coneblog

The Pharma client called with an urgent project need. We set up a meeting to review the situation. At the meeting, it became clear that they needed a replacement for the paper-based tracking system that had become a target of an FDA review.

The new system had to improve on the functionality of the paper-based system (not hard!). The new system had to produce the reporting currently developed by hand for Management. The new system had to be FDA Good Manufacturing Systems compliant (that’s CFR 21 part 11 compliant to you Pharma wonks). And they need it in eight weeks.

Well, I don’t think that you can get “Hello World” designed, developed, validated, thru FAT, IQ and OQ, and in production in eight weeks. But I didn’t say that in the meeting. I expressed optimism for a rapid, high-quality deployment, and said that I’d turn around our estimate and proposal in a week,

Like most PMs, I’d like to have a comprehensive, hi-quality spec on my hands in order to estimate a job like this. I didn’t have any such thing. Because of the way this client approves and spends money, the project had to be done on a fixed-price basis. So I had a week to come up with an estimate and proposal for this job.

I had a few things going for me. I knew the technology set: VB client with an Oracle back end – pretty classic client server. We suggested a browser-based client, but this was a few years ago, and they had a lot of infrastructure in place to deploy and manage client apps, so client server it was.

Plus, we had validated and installed several shop-floor systems for this client before, so I knew what the client required in terms of documentation and testing. What I didn’t really know was the scope of the system