Classification
Let's talk about something I love - Classification. I'm not a super organized person when it comes to clutter, but I can spot a good taxonomy when I see one.
We use classification systems all the time - from books to movies to video games. Most of us probably have our clothes in different bins based on some sort of categorization system. Do you organize your bookshelf by topic? Author? Genre? Color?
Classification is important because it helps us make sense of the world around us. It helps us understand relationships and communicate information more readily. Think about it. Deciding on a movie in group usually starts with genre, finding a restaurant might start with cuisine, sorting Legos might start with shape or color.
Classification can be done in many different ways - by physical characteristics, by topic, by time, etc. It's a flexible tool that we use every day without even realizing it.
Something that I encounter a lot with what I do as a product manager is building systems of classification. And - typically, I'm doing that in an area that I don't know a ton about, or more frequently, don't know what all the things are in the space that even need to be classified.
Take my button collection. No, I don't collect buttons, but I do have a collection. And I do have it organized - so I can find the right button faster. Lauren randomly bought me a big bag of buttons - all kinds of buttons. We were adding them to some pillow covers that I had made. Naturally, my brain wanted (needed) to organize them.
There are a few considerations that I want to call out about a finished system.
What are the physical constraints?
How many bins do I have?
How quickly do I want to access them?
Can I see all of the levels of organization, or is a level 2 locked until I choose level one?
When I'm searching for a button, what am I naturally looking for first: color, shape, size, attachment method, etc.
And a few about the process.
Do I want to build the classification system first? or
Do I want to discover everything about the buttons along the way and build the system as I go?
When I don't know all of the variations or structure to the things I'm classifying, I typically start with the later. Sometimes that means that I need to restart or restructure a few times along the way.
And that's okay.
Building the Classification System First
Especially in a known ecosystem, you can readily build out the classification system at the start. For example:
Dump out the buttons.
Take time considering what makes some buttons distinctive.
Decide on hierarchy (Color -> Shape -> Number of Holes)
Start sorting
Discovery Classification Along the Way
When you don't know the space as well, it can be difficult to decide on the classification system at the start. I've found it better to start with grouping and then decide on the actual classification system later.
Dump out the buttons.
Group like items together - Solid, translucent, metallic for example
Look at these groups individually - do they have obvious sub-classes?
If so, split them out again, if not go back a few steps.
Look at these sub-classes. Should a few be combined to a new top-level class? maybe buttons that have an attachment other than a through-hole?
Keep going until you have a system that makes sense to you. You can always change it later - that's part of the fun!
The best part about classification is that it can always be improved. Just like everything else in life, it's a process. And as you learn more about your buttons (or whatever it is you're classifying), you can update your system to better reflect what you know.
[This part is secret] In addition, knowing the subject and how its classified can help convey meaning and understanding to someone from the outside in a more streamlined fashion.
Happy Hobbying!
