Data is not a quiet thing.
If you listen carefully, it does speak to you. The question is: how do you listen?
1. There are no patterns to be found in isolation (What’s your data for?)
I have said this before and it is worth repeating again and again. A single object, a single thing, could effectively be meaningless because the lack of context means that any definition you apply is as changeable as your mood, as capricious as the weather in the middle of a US Northeast March.
Context gives concreteness.
Therefore, data, when it is bounteous in size, is something that begs to tell what patterns exists in it. Reams of data for a particular subject, survey, et al., (the context) are not random. Depending on the complexity, it may be impossible for your to derive, define, or divine the pattern (and what causes it or if it is the cause itself), but that does not negate its existence.
2. Set the context within the context (What question do you need answered?)
You cannot provide an answer without a question. You can provide a statement, but it does not become an “answer” without an associated “question”. (See what I did there? The statement derives its next definition by the context by which it has been provided by the question).
Therefore, when you’re given a dataset, before you dive in and dig for a pattern, find out if you’re looking for statements (generically, what you see as what you see), and then tighten up the context by posing a question.
- Do I want to know how many times a day this particular number appears in this particular field for this particular set of products?
- Do I want to know what the average cost is over a week, a month, a decade?
- Do I want to know how often I get an error of this specific type after this specific action?
3. Stop looking for the Philosopher’s Stone (Answer your question).
We have all been there. We have posed a question, we have answered it after thorough consideration and search…and we are not happy with it. We do not believe what we see. Is it perhaps too simple, too easy? Or is it perhaps too absurd to be true?
Sometimes, it is just that easy and is just that absurd. And sometimes, it is not – we asked the wrong question and so we got the wrong answer. But, a singular answer, or set of answers, cannot solve the “mystery” of your data, because data, in and of itself, is an isolated set of ones and zeros, bits and bytes that only take on meaning when you enrich, contextualize, and transform it into information.
Information must be tethered to the audience it is trying to inform.
I cannot be everything to everyone and that is okay.
That is my approach to making my data talk – what’s yours?
Please share this with a friend or colleague who you believe would benefit from it by using the buttons below.
Also, let’s start a conversation! You can always send me an email at firstname.lastname@example.org.