Clustering & Office Parties
Team Networking at Office Parties: A Clustering Analogy
Clustering is a statistical learning algorithm. There are thousands of clustering algorithm variations, but the basics remain similar. Clustering groups similar items together based on different features.
For instance, imagine you’re working with a list of food items and their sweetness scores. Suppose you want to create two groups (or clusters). The clustering algorithm minimizes the differences in feature values within each group. If you input rasgulla, tiramisu, and chicken curry with their respective sweetness scores, the algorithm will create two clusters. One cluster will include rasgulla and tiramisu, while the other will contain chicken curry.
Ordinary Analysis is totally free, and I’m happy to share it with you. If you want to receive every new post in your mailbox, you can subscribe by clicking on this cool button below.
I have realised one thing, office parties are like typical clustering algorithms. Specially the ones where multiple teams are involved.
In office, based on familiarity, there can be three types of people (I had blogged about it here):
Type A: Your teammates. You interact with them every day.
Type B: You don’t know them at all.
Type C: The awkward types. You might have worked with them for one project or assignment, but you don’t know them well. So you now have small talks with them.
Now coming back to the office party set up. In most of these office parties, there is “the Leader” who heads multiple teams. So the onus is on her to make sure that her teams interact with each other. Mostly to know the type B (unknown acquaintances) and type C (awkwardly familiar) people. At the beginning, there will be an attempt to randomly order all the individuals that are attending the party. There are quite a few ways to do that such as arranging people based on their date or month of birth or according to the alphabet of their name.
At the start of the party, this works. But the true clustering unfolds when the leader, perhaps fueled by a few drinks or simply caught up in the party’s enjoyment, forgets about team networking. It’s as if the algorithm suddenly shifts gears, and the clusters take on a life of their own.
The Leader, usually responsible for team interactions, now joins the fun without restraint. The initial order dissolves, and people naturally gather with those they feel at ease with. It’s like watching clusters emerge spontaneously—often centered around a single team.
Office parties, especially those involving colleagues from different teams, can feel a bit ‘weird’ to many people. So one would try to maximise how comfortable they are in these parties. The best way to do this would be to to talk and mingle more with the people that you already know, that is the type A people.
Thank you for reading Ordinary Analysis. If you enjoy this post and want to see more, you can subscribe by clicking on this cool button below. It’s free, easy and rewarding.
Alternatively, you can also share it with someone you think will like it.