We can write down a useful decision and still make it difficult for others to use it. I came across this the other week and I wanted to share with you what I've learned as well as the rules of thumb I came up with to avoid problems going forward.
One sure way to limit a decision record's usefulness is to be overly narrow. With our mood chart example, it's when we name the exact instance where we made the decision. Another is to make it so broad as to apply everywhere when it really shouldn't.
For me, the applicability section in the decision record template is critical. This needs to sit somewhere between these two extremes.
How can you tell you've messed it up?
If you're describing where the decision happened, rather than why it mattered, you've gone too narrow. For example, "applies to the weekly mood chart." That may be correct, but the sleep team will ignore it since they're not building a mood chart. So the rule of thumb is, if you name it, you hide the problem.
If you on the other hand remove the conditions that made your approach useful, you've gone too broad. For example, "charts must include a summary and individual entries." Technically, it's the correct solution for the problem you solved, but not all charts everywhere serve the same purpose. So the rule of thumb is, don't turn solutions into requirements.
So how do you write the applicability section?
Start with the user task, then describe the conditions the decision depends on. If it applies to charts where people need to understand changes over time and inspect individual entries, say so, without nailing down requirements or restricting its usefulness. Then name the important limits: situations where the task changes, where the evidence no longer holds or the approach could mislead someone. There's no need to list every possible exception, rather try to preserve the differences that would change the decision.
Give the people reading the decision a reason to consider the approach and let them make up their own minds.
And for the litmus test.
Can another team read the scope and explain why their situation fits or what it needs to reconsider?
If only the original team can use it, you've probably been too narrow. If nobody needs to think before applying it, you've probably been too broad.