Once Bitten, Twice Shy

Great White flashback to 1987 – Is he wearing a robe?

To my wife’s dismay, this is not a post about the 80’s hair band, Great White.

(Thank goodness!)

This is about project communication and estimation.

Mark had done his best. Being new to project teams, he tried to estimate his tasks as well as he could back when Jack (the project manager) asked.

Now it was several months into the project, and was becoming painfully obvious that Mark wasn’t keeping on schedule.

After a particularly painful project status review, Jack called Mark into his office.

“Mark, I expect you to perform at the level you committed to. There is no excuse for this, you are making our entire project team look bad. Your tasks were nowhere near the critical path before, but now you ARE the critical path.”

Now that’s some whiz-bang project communication right there.

Mark walked out of Jack’s office, dejected.

After struggling through that project (Jack went out and got an extra person to help with Mark’s tasks, which resulted in another lecture and brought the project in over the original budget) it was time for another one.

After the whupping he took last time, Mark knew what to do. This time when doing estimates, he remembered all the things that got in his way last time he hadn’t anticipated. He came to an estimate on each task. Then he doubled it.

Sue was running the project Mark was on this time. Jack stopped Sue in the hall one day as they were passing by each other.

“Say Sue, you have Mark on your new project, right?”

“Yep. Why?”

“If I were you, I’d take his estimates and double them. I went through hellon my last project because his estimates were way too low.”

“Wow. Thanks for the tip Jack!”

Of course this is a simplified caricature, but this short story illustrates some key issues that occur on projects often, to a greater or lesser degree. As a new project manager, this kind of scenario is something to watch out for.

  • Group estimates could have been used to get clear on scope and assumptions and bring in more experience
  • Instead of chewing out Mark, Jack should have sought out the root cause and took responsibility upon himself
  • Hopefully, Sue will use the information Jack gave her to sort out the root cause with Mark and ensure solid estimating practices are used.
  • Is Sue’s sponsor going to pad the estimates even further?
  • Will Parkinson’s Law come into play with these heavily padded estimates, so that they will be a self-fulfilling prophecy?

The worst part is probably that Mark has been bitten, and in the future will be very unlikely to trust a project manager who might want to implement something like Critical Chain Project Management, where padding is removed from individual tasks and it’s OK to go over on individual tasks, because you are actually shooting for a 50% chance of completion on time.

No, a barrier has gone up between Mark and his project managers.

Once bitten, twice shy.

Do you have similar stories to share? How about additional comments on this scenario?

Leave a Comment


Leave a Reply

Andrew Krzmarzick

I know you intended for this post to be about project management, but I am definitely seeding a Grooveshark radio session with Great White this afternoon. 🙂

On the scenario – yikes! Measure twice, cut once. We often want to jump into projects, but I think it’s important to take a little more time on the front end to avoid these raps on, uh, the back end.

Josh Nankivel

I think I disagree with you on the sentiment about taking more time on the front end Andy. Getting better at estimates definitely. But I think it’s important for us to realize that an estimate is an estimate, and it will likely be wrong. We hope that the balance of things taking longer than expected and those shorter than expected balance out, but of course it leans towards taking more time than less because of Parkinson’s law.

I’d actually like to see us rely less on estimates and embrace the uncertainty that comes with project work. We should build in the contingency and error bars on our estimates so even if the estimates are missed, we still have the budget MR to manage uncertainty and what happens in the real world.

Faye Newsham

If you employ decent process management, your estimates shouldn’t be that far off, even for a newbie. There should be a chart of basic team assumptions and a tailoring metric for new vs experienced folks. The manager should definately know that a new team member will be the least accurate with estimation and to expect slippage. Padding isn’t a good idea as it builds false expectations and pads project expenses so your proposal team will never get the numbers right (you just won’t be competitive)…assuming these numbers are fed back into the proposal work (which I sure hope it is!). I disagree with @Josh – there shouldn’t be that much uncertainty unless you are doing new, unique work. CMMI and ISO9001-2000/2008 are all about controlling the unexpected. There is little to no reason for it for companies who know what they are doing and support their staff with reasonable tools (templates, process documents, measurement tools, etc.).

Josh Nankivel

Thanks Faye. As you pointed out, the more operational and repeatable the project work is, the less uncertainty there should be. The same goes for organizational maturity.

Conversely, when managing projects that are very unique indeed, especially with brand new software and cutting-edge hardware solutions involved, there can be a lot of uncertainty injected into the mix. There’s a big difference between padding and management reserve, and I’m advocating the strategic use of management reserve, not padding built into estimates. That’s exactly what we want to avoid, but what we get when the scenario described in the post occurs.