Long story short, last Friday my boss told me we needed to release a new version of our ship application as soon as possible that day, without previous noticing. I told him it wasn't properly tested yet, and since the last changes I made were basically "prone to errors" (especially because they were microoptimizations I couldn't test enough and could break the code easily).

I encouraged my boss to postpone the release for Monday, since if there was any kind of error, I wouldn't be able to fix it. Or at least, place this update into a beta stage, where only I am able to test it. However, the way beta and production stages work with mobile apps means we had to be waiting there a lot more than expected, so he wanted to release it straight to production.

After that, he also told me it was pretty important I finished another task before leaving, so I had one hour overtime on Friday to finish a task for Monday morning.

Obiously, the app crashed. It crashed right after I left the office on Friday, and our ratings went down a lot. He saw this on Saturday and called me at home, after lunchtime, asking me to guide him on how to solve this on the phone. I wouldn't say he isn't capable, but I was totally unable to do that from memory, so I had to go to the office and spend another hour of my free time there, fixing an issue I encouraged him to avoid.

Also, during today (Monday), he's come to my desk, asking me to keep the same schedule we had last Friday, but also applying more improvements to this. This means that I won't have my two hours off (since I already worked on them during Fri and Sat) today, and probably not tomorrow as well.

I have a bad taste since it happened the same to me in my previous job: the boss felt fine to make me work overtime without paying the hours or letting me be off earlier, since there were more tasks everyday. Worse than that, despite forcing me to come to the office on a Saturday, he didn't apologize in any way (for either calling me, or having taken a decision that force both of us to go to the office on Saturday). Instead, he kind of put more pressure by tightening the schedule even more.

In a situation like this, which will definitely end up burning me, and where the boss clearly wants to rush even at a cost none of us is willing to pay, how can I tell him to, let's say, stand back and trust my expertise?

I'm in Spain, hired for 39 hours/week, no paid overtime (instead, any extra hour done is compensated with an hour off in the following days).

  • First, I don't understand why you are "burning me" for a couple of hours extra, assuming you are salaried. It is to be expected...OCCASIONALLY. If it becomes a habit then that's another story. Also, if you aren't willing to release code you've worked on then simply say "It isn't done yet". You don't need to say that you want to test it before releasing it. If you haven't tested it yet then it isn't done. So IMO, you caused the problem just as much as your manager because you made him think that your changes were done. – Dunk Dec 15 '14 at 18:42
  • Hi @Dunk, thanks for the comment. Well, it's not been only those hours, but some more (unrelated to the question). Unlucky myself, as I mentioned, I was doing microoptimizations, something he already knows, which means I could apparently stop whenever we want. If I was to say "it's not done yet", the answer could be "then leave it as is and upload now", which is not the best thing I could do. That's why I try to show him what I'm up to now (testing, in that time). Also, he's seen before that releasing without testing leads us to this situations. – Korcholis Dec 16 '14 at 11:44
  • Korch:Your job is to deliver code that works. If you want to do micro-optimizations to your changes then that is fine. However, you should first incorporate your changes then test them to be sure they work. After that create a baseline of code that you know works. Then do your micro-optimizations. If your boss insists on you putting your changes in the main build before you have tested your micro-optimizations then give him the build that you have already tested without the micro-optimizations. Do not ever give code that hasn't been tested that will be delivered to the field. – Dunk Dec 19 '14 at 18:17
  • BTW The Workplace is full of questions regarding deadlines – user8036 Dec 21 '14 at 13:03

First, you need to clearly understand what you can and can't control. For example, you can control whether you take calls on a Saturday, come into the office on a Saturday, and so on. You can't control what your boss asks you to do.

Second, you need to entirely abandon the approach in your bolded final sentence. Your boss is "the big boy". There is no way you can tell him to stand back and let you do it without his input. No way. If that is your goal, give up now. But let me suggest a better way. You want him to understand the business consequences of doing things the way he has been. Perhaps that will lead to different decisions from your boss.

So, while the misery of the weekend is still in everyone's minds, have a "lessons learned" exercise. Ask yourself whether some of these things would be good to keep doing, or not:

  • deploying late on a Friday
  • adding more items to a delayed project phase and expecting not to delay it further
  • rushing things into production before they're tested

Now to you it's obvious that those are all terrible things to do. But your boss wanted them done. Don't assume your boss is an idiot. There may be tremendous first-mover advantage or some other reason behind these requests. Say there's a 50-50 chance the app will crash and your ratings will go down if you release without testing. Maybe your boss believes there's a 100% chance your ratings will go down if another week goes by that the competition has feature X and you don't. So your boss decides to take the chance since there's actually nothing to lose. (And of course, there's even less to lose if the boss knows Korcholis will come in on the weekend and make everything ok again, right?)

So understand what motivated the boss to ask you to do these crazy things. Help your boss understand why you weren't surprised it crashed. Talk openly about avoiding overtime, weekend emergencies, and so on. Reach a point where you know why these pressures and demands are coming to you, and where you boss knows why you push back on some of them.

And next time the boss asks for a late Friday deploy, you can include a warning in your push back: I'm not available at all this weekend, from 5pm when I leave here until Monday morning when I wake up. (Don't say why.) I may not be able to even answer my phone, but if I do, I won't be able to talk long or come into the office. Let's not take the chance you might need me over that time. Surely it's wiser to wait until Monday morning?

The decisions are not yours to make; that is the role your boss is here to fill. You can provide the relevant information and you can control your own free time.

| improve this answer | |
  • 3
    Sometimes it's easier to make these fights with numbers, as you're not getting paid for this extra time (as most salaried people aren't) this direct cost isn't as easy. In this case the app broke, how many people were affected, is there anyway to gauge the revenue lost during the outage? are we talking hundreds, thousands, tens of thousands lost over this? If you can figure out the impact you have a lot more ammo that Friday evening when the boss says "$%@@ do it live!" to say... "you know boss last time that cost us ~23,000$ we should probably wait until Monday" – RualStorge Dec 15 '14 at 18:24
  • It is not a new release until it runs. Dropping untested software in production does not speed release. – paparazzo Dec 15 '14 at 19:04
  • First mover advantage to make a bad impression on the customers is no first mover advantage, especially since bad impressions tend to be lasting. The boss could not even stand a weekend's delay, so this kind of rush has to be concerning. I agree that the boss could not care less about the impact of the way he does things on Korcholis, so any argument has to be centered on something the boss cares about - in this case, the impact on his business. – Vietnhi Phuvan Dec 16 '14 at 10:47
  • In our case, the app is absolutely and completely free, no ads, no subscriptions, no initial payment. In fact, it's, to my opinion, more like an experiment for now, but he wants to give it other goals in the future (really far future). He doesn't know how to monetize it, so there are no numbers in this case. Only ratings, user comments and installs/uninstalls. He's seen, however, that a bad release reduces our ratings by 0.03 points out of 5 in one day, while a good one gives us 0.01 every two or three days. – Korcholis Dec 16 '14 at 11:49
  • Heh... after 2 years, @KateGregory, I clearly see the fails of my reaction. Here's comes the tick and a +1 – Korcholis Jan 12 '17 at 14:39

Not the answer you're looking for? Browse other questions tagged .