reporting kpi's with chernoff faces by super analytics
DESCRIPTION
This document is a series of studies, whitepapers, presentations, instagrams, and other forms of publications in which the Super Analytics team seeks to find new / old smart and even crazy as yet innovative ways to report performance or to visualize data.TRANSCRIPT
Chernoff’s Faces Generator
How to create a simplistic yet powerful data visualization engine that generates complex data in
Chernoff’s Faces style.
Purpose of this Document
This document is a series of studies, whitepapers, presentations, instagrams, and other forms of publications in which the Super Analytics team seeks to find new / old smart and even crazy as yet innovative ways to report performance or to visualize data. Prior to making this document we wrote a blog post at www.superanalytics.fi/blog about the Chernoff’s faces We hope you’ll enjoy! J Best, Kalle Heinonen Super CEO @super_analytics
What is Chernoff’s faces (wikipedia)
Chernoff faces, invented by Herman Chernoff, display multivariate data in the shape of a human face. The individual parts, such as eyes, ears, mouth and nose represent values of the variables by their shape, size, placement and orientation. The idea behind using faces is that humans easily recognize faces and notice small changes without difficulty. Chernoff faces handle each variable differently. Because the features of the faces vary in perceived importance, the way in which variables are mapped to the features should be carefully chosen (e.g. eye size and eyebrow-slant have been found to carry significant weight)
Example of Chernoff’s Faces
Getting started
The ideology behind the simplistic Chernoff’s Faces Engine (CFE) is quite straight forward. Each of the ‘facial indicators’ are generated by their predefined min-max locations on x-y graph. At first we will need to define the variables or the facial indicators.
Facial indicators (Variables)
Eyes = Sales increase /decrease in % Gaze = EBITDA increase /decrease in % Eye-brows = Customer Retention % Ears = Customer Acquisition Cost Nose = Market Share Mouth = Brand Awareness increase /decrease Head = Employees
Variable Values = Eyes
Eyes = Sales increase /decrease in % Sales
Excellent
Good
Sa/sfactory
Poor
Bad
2
1
0
-‐1
-‐2
Variable Values = Gaze
Gaze = EBITDA increase /decrease in % Excellent
Good
Sa/sfactory
Poor
Bad
2
1
0
-‐1
-‐2
Variable Values = Eye-brows
Eye-brows = Customer Retention % Excellent
Good
Sa/sfactory
Poor
Bad
2
1
0
-‐1
-‐2
Variable Values = Nose
Nose = Market Share % Excellent
Good
Sa/sfactory
Poor
Bad
2
1
0
-‐1
-‐2
Variable Values = Ears
Ears = CAC (Customer Acquisition Cost) Excellent
Good
Sa/sfactory
Poor
Bad
2
1
0
-‐1
-‐2
Variable Values = Mouth
Mouth = Brand Awareness increase Excellent
Good
Sa/sfactory
Poor
Bad
2
1
0
-‐1
-‐2
Variable Values = Head
Head = Employees Excellent
Good
Sa/sfactory
Poor
Bad
2
1
0
-‐1
-‐2
Sample Chernoff-KPI
The sample Chernoff-KPI on the right is based on the 5-level threshold-driven character engine. Each set of 5 characters is defined for each KPI individually. As a combination these form holistic performance report, with a grin.
Pretty much everything is down. The market share and amount of employees remains high, yet all other fronts are performing poor. Sales & EBITDA is going down.
Super Analytics CFE applied into realistic perspective.
We will create a simplistic comparison of Apple Inc and Nokia Corporation to form an Analysis of their Competitive Positioning via Market Share, Growth and Turnover in FY2006 and FY2012.
NOKIA CORPORATION APPLE INC.
Competitor Analysis (Apple vs. Nokia 2006)
Gro
wth
Market Share
HI
LO HI
HI
Nokia = $41bn
Apple = $19bn
Competitor Analysis (Apple vs. Nokia 2012-2013)
Gro
wth
Market Share
HI
LO HI
HI
Apple = $156bn
Nokia = $39bn
What else could be done with the Super Analytics CFE?
Customer Experience
Measurement
Marketing Campaign Happiness
Customer Satisfaction
Analysis
Employee Satisfaction
Analysis
Marketing Channel Analysis
Production Quality
Analysis
Marketing Channel Analysis (Bought vs. Earned Media)
Ret
urn
on In
vest
men
t
Cost Per Acquisition
HI
LO HI
HI
Earned media
Bought media