really interesting topic, Jad! I appreciate what you pointed out on the negative impact on culture and subgroups. It seems that this algorithm would narrow in on a homogeneous “acceptable” humor. This in itself subverts the end goal of building trust and creativity by alienating anyone outside the norm. Beyond that, so much of humor comes from uniqueness of perspective or performance, and the art of this would easily be lost in an algorithm.

  4. Catchy title! You definitely got me drawn in here!

    You lost me a little bit with “it’s possible to build a relatively accurate Naive Bayes classifier that uses social action data, without the use of content or media-specific metadata, to understand if something is ‘funny’”. I tried to Google it, but I think I need a proper Googler to explain ;-).

    Taking a step back, I wonder whether humor is really measurable without the use of voting buttons, polls, or perhaps audio measurement (laughs) and/or face recognition (smiles). I also wonder whether we actually need it. As you rightly point out, humor is used with a specific cause in mind:
    – on a granular level, is this piece of communication good / positive / professional?
    – on a higher level, how do you perceive this leader?
    – on an even higher level, how would you rate our culture on these parameters?
    Taking this all together, perhaps it’s easier to measure the effect rather than the input here?

  5. Interesting thoughts Jad. And I can definitely see the worry here in attempting to define humor based on a static, specific sense of what’s funny. Not only does it lead to more homogeneity but also a lack of creativity which can be counterproductive. Something said in one context can be hilarious, but completely fall flat in another. How do you account for that?
    In a way it almost brings to mind art in the middle ages when the church would dictate what constituted art (or didn’t). Only much later did we discover that this was not the way to go.

Leave a comment