WEBVTT

00:00:01.200 --> 00:00:03.433
This case study grew
out of a real technical

00:00:03.467 --> 00:00:05.967
assessment for a
data analyst role.

00:00:06.433 --> 00:00:08.767
I rebuilt the challenge
with fictional books and

00:00:08.800 --> 00:00:13.067
prices, so I could show
the process publicly.

00:00:13.933 --> 00:00:16.967
The task was to analyse
a publishing price list.

00:00:17.433 --> 00:00:18.533
The catch?

00:00:19.000 --> 00:00:21.233
You cannot compare two
books fairly until you

00:00:21.267 --> 00:00:22.800
account for the level,
the edition,

00:00:22.833 --> 00:00:25.833
the format and
everything included.

00:00:26.700 --> 00:00:29.833
Six product groups initially
looked inconsistent.

00:00:29.933 --> 00:00:32.133
Four had a cheaper
Starter level.

00:00:32.167 --> 00:00:33.667
But two were different:

00:00:33.767 --> 00:00:36.467
the books had been filed
under the wrong series.

00:00:36.933 --> 00:00:39.000
What looked like pricing
problems were actually

00:00:39.033 --> 00:00:41.033
classification problems.

00:00:41.933 --> 00:00:44.033
I left the raw data untouched,

00:00:44.067 --> 00:00:47.467
corrected the reporting view
and reran the comparison.

00:00:47.933 --> 00:00:52.700
Consistently priced groups
went from 54 to 56 out of 60.

00:00:53.033 --> 00:00:54.300
The interactive case,

00:00:54.333 --> 00:00:56.533
the fictional dataset
and the calculations are

00:00:56.567 --> 00:00:57.867
on the website.

00:00:58.233 --> 00:00:59.833
I'm Alfonso Cifuentes.

00:00:59.867 --> 00:01:01.667
This is Pricing Demo.