Need insight into the current situation
Posted by want-Dreams@reddit | ExperiencedDevs | View on Reddit | 15 comments
I developed a functionality to support the requirement of another team. They had discussed one of the negative use case that can happen and I took care of it. But my product manager said we should not worry about it and let the customer raise an issue and then we will fix it. He said this has a very rare chance to occur in production but the team which gave this requirement will definitely test it.
He is a good product manager, has a dev experience of more than 15 years and is very involved. Actually reads the code, tests application regularly. I have always seen him to be very level headed and with good inputs.
But I am struggling to understand his pov as to why should we wait for customer to raise an issue when we have already developed a code to take care of it.
Please provide some insights on this.
datacloudthings@reddit
Product managers are in constant negotiations with stakeholders over prioritization. Different stakeholders (who all underestimate the difficulty of the feature they want, and say things like "well, it's just one little thing" about everything) battle it out with each other via the product manager to see what gets prioritized and the product manager needs to keep them all reasonably not unhappy while not swamping the dev team with too much work and too many commitments caused by caving in to stakeholder demands.
Which exact piece of the crazy Jenga puzzle this product manager is holding up I can't tell you, but I would NOT assume they are acting from either ignorance or malice. They are probably holding a world of pain at bay for you.
What you DON'T want to do, in cases where there are far too many stakeholder demands ever to fulfill, is have your roadmap suddenly change because some developer went ahead and built something on their own that the regular product process didn't prioritize. And it sounds like that's what may have happened here.
It is perfectly understandable to me that the product manager may decide to keep this feature in their back pocket and have it ready later when it does surface through the prioritization process.
I am curious actually about how you worked on this without it being prioritized. Was it in your sprint? Or in the backlog? or you just went out and did it after a meeting with another team? Because I REALLY don't like engineers picking and choosing what to work on after having random conversations with random stakeholders. That is how roadmaps get frayed.
Kuliyayoi@reddit
Haha I've done stuff like this. Basically he's playing chess and trying to be 5 steps ahead of his "opponent". He's trying to generate leverage. Small things like this add up and he can use them in the future to continue guaranteeing comfortable workloads. It's all a resources game. With how much experience you say he has, you should honestly just listen to him. He's playing the office politics game for your sake.
intinig@reddit
My .2c, from the point of view of someone that has been on both sides of the PM/Engineer barricade.
A good product manager will need to make hard decisions like this one. Deciding to postpone fixing a known issue because the projected impact is not big enough to be worth the investment _at this point in time_.
There's always a million important things to juggle, and knowing what to prioritize is what makes a good PM.
I don't know your PM personally, but from your description, and the description of this situation, I have a very good opinion of them now. :)
want-Dreams@reddit (OP)
Yes I understand but the code was already developed and delivered. He was hinting at reverting it back stating the reason I stated in the post. He didn't outright ask to do it but in every discussion it is coming up and I am struggling to understand his pov. I even made the changes he suggested. My manager is also a bit lost on understanding why he wants us to revert it back.
Hziak@reddit
Two thoughts (playing devil’s advocate) - there’s a dollar and effort cost associated with QA/UAT that he doesn’t want to pay. - there’s a risk associated with prod deployment that he doesn’t want to take.
Could be either of these if the edge case is very small. That said, it could just be wrapped into the next set of regression tests, idk. Hard to say the LoE required to test it without knowing the problem/fix…
foreveratom@reddit
That still doesn't make much sense. OP specified that the upstream team is already planning to test that negative case.
OP also mentioned that the manager said "this has a very rare chance to occur in production" which means it will happen (because we all know that if something can go wrong, it will).
Since the code is still written and someone has already committed to test it, the cost of removing the code and not deploying it will be higher than doing basically nothing...
Hziak@reddit
I was assuming it was on another branch and the cost of “removing it” was actually just not merging it in.
Acknowledging that they were already planning to test it, maybe now they freed up some bandwidth to test something else?
Unfortunately, I still have a trade modicum of intelligence so I can’t really get fully into the head of a career manager, so I’m just kind of shooting at the hip at possible things they might be thinking. IMO, it sounds like a slam dunk, so really your guess is as good as mine…
dusnik@reddit
There can be many reasons we couldn't possibly know. Maybe he was specifically ordered to not spend time fixing this, and if your solution is included in the release log he might be called out
utihnuli_jaganjac@reddit
Extra money
want-Dreams@reddit (OP)
Can you please elaborate?
utihnuli_jaganjac@reddit
Once the initial requirements are completed and they accept it, every additional change is charged $$$
want-Dreams@reddit (OP)
Please correct me if I am wrong. But this would be for service based companies right? I work in a product based company. The product is ours, and the new functionality is also ours, not something asked from the client. What you have stated, would it be there for product based companies too?
overdoing_it@reddit
Are you sure he understands that the code is already written?
want-Dreams@reddit (OP)
Yes, I showed him. He went through it line by line and asked questions.
overdoing_it@reddit
Well then I would just ask how he wants to proceed with that in mind, remove the code, disable it somehow, or what, then follow up with asking why.