I know it's crazy, but hear me out. I've got a huge (n = 7) sample size and everything, okay? And I can tell ya this much for free, bucko: I always got an offer when I wore makeup, and I usually didn't get an offer when I didn't. Just take a look at this totally legit table:
|Interviewing company||Wore makeup||Got offer|
The verdict? Makeup makes you a more skilled software engineer.
But seriously, I'll admit there's an obvious confounding factor here: Maybe I just cared more about the interviews where I wore makeup, so I put more effort into those. While I'll concede that I really did care more about those, I still think my performance in all the interviews was roughly equally as bad.
And maybe the correlation is just spurious. The sample size is pretty small after all.
But I don't think so. I'm pretty sure makeup just makes people like you more, otherwise celebrities, news anchors, and women wouldn't waste so much time painting their faces.
If makeup really improves your odds of getting through a software engineering interview, you might see this as a poor reflection on the industry. Candidates should be assessed solely on their technical abilities, not on their looks, you could argue. But chances are you also think that candidates should come to interviews well groomed. If you agree with both these things, I'd put it to you that you hold self-contradictory beliefs. "But," you say, "there's a difference between 'looks' (i.e. attractiveness) and grooming!" I agree, but people don't groom themselves to look like Oscar the Grouch. People groom themselves to increase their attractiveness.
While I don't want to prescribe a set of guidelines to candidates or the industry for how they should present themselves or conduct interviews, I can see why looks matter. In my case, the dark circles under my eyes give an impression that I lack mental acuity—Obviously a useful trait for software engineers. Even though this is a pretty coarse-grained signal about the quality of an engineer's work, interviewers resort to using it because interviews are too short to give information which is much more accurate than that.
The paucity of good quality signals suggests that the hiring process should be longer. (I can already hear a mob of angry engineers marching to my door, brandishing pitchforks and tiki torches). But it's hard to say how a longer interview can be achieved in a fair and practical way. Maybe the provisional period typical jobs have could have an "even-more-provisional" part for the first few weeks. Of course, to make this worthwhile for employees, the post-provisional period would have to promise even better pay. And even then, I'm not sure such a casualised workforce is a good thing for society, so forget I said anything.
We learnt nothing. But it might be cool to keep the above table up-to-date with data from future interviews.