Coping with turnovers in a project

[Hu Chong]

I know this is not a new problem and many times you will encounter this:

XXX promised the customer that the system can do YYYY and ZZZZ during requirements gathering. During implementation, XXX left and the new team lead got to deliver on the promised feature. The new team lead found that what has been promised is not feasible to implement. But he cannot tell the customer that. And the management is forcing him to deliver on the deadline. What will you have done if you were him?

[Lasse Koskela]

Originally posted by Hu Chong:
I know this is not a new problem and many times you will encounter this:

XXX promised the customer that the system can do YYYY and ZZZZ during requirements gathering. During implementation, XXX left and the new team lead got to deliver on the promised feature. The new team lead found that what has been promised is not feasible to implement. But he cannot tell the customer that. And the management is forcing him to deliver on the deadline. What will you have done if you were him?

If the truth is that you cannot deliver the feature someone had promised, you should try to find a way to tell the truth to the customer instead of letting him believe that the feature will be delivered and eventually destroy the illusion anyway.

I'd focus on the underlying forces behind this:

But he cannot tell the customer that

[Ilja Preuss]

I agree with Lasse.

I'd probably tell everyone who listens that without close collaboration and *everyone* understanding both what *can* be reasonably expected to be done, and what really is *vital* to be done, the project will be sure failure.

If management still insists on not telling the customer, depending on the situation I might do one of the following:

- tell the customer anyway, and hope that something good follows from it,
- dig out my copy of the book "Death March" and do the project without investing much of my soul and body into it, or
- simply refuse to do the project, possibly even quit.

[Hu Chong]

Originally posted by Ilja Preuss:

- dig out my copy of the book "Death March" and do the project without investing much of my soul and body into it, or
- simply refuse to do the project, possibly even quit.

You are right, most of the project members are doing either of the above.

Some have even gotten "creative" with their language. They will tell you their module can be integrated with yours. When you try using and it cannot run, they will tell you they didnt tell you it can run
[ March 28, 2005: Message edited by: Hu Chong ]

[Ilja Preuss]

Originally posted by Hu Chong:
Some have even gotten "creative" with their language. They will tell you their module can be integrated with yours. When you try using and it cannot run, they will tell you they didnt tell you it can run

Mhh, that doesn't sound like something I'd do, at least not deliberately. I think honesty is a very important value, always. It's important to notice though, that when people experience huge stress, communication problems will invariably increase. Depending on the situation I could imagine that it seemed "obvious" to the developer mentioned above that the module couldn't run (yet?). So instead of blaming the "liar", I'd try to find the causes and remove them. Perhaps you need to slow down a bit and communicate more to get more effective.

[Kishore Dandu]

I have experience of getting yelled at for proposing to create a frontend test page instead of a turnaround until the integration phase. I even worked with folks who does not understand a temporary metadata creator(in the lines of mock objects) instead of depending on deliverable that is expected to be delivered at the same time as theirs.

Basically depending on culture of your colleagues, there can be situation where people will start banging on your door much before the integration timeframe.

[Naina Si]

I agree with all of you,

In case of situations where it is impossible to deliver the agreed functionality, i would try to give possible alternatives to the customer. If the customer likes one of those, would be good for everyone.
Else as Lasse said, its better to inform the customer asap. Just my opinion..