Data Analysis - Fundamentals (Patreon)

So, we talked about KPIs earlier (key performance index); but I just had a conversation about what kind of data to track. I thought I'd mention a few things that happens when one ends up in the digital marketing arena and things to watch out for.

Information Overload

In comparison to traditional advertising, one of the biggest advantages of digital advertising is how detailed your results can get. You can often see the results of advertising immediately - everything from clickthrough rates, impressions to readthroughs and sales.

In fact, there's so much data that you can get that it can be overwhelming. Or worse, you end up recording everything.

I call it information overload - having so much data that you record everything, thinking you can analyze it all for some use.

And you can... but...

What's Important?

One of the most important things to do is make sure you know what the information you are recording is going to be used for. It's all too easy to record information that will not be useful. Or, will only be marginally useful.

As a small business owner, you aren't a data analyst for a Fortune 500 company. You don't need or care about the miniscule changes, you're more worried about the big picture.

Figure out the most important things that you need to know when you record information and focus on that.

A Concrete Example

I recently ran a Bookbub (UK Only). Now, I want to know how it did. So, what do I need to know? Well, I have access to a slew of information, from sales, page reads, individual book sales and series sales, as well as sales by marketplace for each of those.

All that is useful, but what do I really need to know?

Here's the basics:

- Cost of the promotion ($160)

- Unit Sales & Revenue for previous period (5 days / 30 days) in the UK (to create trendline / base data)

- Unit Sales & Revenue during promotion period & immediate aftermath

- Unit Sales & Revenue for period after promotion (5 / 30 days)

- Repeat above information for series data

While running read-through numbers from book 1 to book 2 and book 3 (and the change in that) might be useful, I really want to answer a few questions:

- Did I earn more money than I spent?

- How long would it take for that profit to happen?

- And is that earning out from the book itself or via the series?

Everything else is just finagling minor data. I could record more information, and see the % increases, the variations of readthrough and what not, but the question is - what am I going to do about it? 

Example. I learn that Readthrough rates are lower on Boookbub promos. So what? Am I going to rewrite the book? No. Then that data isn't useful.

I learn that it takes me 5 days to see an increase in page reads. Great. That's interesting. But is it useful? Not really. I don't get paid for 60 days, so it isn't affecting my cashflow. So it's not really useful / actionable information.

And so on, so forth.

End of the day, the ONLY thing I really care about is - would I do this promotion again? And if I had a limited budget, is Bookbub better or worse compared to other promos of that type?

That's really it. Everything else - and I do mean everything else - is just noise.

Final Words

When you do data analysis, whether it's for sales or whether your wide strategies are working or if you should move your audiobooks over or whatever, the question you need to ask yourself is simple:

- what is the data useful for and;

- what ACTION will I take based off this data

If you are analysing things for the fun of it, great. But, really, you should be taking action based off the analysis. If you aren't taking action based off the data your collect and analyze, why bother?

You should be writing instead.