We've worked on a lot of data projects, and the ones that performed always came down to what happened before any analysis started.
So now, before we touch a single spreadsheet, we ask three questions. They may sound simple. They're not always easy to answer. But they save a lot of time.
1. What decision does this need to support?
Not "what do we want to know?" What decision. Because data without a decision attached to it is just information. It might be interesting, but interesting doesn't really move things forward.
When we push people on this question, we often find that what they thought was a data problem is actually a clarity problem. They don't need more data. They need to get clearer on what they're trying to decide, and who's doing the deciding.
2. Who needs to be convinced, and what would convince them?
This one changes the shape of everything. A CEO and a junior analyst need completely different things from the same data. One needs a story with a clear direction and implication. The other might actually want the methodology.
Getting this wrong is expensive. We've seen great analysis fail to land because it was presented with the wrong intent, to the wrong person, in a poor format. The data provided was right. The communication wasn't designed for its audience.
3. What would we do if the data said the opposite?
This is the uncomfortable one, and it's the most important.
If the answer is "we'd probably still do it anyway" that's useful to know. It means you're not really looking for insight, you're looking for validation. Which is fine, as long as everyone's honest about it. But it changes what we'd recommend spending the budget on.
The best projects we've worked on are the ones where clients were genuinely open to the data, challenging their assumptions. That's where the most interesting stuff tends to happen.
Three questions. Sometimes they take five minutes. Sometimes they take a week. Either way, they're worth asking before anything else.
This is part of an ongoing series where we share thinking on data, storytelling, and the space between the two.