How loudly should I warn about a failing project?

Question

A few months ago I was asked to estimate the staffing and the time needed for a proposed new project. I estimated it as needing 10 staff over 5 years. A co-worker estimated the project as needing 7 staff over 9 months. Management decided to go with the 9 month estimate.

A project manager was appointed who had a huge reputation within the company, having successfully guided many projects to completion. He was totally on-board with the 9 month estimate and promised that it would be complete within 9 months.

I was brought on board as an internal consultant at 0.1, but spend 90% of my time on my old project. The work was broken down and an extremely detailed schedule was produced. I pointed out that a component that had been estimated as taking 8 hours actually had 200 features requiring the engineers to complete a feature every 2.4 minutes. They re-estimated it as 80 hours which is still too low. However the total estimate of the project never changed. Every time I pointed to a problem area they would reduce their estimates on the areas I wasn't complaining about so that the total stayed the same.

The senior manager asked me into his office and asked me whether I thought the schedule was unrealistic. I told him the extra staff had already been hired, the equipment bought and suggested a 6 week trial to get a clearer idea on a realistic schedule and that after the 6 week trial we should reevaluate the project. The project managers schedule said that the first module would be completed at the 4 week mark.

At six weeks the first module hadn't been finished. The first sub-module, the first sub-sub-sub-module hadn't been completed. Not a single feature had been finished.

I talked to project manager. I asked the project manager about the progress of the project. He said the team had produced a lot of work. I pointed out that the methodology that the company used did not count partially completed work and if you calculated the projected completion date the way you were proposed to calculate it the result came out as never. He said that things were more complicated than that and that the project was actually on track. I asked what he was basing that on. He said that he was basing it on his many years of experience.

I talked to the senior manager and pointed out that we were going reevaluate the project after 6 weeks. He said there was no need as the project manager had assured him that the project was on track.

Now I am kicking myself. I should have just told the senior manager that the project managers schedule was delusional fantasy and that he should cancel the project straight away. Why did I have to wimp out? The 6 week trial seemed like a sensible idea at the time, however it was clearly a mistake.

Its now 10 weeks and the team still hasn't completed a single feature. Even if I bend the rules beyond breaking point in the teams favor saying that they had completed 5% of the module that they were supposed to have finished 6 weeks ago is generous. The project is clearly doomed. The project manager still says that they are on track.

Options going forward are

• Build relationships and trust to increase influence
  • However the problem seems to be that my views are inconvenient
• Build more evidence
  • However the existing evidence is clear
• Just talk to the senior manager
  • However the project manager has a god-like reputation
• Just talk to the project manager
  • However when I started listing the things that need to be completed for the project to succeed he exploded in anger, yelling that this must be the largest most complicated project in the world. He made it clear that he thought my views were completely unrealistic.
• Talk to the team and build consensus
  • However the team are all new hires and they put down their lack of progress to teething problems
• Pull some brilliant idea out of my butt
  • However I seem to be short of those
• Distance myself from the project and just concentrate on my main project.
  • However that feels unprofessional.
• Ask 'The Workplace' if anyone has any better ideas

One thing I worry about is learning the wrong lesson from this. A cat who sits on a hot stove will not sit on a hot stove again, however they will not sit on a cold stove either. Obviously I am going to be nervous about proposing a spike or a trial period again. However looking back there were little signs and indications that my managers were not interested in evidence based engineering. The real lesson is probably that I need to know my audience. That and I need to be more vocal and assertive.

So my questions are what should I do now? and The next time this happens what should I do differently?

Answer 1

From your description, you are not the project manager nor in charge of the project. You were a consultant, your job was to provide expert advice which you succesfully delivered. Unless you hold shares in the company, the health of the project is not your business. You had no authority or responsibility over somebody else' works.

Answer 2

This whole situation seems bizarre to me. If I asked two members of my team how long a project would take and they come up with slightly different estimates, then that's OK. If one of my team said 63 person-months and the other said 2600 person-months, I'll tell them to go and sit down in a room together and come up with some explanation as to why their estimates are so immensely different, because they've obviously not estimated the same piece of work.

Similarly, if I've been told that I'll see a first deliverable 10% of the way through a project (4 weeks out of 36) and I still haven't seen anything 25% of the way through a project (10 weeks out of 36), I'm not just going to sit back and believe the project is on track because one person tells me so, I'm going to be asking some very pointed questions and either getting into the details of the delays myself, or making sure somebody else is.

To move forward, I think you need to understand why people aren't raising a whole load of red flags. That's something really only you can do, but when I've been in this sort of situation in the past the most common explanation has been that people are, rightly or wrongly, afraid of reporting bad news up the management chain: often they've staked their personal reputation on the success of a project, and it's very hard to admit they were wrong unless there's a culture in place which allows that kind of failure.

If you still want to fight the good fight on this project, realistically the only thing you can do first of all is to talk to the senior manager again. Express your concerns as well as you can, trying as hard as possible to make anything you say backed up with as much evidence as possible. Hopefully the senior manager will start taking a more direct interest in the progress of the project.

If that doesn't work, the next step has to be to go up the chain to your VP. That's pretty much the nuclear option as it this point you're throwing the senior manager under the bus - but it's not necessarily the wrong thing to do if the senior manager isn't doing his job effectively. Again, only you can really judge whether this is a good option or not.

Or of course, you can ignore the problem. You were asked for your opinion, you gave it, it was ignored. Some fights you can't win, no matter how much you want to. Move on, do the rest of your job to the best of your ability and let other people worry about the hole they've dug themselves into.

Answer 3

You went above and beyond as far as reporting your observations. Just concentrate on your main project. From your description, I'm even surprised you were allowed that far on a side project.

The important thing to realize is that companies must fail some projects. Imagine you were the CEO. You want your company to go as fast as possible. You don't want a complacent company. How do you achieve that? You try various speeds, various stress level on different projects. If they all work, you know the company is too slow, so you push for faster.

(From the description I assume knowledge work, especially like software engineering, consulting. This is less true in more classic or dangerous industries like energy, mining)

See the one-off project that seems completely delusional as a measurement tool for high management. They're checking if it is possible. If it's not, the best you can do is to let it fail so they get the correct measurement.