Since the dawn of time, one of the questions most likely to strike fear into the heart of even seasoned project managers is, “So how much is this project going to cost?” In fact, at Brightwork says there are hieroglyphs on the wall of the tomb of the great pharaoh Khufu, depicting the pharaoh asking Vizier Hermiunu, the pyramid’s project manager, this very question about his burial pyramid and, a few walls down, a second depiction of the project manager being thrown into a nest of crocodiles in the Nile after the project overran its budget by a few thousand debens.
As evidence of how little project management progressed, Mr. Kreha relates how he sat in a questioning-techniques-in-training estimation meeting and watched an “agile” project team assign “points” to “stories” in a vain attempt to estimate how much work they might get done in the next two weeks, aka the next sprint. And inevitably, a new team member would ask at some point, “so how many hours are there in a point?” Immediately, this agile novice is mocked mercilessly! “You don’t understand,” the scrum master and other developers say, “points aren’t convertible into hours or dollars. We use a Fibonacci sequence – you know, 1,2,3, 5, 8, and 13 – to estimate how much effort a story is. It has nothing to do with hours or money.”
And so we project managers are still left holding the bag for estimating projects, often early in a project’s lifecycle, and then being held accountable for them as if we were clairvoyant. What can be done?
Brightwork’s Kreha offers some hints on how to stay out of the croc pond. Start doing them, that might help:
Separate hard costs from soft costs
When you’re estimating. Hard costs are things like license fees. Once you have a quote from a vendor (stall until you have one) you can be pretty confident that’s what the cost will be. Hourly labor, and time and material contracts in general, are obviously softer since you’re funding time and not deliverables per se.
For softer costs, use ‘burn rates’ to look at low, likely, and high ranges for labor costs. For example, if you have a team of 10 with an average bill rate of $100/hour and they will be working on your project more or less full time for the next 5-6 months, you’re looking at $860k to $1M if I did the math right. Don’t get suckered into estimating hours without thinking about time, because things ALMOST ALWAYS take longer than you planned.
Use ranges wherever possible
Early in a project, it at least helps to subliminally communicate to stakeholders that the project costs are still a bit squishy. I am sure we’ve all seen estimates that have line items down to the dollar. Like $365,750.00. That’s a terrible thing to do – it implies a precision that just isn’t there.
Don’t EVER leave out contingency! At the project outset, make sure it’s 25% of the total project estimate. And try your best NOT to tell anyone it’s there. That’s YOUR insurance policy to keep you out of the croc pond
Get estimates from multiple sources if possible. Have a technical team do an estimate. Have a trusted project manager do one. And maybe even ask the stakeholders what they think the project should If the numbers you get back are wildly variant, you have a lot of work to do to rationalize them down to something plausible.
Relentlessly track your actual costs as you incur them! And more importantly, once you see them drifting away from the estimate or any underlying assumptions you made, TELL someone right away. Delivering financial bad news is one thing; delivering financial bad news 75% through a project is PM malpractice.
Figure out who, if anyone, is likely to be joining you in the croc pool. Trust their numbers more than someone who will skate out the side door faster than Usain Bolt if the project costs start going sideways.
rb-
We have all been there, in the croc pond, under the bus, or in front of the train. Someone didn’t complete their task on time or misunderstood a requirement or just screwed up. These suggestions can help insulate you from some of the inevitable problems that are part of being a project manager.
Related articles
Ralph Bach has been in IT long enough to know better and has blogged from his Bach Seat about IT, careers, and anything else that catches his attention since 2005. You can follow him on LinkedIn, Facebook, and Twitter. Email the Bach Seat here.
