An investigation card is intended to allow a team to spend time looking into an issue before they launch into actually doing it. There are good reasons to do this: you can’t simply charge into every piece of development; some things have to be planned.
But I also feel investigation cards are over-used. It’s often the development equivalent of an unconvincing sick note at the games lesson, excusing the pupil from jumping into the freezing cold swimming pool… until next week. It’s too easy to hold back from working on an unfamiliar area of a system, spending time looking rather than doing.
The question for me is: What difference will the investigation make? If it’s to assess the best design for several stories then it’s valuable, and one piece of work will benefit several others. If it’s to make an estimate that will in turn determine whether or not a piece of work will be done then it’s valuable.
But if it’s just to get a better estimate, then it’s not useful if the work is going to be done anyway. If it’s a couple of days’ investigation ahead of a week of implementation then that’s just putting off the implementation until it’s brought into the next sprint.
Development isn’t just the act of developing. But we shouldn’t shy away from difficult work; sometimes we need to brace ourselves and take the plunge.