We need to stop firing low-level people for mistakes

Some of you may have heard the story of how a customer of US Airways complained about something on Twitter and the response from an employee in the customer relations division came with an attached photograph of a women in a sexual pose.

There were immediate calls for the employee to be fired for this act but to their credit, US Airways said they would not do so because it was an ‘honest mistake’.

We all make mistakes in our jobs, especially if the job requires us to think and act with some autonomy. Most of the time we are lucky because our mistakes are not as spectacular as this particular one or because very few notice them. If people fear being fired for every mistake, they will be miserable employees who do not take any initiative and do no one any good. People should only worry for their jobs if they are irredeemably incompetent or systematically engage in rudeness towards customers or commit acts of malice or fraud or sabotage against the company.

And all of us should refrain from calling for the heads to roll of people who simply make mistakes, however egregious, especially if they are low-level. These people and their families need their jobs and calling for someone to be fired is not something to be done lightly.


  1. mastmaker says

    Easy now, Chiroptera.

    Mano CAN be fired for this mistake, since he is not ‘low-level people’!

  2. jamessweet says

    I agree with this, with the following caveat: I’ve come to accept that us not-so-hairy primates seem to have a very real need for some bizarre ideal of “justice” that doesn’t always serve a practical purpose, and sometimes in fact runs directly counter to practicality. Sometimes, depending on the mistake, I think one can make a case that it’s worth making a quite literal scapegoat out of the person who (however innocently) was responsible for said mistake, as a way of satisfying this weird need for “justice”. People just seem to be happier when they live in a world where they feel like people get what’s coming to them — even when it serves absolutely no useful purpose whatsoever.

  3. moarscienceplz says

    Either the ‘low-level employee’ was 1) using their personal electronics to do work, or 2) they were copying porn to company property. If the first – why? Did US Airways not provide equipment to get on to Twitter, or was the employee cutting corners? If the second – they seriously screwed up BEFORE the mistake.

  4. jamessweet says

    To be clear: I’d rather live in a world where there was no need for that kind of “justice”, and I don’t generally desire that kind of “justice” myself. But I’ve come to believe that the human need for that kind of “justice” is sufficiently ingrained that it may at times be ethically justifiable to pander to it, for the greater good.

  5. jamessweet says

    @moarscienceplz, I think you missed some important details to the story. Somebody who was unhappy with US Airways tweeted the picture in question at them first, and then the employee in question went to flag the tweet as inappropriate. Immediately afterwards, (s)he attempted to retweet a different tweet, and… that happened.

    My guess is that it was some pretty ordinary multi-tasking that resulted in a mis-click. “Respond to A”, “Retweet B”, “Retweet C”, “Respond to D”, “Flag E”, “Respond to F”, “Flag H”, “Retweet I”, etc…. became “Retweet H”, “Flag I”. Or something like that. Oops.

  6. moarscienceplz says

    @#7 jamessweet

    I don’t know much about Twitter, but that is not the take I get from the story Mano linked to. Maybe you saw some other story with more info than this one has?

  7. says


    Just to address your 1): BYOD is only becoming more prevalent, despite all the security issues which were never really addressed (by corporate culture at least) before BYOD became a thing. So, it wouldn’t surprise me if it was a personal device, relevant or not to this particular incident.

  8. Arlen says

    I work as a retail supervisor, and I get people that demand employees be fired over simple, easily remedied mistakes on at least a monthly basis such as forgetting to scan a rewards card or counting change incorrectly. Usually the more the customer complains and less reasonable they are in demands/complaints the less likely I am to do anything or even investigate the problem. The only circumstance in which we’ll fire an employee after the first issue is if they get busted for selling alcohol/tobacco to a minors.

  9. Mano Singham says


    I am really glad that you take that humane and enlightened attitude. The customer is not always right.

  10. leni says

    I had a job in an industry that was harsh about mistakes, and mistakes were visible to pretty much everyone (no, I was not a surgeon!). We handled thousands of samples a week (sometimes in a day) and we would obviously periodically make a labeling mistake or something that might require rework. Sometimes bigger mistakes, but in those cases you could usually identify a hole in the process and close it.

    The thing is, we had an electronic system for cataloging our mistakes because the strict documentation guidelines (cGMP for those familiar with that wonderful world, pharmaceutical manufacturing for those of you who aren’t). You wrote the date wrong yesterday but only discovered it today? Document it in the system and provide evidence or witnesses for the fact that you did indeed merely write the date incorrectly and did not back date! All that is good, don’t get me wrong. It provides a lot of necessary transparency, traceability and accountability. It also meant, at least if you were a supervisor or study director or in QA, that you could get several of the damn things a week, mostly for really petty things like this:

    Entry1 (Guilty chemist) ” “Opened on” date was entered incorrectly on lot ABCXYZ, bottle 12345 of test chemical ABC. The date was signed as 29Apr14, but should have been 28Apr14. Please refer to the attached test data sheets documenting that this bottle was used for tests on 28Apr14. Additionally, all test data packets from 27Apr14 were checked and contained only references to the previous bottle (12344). The original bottle label was corrected and contains a reference to this eNote.”

    Entry2 (Supervisor):”Approved. Data packet review from prior day is sufficient. No other action required.”

    Entry3 (Study director): “Approved.”

    Entry4 (QA): “Approved. No other corrective action required.”

    Ugh, it was horrible! I’ve had to correct signing my own initials wrong lol. If it were a regular office I could just pull out the white out and hide my stupid mistake from the world, but no, not in cGMP land! There you must share every small embarrassment with several of your coworkers, your supervisor, your customer, and possibly the government 🙂

    So it also means that even small, inconsequential mistakes are visible to everyone, which can understandably leave a bad impression. We did a six sigma project to determine our error rate (vs the number of samples we handled) and found it was only something like .9998%. Yup. That was too high. Perception, perception! Nevermind that 20 people handled 150,000 test samples multiple times over the course of 3 years.

    Ugh. I’m glad the rules are there and I’m glad that was my first experience because I would not want to learn lazy habits and then move into that setting, but man I am so glad I don’t do that anymore. Life is so much better when job expectations are realistic and improvement is actually achievable.

  11. leni says

    Er, lol, I think that should be our error rate was .0002%

    (Sorry- I never took statistics and deeply regret it.)

  12. leni says

    You are very kind, Mano, thank you, I appreciate that 🙂

    I forgot to mention in all that, the reason I chose that example was because a coworker there received a formal reprimand for not doing what I described. He corrected his mistake on the bottle label, but didn’t document it in the Formal Fuck-up E-Catalog. To give it context: we were having a problem with that customer, it seemed like an all too easy opportunity for management to save some corporate face by writing him up. And they did, gleefully. Despite the fact that he was good employee with a good record and the mistake was pretty innocent.

    I understand why (perception, perception!), but at the same time we were doing end-phase testing for large, well-funded pharmaceutical companies who don’t want to hear that anything is wrong with their product. Any aberration was a Big Huge Potentially Contract-Breaking Deal. If they could blame it on the contract lab, you can bet your ass they would.

    Ugh. So glad I’m not in that industry anymore.

Leave a Reply

Your email address will not be published. Required fields are marked *