I recently taught a workshop for a client where I gave the participants a good news/bad news scenario. The good news, I suggested, is the documentation work practices I observed in their company were like what I saw in other pharma companies. I then suggested the bad news is their work practices were like what I saw in other companies. My point to them is that the bar for sophisticated work practices is set pretty low in the pharmaceutical industry. I used as my line of reasoning our McCulley/Cuppan modification of JoAnn Hackos’ Document Capability Maturity Model (DCMM) she presents in her book: “Managing Your Documentation Projects.”
To this group I suggested their documentation work practices would come out at about a 2 on our 6 point scale, which suggests also the point on the scale I will place most of the documentation groups in the pharma industry. Most were surprised at the low ranking and a few were even offended I think. By the end of our workshop, all were feeling different about what it takes to create high quality technical documentation in the life sciences.
I’ll characterize here a documentation project undertaken by this group I had in the workshop and suggest such work is emblematic as to why I give them (and the industry) a low DCMM score.
The group generated a briefing document to be submitted to a regulatory health agency in advance of a request (a standard practice across the industry.) The briefing document is intended to inform the health agency as to what topics the company wanted to receive feedback from the health agency regarding an ongoing drug development project.
The company had an hour meeting with the agency and ended up wanting to address 8 topics within the hour (an ambitious agenda.)
The documentation team went through a three draft process to create the briefing package where the number of questions and the actual questions changed w/ each draft. I mean they were truly using a “fire; aim; now we are ready” approach to create this briefing package. The approach of: "activity is everything, so we are way too busy to plan."
Each draft looked dramatically different than the previous version and required massive investment of time and energy from the team. In the end the briefing package was 83 pages—that is a little over 10 pages per question. The first question did not appear until page 25 (which means there were 24 pages of introductions and throat clearing before we get to the meat. Curiously to me, none of the questions appeared in the Table of Contents of the briefing package and the document did not have an Executive Summary (we want you to work as hard in reading this document as we did in crafting it.)
Then I asked the authoring team what they did as a documentation project post-mortem for lessons learned, such as solicit usability feedback from the customer using the document or to identify how the documentation work practices impacted the process and the product. The answer was, “Oh we did not do anything like what you suggested. The document must have worked because we had our meeting with the agency. We just moved on to the next document.” I asked if they got what they wanted out of the meeting. The answer was a categorical no. “We did not get an answer to our most important question as we had placed it as 8 of 8 in the queue. We ran out of time to address that question.”
I asked how uniform their documentation practices are for the creation for briefing packages (they will generate such packages numerous times in a given year for various development projects). Response was “We have no set approach, so each briefing package will be generated utilizing variable methods and processes.”
How they could be surprised with a low rating for the sophistication of their documentation practices is the interesting question for me. Where comes the comfort with such outcomes? In part, I think, it is due to erroneously assigning the sophistication in conduct and analysis of science as carrying over to documentation work practices.
Originally published on our Knowledge Management blog
No comments:
Post a Comment