Gallop openly acknowledges that of all 12 questions, this one is met with the most questions and skepticism, but they insist that it is predictive of top-performing teams.
Do you trust the research behind that question, and do you think it's a fair question to ask on engagement surveys?
Given a large company you can determine the cohesion between all employees and compare that to other (successful) companies.
I am no expert in any of this but this appears to be a proxy metric for tenure. As tenure increases people are more likely to consider someone a friend. So being below average on the cohesion would mean to me that the company needs to work on employee retention.
My 2 cents
I'll go to bat for my team/trust they'll do the same, but I avoid friendships at work for the same [personal] reasons one may avoid working with family.
I have a lot of acquaintances. None of them know what really makes me work, so I don't consider them friends. I'm pretty sure they wouldn't like what's behind the masks...
Now, in terms of effectiveness, it's a wash in my opinion. Distance has value, I can be the bad guy and not worry about it too much. I miss out on a lot but that's fine, I've seen a lot too.
If nothing else I want to offer this distinction: friends and friendly
I agree with the notion that you should be careful about letting many work people know your personal business. However, close teams do work better together and good friendships make work so much more pleasurable and enrich your life. Part of the problem is that our work relationships are a huge part of our job and we need to manage them carefully. But all you have to do is look at people in working class jobs where the relationships have no impact on the job, and you see what great camaraderie they often have. I miss that.
Having experienced both, I’d actually choose a boring job with good friends vs an engaging job.
That question is one of the most ridiculous -- so ridiculous that I (and the rest of my team) aren't even sure what is actually being asked.
It can't possibly be "best friend" in the traditional sense, as that simply doesn't make any sense. I have rarely seen a workplace where people are working with their best friends, mostly because most people have established their "best friends" well before they took their current position.
The consensus our team has reached for that question is to just answer "yes", because if too many people answer "no", then the result is that we have to develop some sort of action item to address this "problem".
In the end, the entire thing is just bullshit intended to make upper management feel better.
TL;DR; its a gauge to see how willing to leave you are.
Long version: For some crazy reason I am often interested in why people leave a job. Of course there is a corollary, why do people stay? Especially if the job is terrible. In talking with a lot of people in many professions, the most common theme applied to why people stayed at a bad job.
People fail to move on from a bad job if they have a sense of loyalty to the co-workers and "cannot leave my friends behind".
Although it should be obvious, the fewer ties you have the more willing you are to move on.