2. Obvious to Some Analysis
|Radiologist : Minimal things to consider like in a simple fracture of the finger. Only 1 or 2 abnormal findings. Obvious to everyone its a fracture and a radiologist can be definite about diagnosis. Reported in either of the ways below depending on the situation
|Data Analysts: Multiple reasons that are not obvious as to why sales increased but analyst should be able to bring everything together and say why.Example: Let me explain this using an example of entrepreneur Ramakrishnan that I read in Effective Data Storytelling: How to drive change with data, narrative, and visuals by Brent Dykes. Ramakrishnan mentions that a few years ago, he looked at transaction data from a large B2C retailer.One of the fields in the dataset was ‘transaction amount’.
What did we expect to see? Well, we expected that most amounts would be around the average, but there will likely be some smaller amounts and some larger amounts. So a histogram of the field would probably look like this:
But when we checked the data, this is what we saw:
Ramakrishnan termed this unexpected bump ‘hmm’ and further shares that these transactions weren’t made by their typical shopper — young moms shopping for their kids. They were made by people who would travel to the US from abroad once a year, walk into a store, buy lots of items, take them back to their country and sell them in their own stores. They were resellers who had no special relationship with our retailer. This retailer didn’t have a physical presence outside North America at that time nor did they ship to those locations from their e-commerce site. But there was enough demand abroad that local entrepreneurs had sprung up to fill the gap. This modest “discovery” set off a chain reaction of interesting questions on what sorts of products these resellers were buying, what promotional campaigns may be best suited for them, and even how this data can be used to inform global expansion plans.
Now, only an experienced data analyst will be able to put the information in the graph, analyse the graph and investigate the hmm. Others, may just resort to rationalising the increase in sales to multiple reasons like promotions etc, with no clear insight that can lead to strategic direction for better business outcomes.
The data analyst adds value by investigating the minor subtle causes here and doesn’t stop at just looking at the biggest and brightest cause.
"I attended your story telling course some time back. And I've enjoyed keeping up my knowledge with your blog. You may not have realised however, that the Whole of Government is implementing Internet Seperation. Hence I'm not able to access the links to read your articles. Could I suggest including a QR code in your emails so that I can use my mobile to scan it and gain immediate access to the article? It would be most helpful"