Rootdev

You want the Moon on a Stick.

Or: Project Management the Wrong Way

Having spent the last 6 months working in Project Management after being thrown in the deep end, the best advice I can give is make sure your requirements are signed off before you start.

It all started so simply. I was taken on as a technical specialist on a project and given a simple brief with a clearly understood, seemingly straightforward requirement. When I took over management of the project very soon after I discovered that there was no supporting documentation and even though everyone knew what the requirement was, it had never been written down anywhere.

In reality, this project should probably never have been given funding. As it was, despite the fact that I had initially been asked to undertake the product evaluation phase of the project, I had to abandon that and go back to square one, creating the supporting documentation that should have been prepared in order for the project to be given the green light in the first place.

And that’s where the trouble started…

Requirements are Everything

Everything depended on the requirements, and even though everyone agreed to the requirements verbally and in principle, no one would agree to them in writing. And since no one would agree to the requirements in writing, none of the other documentation could be completed.

What ensued then was a couple of months of shifting the goalposts. Every time I thought I had pinned down the requirements and got agreement from the key stakeholders they would suddenly be changed again at the last minute.

With deadlines looming, we decided we had no recourse but to push ahead with the product evaluation regardless (4 months after I was initially made to call a halt to it). We arranged a two week evaluation period and stressed that the scope of this evaluation was limited to out of the box functionality only. But of course, we still had no sign-off on the requirements, so it became open season on scope creep.

Enough already. After calling yet another group (read ‘crisis’) meeting, and yet again getting verbal consensus around the table, I put my foot down and got assurances that each department would forward their requirements to me in writing.

Three months after I first asked (and only one department bothered to respond), I finally got written requirements from the other departments. And from the team with realistically the least vested interest in the outcome came the most off the wall list of requirements I have ever seen.

MOAS

From the simple premise of “What improvements would you like over the current system?” came the response: “Please give us the Moon On A Stick”.

Underlined with the proviso that they would refuse to sign off on the project’s deliverables unless all of their requirements could be met.

So, what was supposed to be a two week product evaluation turned into a 5 week long endeavour to prove that even though the Business Requirements had finally been accepted and signed off, that if something was impossible it could not be held up as a reason to not accept the outcome.

The purpose of the evaluation was to find a ‘best fit’. If we could prove that the base requirements could be met, along with some of the more advanced requirements, then this should be documented and used to seek approval to proceed to the deliverables phase. If not, it was going to be time to pull the plug.

Even after I finally called a halt to the evaluation and presented the results in a wrap up meeting, where to my surprise I actually got agreement to go ahead, I was still being asked to consider other options and to allow yet more shifting of the goalposts.

All’s well that ends…

Thankfully by this point it was too late. The requirements were already signed off and the outcome had been accepted by the project sponsor, so I was free to proceed and hopefully put an end to the whole process. Free to deliver an outcome that nobody really cared about in the first place and which none of the key stakeholders had any vested interest in. But at least I could be happy that I had done my job well and delivered what I had promised. On time, and under budget.

Next time

Despite everything, this experience has proved incredibly useful. Next time I will do things differently - starting with demanding to see signed copies of the supporting documentation and the _signed_ business requirements before I start.

And if you want the Moon on a Stick? Well I’m afraid you just can’t have it.

With thanks to Lee & Herring