1

Summary: boss asks for automation of task. Doing it manually will cost 30 developer-days of work, but creating custom CMS to automate it will cost 120 developer-days. How can I bring this to manager's attention?


I want to alert my boss to the fact that he maybe isn't using his developers (me being one of them) in the most efficient way. Basically he's asked us to build a completely custom CMS with a whole bunch of features and we've scoped out 2 months of time for this. That's 3 developers, working on a 2 month project.

The CMS is used to form a survey/quiz that can be selected or assigned at user creation. Now reasonably speaking, the company would make 30 of these surveys, at most throughout the year. With everything that's going on our director has actually said to expect a downturn of clients.

The surveys themselves aren't difficult to create. I could hardcode one in a day, maybe a day and a half easy. It's the automation that will cause the time sink and making sure those in the company who'd use the CMS would be constrained in what they can/can't do. Not to mention that 30 would happen over the course of the year, not all at once, and 30 is being really safe.

So that's 30, maybe 45 days of work vs 120 days of work upfront (not to mention maintenance & patches) being spent on an infrequently changed feature. Our plates are full too. There's much more stuff we could be working on.

That being said, I don't want to come off like I know the job better than him, especially given that he's likely privy to information that I'm not. I want to more so approach him with a food for thought mentality and let him make a decision.

  • Can you meet your manager halfway? Implement some minimal system that can handle the first user survey, but then can have more surveys added in easily? But doesn't have all the bells & whistles? So maybe a couple of weeks worth of work that then pays back as future surveys as more easily implemented? That way you can grow the system as needed. – DaveG Apr 8 at 0:13
  • good illustration of the problem: xkcd.com/1205 – aaaaa says reinstate Monica Apr 8 at 1:16
  • what are you trying to achieve? – aaaaa says reinstate Monica Apr 8 at 1:18
  • When you create a CMS for this task, how much time will it take the next time you need to do the task? Maybe that is why the manager is asking to automate it? Because it needs to be done frequently – Khalil Khalaf Apr 8 at 1:48
  • Why is this even a problem you feel a need to solve? – Kilisi Apr 8 at 8:32
5

To answer the direct question:

It's reasonable to simply approach your boss and casually ask what the bigger picture is. You can mention that you could create four year's worth of surveys in the time it takes to create the CMS, but your boss is probably already aware of the situation. There's likely a reason for the decision, and it shows initiative that you're interested in understanding the rationale.

Potential reasons to build a CMS:

  1. There may be greater demand for surveys in the future
  2. Surveys will be more difficult to produce in the future, so automating the task now will pay off more in the future
  3. The survey data needs to be standardized and integrated with another system, and the CMS will facilitate standardization
  4. The boss wants non-technical employees to develop surveys
  5. This CMS is an "easy" pilot project to decide if it's worth automating other tasks
  6. The boss needs a flashy automation project for his next performance review
  7. The boss thinks that developers get bored of manual tasks, and is gifting you a "fun" project instead
  8. This is something that has been desired for a long time, and this is the first opportunity to achieve it
  9. Everyone technical will be laid off next year, and those who are left won't be able/willing to make surveys
  10. (Unlikely) The boss isn't aware that it's a waste of time
| improve this answer | |
  • Another explanation is that delivering a CMS is a potentially positive item the boss can use at their next performance review. Doing things simply and efficiently is not the way to get promotion and bonuses. Delivering complex ‘innovation’, even when wasteful, is what is rewarded in many companies. – jwpfox Apr 20 at 22:25
  • 1
    @jwpfox good one! I've added it to the list. There's often a disconnect between perceived value and actual value when it comes to automation. It works in reverse too; sometimes management doesn't want to invest short term profit into much greater long term savings – Thomas Zwaagstra Apr 22 at 8:15
0

Will this be used for a single year, and then it's done? Are you going to be doing an interface that some rando off the street can use?

You should take a day (or 2 - 4 hours) to go over the business requirements for the project, understand the drivers, the lifetime, and the goals before thinking about shooting down someone's plans, especially your supervisors. Maybe your CEO wants to make surveys and requested that this be done. Maybe the plan on ramping up the volume of surveys? That 120 (360 person-hours) hours of dev time might end up saving a lot more than you'd have to spend hardcoding them. In addition to that, you might not get interrupted in your other work to write a survey again, and again, and again. 3 years of surveys, interruptions, and having to rely on some dev to hardcode them time after time? Seems like there is a value proposition here worth looking at.

| improve this answer | |

You must log in to answer this question.

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