Sentiment analysis and project debriefs/lessons learned

This post by Giles Palmer at Brandwatch (here) hits on something I’d like to try with our increasing volume of project debriefs.  We’ve done a couple of analyses of project success and trouble factors (I’ve posted on them a number of times, especially here, here, and here), but I’d like to take it to the next level.

We haven’t had a lot of success having our project managers analyze the debriefs.  First, they’re not exactly expert media analysts.  Also, because it is “their” domain, they’re very invested in the topic.  Everything seems important to them, so we end up focusing on what are essentially “neutral” responses and issues.  It appears that most PM’s “wetware” doesn’t discriminate finely enough.  And frankly, the opportunity costs of having PMs do this kind of work are substantial.

The sentiment analysis technique seems like a great filter to ID those questions that provoke the most positive or negative responses.  Once we go through that analysis, we can then cross-tab on the debrief questions, industries, solutions, projects, etc. that provoke the strongest responses.  Only then would be bring in senior folks to refine the analysis.  Furthermore, debrief questions that prompted a lot of neutral responses may need to be reworded or removed.

I like the idea of cross-pollinating across functions, and this idea seems promising.  I hope to pilot this approach later this year and early next…

%d bloggers like this: