IBM’s Asshole Test - Positioning in current enterprise context

Context :

“ Some years back I applied to join IBM's grad scheme, there was a peculiar stage to the process I've not seen elsewhere.

It was during the onsite day, where a batch of 20 or so applicants were put through various tests in an IBM office. They called it the "group test"; around 8 of us were led to a room and asked to solve a puzzle together. Each of us was given an information pack, there was a white board, and a timer ticking down from 60 minutes.

At first there was silence as we looked at our packs, then the first voice: "Let's pool our information", someone stands up by the whiteboard, grasping a marker. Silence, it's not clear how this information should be parsed. One person starts reading theirs out word for word. This is not going to scale. Someone interrupts. Before long the whiteboard leader has been deposed and another is wielding the marker. Then another fights for control. The 60 minutes has run out, the puzzle is unsolved. Confused and drained we head off to the next test.

At the end of the day the group is split into two rooms, my room is given the good news and I go on to join the grad scheme.

Six months later I am shadowing a colleague who is running the "group test".

I asked him if he'd ever seen a group complete the test?

"Oh, it's not about that, this is an asshole test.

You see who turns into an asshole under pressure and they don't make it to the next round ".

Referenced from the blog - https://johnpublic.mataroa.blog/blog/the-asshole-test/

Analysis:

There are quite a few critics of this approach, and rightly so.

Is it really possible to identify a non-collaborator, or potential disruptor of team hygiene in the context of an interview setting, that too in a small-time frame of approximately 60 minutes, or so ?

Most interviewees are anyhow nervous during interviews, and it might come across as being rude, or dysfunctional, or not polite, but the reality could be that they were just tensed, or not prepared.

In any case, there is a rational that one cannot judge a man’s nature, or character in such a short time span, and certainly not by these sorts of quick-fix tests/assessments, which do not have any backing in science/research.

Some critics have gone so far as to claim that the entire process is flawed, and they strongly discourage prospective talent from joining companies who try to judge competencies of potential hires based on these sorts of assessments, which according to the critics, are biased in nature, and not perceptive of the true depth of human character.

Modern adaptation:

From my perspective, this is a nuanced subject, and needs to be set in the right context, in order to make it successful.

It is indeed true that many interviews in the current enterprise ecosystem, especially those in the technology space, are more focused on the technical/cognitive capabilities of the interviewee and does not equally focus on :

• whether they have a strong collaborative mindset
• whether they can seamlessly work within, and with their teams
• keep their composure, and react well under pressure
• or at the least, be a good team player.

It is unfortunate that these skills, which are key to unlocking the true value of an enterprise are deemed less important, than say the ability of the interviewee to reverse a binary tree in O(n), and so on.

However, more and more industry specialists are emphasizing that team work and collaboration are more important than ever, especially in the current context of DevOps, Platform Engineering, and agile workstreams.

Most software teams have adopted the agile methodology which is built on the essence of strong collaboration, and team work; where engineering teams work together in sprints to deliver a usable software component at the end of evert development cycle.

Conclusion:

Will modern software development be possible without collaboration ? I strongly believe no.

Will a non-collaborative, disruptive team member be able to negatively impact the team hygiene to such an extent that it starts impacting actual deliverables ? Yes, I am confident of that also.

This is why it is so important to not only have individual performers who are good at what they do, but also ensure that every hire has a collaborative mindset, is patient and respectful of their team, have the cognitive and social capacity to interact with the extended team in a coherent, consistent, and effective manner.

We need an efficient, and effective screening system to ensure that collaboration and teamwork is provided as much weightage as coding/cognitive competency/technical expertise.

The so-called IBM asshole test might not be a full-proof mechanism of figuring out the same, but it provides enough context for enterprise leaders to figure out how they want to modify the assessment, to ensure they have some sort of a benchmarking mechanism to figure out who are better suited to work in a collaborative environment than others.

The actual test/assessment does not hold as much importance, as the overall message it wishes to convey. We have different types of folks in our society, some more collaborative, open-minded, composed under pressure, and in-general more compassionate than others, and an enterprise needs to figure out what kind of talent they want to introduce, and accordingly setup a screening process to encourage the same.

Note : I am not aware of any official briefing/memo/documentation from IBM where they explicitly confirmed/accepted that they conduct such sort of an assessment. It might have been an individual initiative from a business unit, might have been discontinued, who knows.