I’ve been working on some analysis about the usage of our various mobile applications for Basecamp in advance of our next company-wide get together. As I’ve been going through the process of gathering and analyzing various sources of data, I was reminded yet again of a fundamental question you have to ask yourself before you can really do anything with all that data. Are you writing a reference book or are you trying to answer a question?

Reference books like encyclopedias or statistical abstracts contain many facts about a topic with little or no higher level analysis. The point is to take raw data and summarize it in a form that’s easier and faster to refer to for future analysis, not to break new ground. They’re wonderful to have (I happen to have copies of The Statistical Abstract of the United States ranging from 1880 to 2013 on my bookshelf), particularly if someone else compiles them. That’s because making a reference book is thankless work; the goal is to create accurate summaries of data, and comprehensiveness is valued over creativity.

Analysis to answer a question is far more interesting. Instead of writing an encyclopedia of facts you get to tell a story. You start with a hypothesis that you seek to prove or disprove, and because of that you’ll end up looking at data in a different way than if your goal is just to catalog it. Sharing a story with others is far easier and more impactful than sharing a compendium—people like hearing stories more than they like reading a thesaurus, and they’re far more likely to remember the story. Most importantly, if you choose the question you’re trying to answer properly, you can deliver real and immediate value to a company instead of delivering a work product that may someday be used to answer a question.

It’s hard to go from making an abstract to telling a story, but it’s even harder to tell a story without knowing what question you’re trying to answer. Sometimes it takes a little bit of abstract making to figure out how to articulate the problem you’re really trying to solve. Eventually though, you need to find that key question to tell your story around if you want to be relevant. Think about where people keep their reference books: do you want your analysis to go on a high shelf next to the dictionary, or sit on the top of someone’s desk?

I’m forever thankful for people who produce reference books, but I’ll choose to spend my time and energy telling a story over compiling a fact book any day of the week. Hopefully I have an interesting story to share with the rest of Basecamp in a couple of weeks, but I’m certainly more likely to have an impact than if I just set out to give them a catalog of facts.