The principles of good data visualization ft. Amanda Shendruk

That makes complete sense.

Yeah.

Let's talk about more of these like principles.

You're sort of getting into this, but I want to know some of the principles of good data is right.

So I don't know.

They're like, this might be begging the question too much, but like, do you have like three like basic rules or principles that would apply regardless of the audience? Right.

Maybe you're getting to some of this, like this idea of useful redundancies, like always include useful redundancies.

But I'm just wondering if there are some like, universal general 101 type of principles you can give us of good data viz

well so when i think about good data viz um well there's so many different things you know there's entire books there's entire courses it's like

i'll take i'll look we got time

But let me talk about sort of some of the questions that I still ask myself every time I do a chart to like ensure that it's, it's the best, most accurate, most useful piece that I can put out there.

So the first thing I always, and we've already talked about this, but the first thing I always ask myself is who is my audience?

Like,

because it makes such a difference.

If, if, if I'm doing a piece on heat waves, for example, and my audience is academics and climate scientists, um, I'm going to include a lot more detail.

I'm probably going to include error bars or uncertainty lines in there because these people, they are very well versed in this kind of data or data that's similar.

You know, they're in a world where that makes sense to them and doesn't scare them off.

But that's not my normal audience.

The audience for NotShip is a little bit more general.

It's people who are definitely interested in data, who want to know more about data, who can read a chart, but not people who are, well, I guess I do have some data scientists, but it's not usually data scientists.

So for my audience now, I question everything and I think, well, okay, for my current audience, who's like an educated person who's interested in the world, is this the right level of detail? And this is also where the useful redundancy comes in, because I don't think that if I were doing a bunch of charts for academics...

I would necessarily, if I were doing a series of charts with the same axes, that I would necessarily label the axis exactly in detail every time.

But because I have this audience, I do label it in detail every time because I don't want people to have to go back and think, oh, wait, what is this showing? I want it to be as easy as possible.

And as easy as possible is very different for different audiences.

And something I should say, these kind of questions that I ask myself, None of these are like specific design rules, but all of the design can come out of this, if that makes sense.

That makes complete sense.

Yeah.

Let's talk about more of these like principles.

You're sort of getting into this, but I want to know some of the principles of good data is right.

So I don't know.

They're like, this might be begging the question too much, but like, do you have like three like basic rules or principles that would apply regardless of the audience? Right.

Maybe you're getting to some of this, like this idea of useful redundancies, like always include useful redundancies.

But I'm just wondering if there are some like, universal general 101 type of principles you can give us of good data viz

well so when i think about good data viz um well there's so many different things you know there's entire books there's entire courses it's like

i'll take i'll look we got time

But let me talk about sort of some of the questions that I still ask myself every time I do a chart to like ensure that it's, it's the best, most accurate, most useful piece that I can put out there.

So the first thing I always, and we've already talked about this, but the first thing I always ask myself is who is my audience?

Like,

because it makes such a difference.

If, if, if I'm doing a piece on heat waves, for example, and my audience is academics and climate scientists, um, I'm going to include a lot more detail.

I'm probably going to include error bars or uncertainty lines in there because these people, they are very well versed in this kind of data or data that's similar.

You know, they're in a world where that makes sense to them and doesn't scare them off.

But that's not my normal audience.

The audience for NotShip is a little bit more general.

It's people who are definitely interested in data, who want to know more about data, who can read a chart, but not people who are, well, I guess I do have some data scientists, but it's not usually data scientists.

So for my audience now, I question everything and I think, well, okay, for my current audience, who's like an educated person who's interested in the world, is this the right level of detail? And this is also where the useful redundancy comes in, because I don't think that if I were doing a bunch of charts for academics...

I would necessarily, if I were doing a series of charts with the same axes, that I would necessarily label the axis exactly in detail every time.

But because I have this audience, I do label it in detail every time because I don't want people to have to go back and think, oh, wait, what is this showing? I want it to be as easy as possible.

And as easy as possible is very different for different audiences.

And something I should say, these kind of questions that I ask myself, None of these are like specific design rules, but all of the design can come out of this, if that makes sense.