Anxiety with mistakes at work

In my bank in Sweden we have a saying that goes:

”Det som sker det sker”. Which basically means, what happens, happens. The underlying meaning is that, whatever has happened - there’s nothing other to do than to move on.

It’s a terrible feeling and it in some way or another happens to most analysts (though varying degrees of mistakes for sure.) I understand your anxiety about making another mistake but you gotta let it go in order to improve. Your anxiety is going to give you tunnel vision the next time you are running through a check of the model. You’ll be so worried about not having any typos (being perfect) that you won’t catch any obvious mistakes that you would have otherwise caught. The amount you care is a good thing but you need to channel it properly.

I made a similar mistake a few months ago and had that same crippling anxiety feeling. In about 2 weeks you will realize how short term people’s memories are.

Additionally, you just unlocked a whole knew technical / quality control skillset in a very quick amount of time.

Eat the loss, turn it into an advantage and remember that at the end of the day no one actually cares.

You’re having a very typical problem.  People try to be careful, they triple check everything, and inevitably the mistake makes it through.

I noticed this early on my career and have had some thoughts since.

One in particular: becoming more detail-obsessed (particularly checking every number again and again, or any other checking exercise that practically resembles re-doing the work) won’t solve anything, and will actually lead to more mistakes. 

The checking process needs to be a completely different lens/angle than the work-creating process.

Instead of combing through every detail, learn to ask yourself “what would be fucked up if there’s a mistake in the model?”

For example, balance sheet likely wouldn’t balance. IRRs would probably not be sensible overall, or the progression of IRRs across different time periods/multiples wouldn’t be consistent. Margins should progress naturally unless there’s an inflection. Etc etc etc

Most of the clues will be at the end result of the model . . your returns, cash flows, debt pay down etc. Focus on that handful of items and get comfortable with the idea that of all those look reasonable, the model is probably right. 

Yes of course it’s possible to still have mistakes despite all those items looking OK . . but not big mistakes. And if you focus on avoiding the big mistakes, you progress without any blowups and eventually become that person who catches everything.

Totally okay to make mistakes like this and frankly, it's your associate's job to catch these errors as being deep in the weeds on a model often times leads to errors given how myopic your lens becomes. That being said, the cardinal sin is making the same mistake twice - that is what you need to make sure you don't do as that'll peg you as someone who isn't teachable. Don't be afraid to ask for a second set of eyes from another analyst or your associate before sending out deliverables either as it can often help catch simple errors you may have overlooked. And most importantly, don't beat yourself up about mistakes as you'll spiral into a state of paranoia that'll further worsen your work quality. Remain confident and request frequent feedback from your peers prior to distributing widely. 

Let me put it this way. If I’m given a few hours to go through materials of an already closed transaction, there’s a 99% chance that I’ll find a mistake. The volume of work is just too high and things often get overlooked, despite the amount of time spent reviewing by the deal time. The reality is that mistakes happen, some are immaterial, some are not. The latter are easier to spot as they tend to have a greater impact on the output.

I’ll caveat this by saying, that repeatedly making the same mistakes is unacceptable, but materials are NEVER error free. Don’t beat yourself too much over this. Try to have checks throughout. Then take a step back and ask yourself whether outputs make sense and go from there. This will help you clean up 80% of material issues in early drafts.