June 01, 2016

The Brand Ecosystem - An Image

We're going to spend a little bit of time talking about images like this image.

This image is what I call the "Brand Ecosystem". The image shows how different aspects of an order all relate to each other.

Points that are close to each other in the ecosystem are points that commonly appear together in an order.

Tomorrow - I will talk about the data points.

Today - I will share the geeky methodology I use - if you don't like math, stop here, and join me tomorrow.

Dataset:  1 Row Per Order, Past Twelve Months.

Variables In The Dataset:
  • Household_ID.
  • Life-To-Date Order. If the customer placed the 3rd and 4th orders in her lifetime, then there are two rows for this customer ... one with a life-to-date-order value of three, one with a life-to-date-order value of four.
  • 1/0 Indicator:  Is this the first order ever for this customer (1 = yes, 0 = no)?
  • 1/0 Indicator:  Is this the second order for this customer?
  • 1/0 Indicator:  Is this the 3rd-5th order for this customer?
  • 1/0 Indicator:  Is this the 6th-10th order for this customer?
  • 1/0 Indicator:  Is this the 11th-25th order for this customer?
  • 1/0 Indicator:  Is this the 26th+ order for this customer?
  • 1/0 Indicator:  Is this order tied to a discount/promotion?
  • 1/0 Indicator:  Did this order have a single item (0) or multiple items (1)?
  • 1/0 Indicator:  Was the average price of items purchased in the lower third?
  • 1/0 Indicator:  Was the average price of items purchased in the middle third?
  • 1/0 Indicator:  Was the average price of items purchased in the most expensive third?
  • 1/0 Indicator:  Affiliate Channel.
  • 1/0 Indicator:  Flash Sales Channel.
  • 1/0 Indicator:  Subscription Channel.
  • 1/0 Indicator:  Cost per Click Channel.
  • 1/0 Indicator:  Email Channel.
  • 1/0 Indicator:  Housefile Catalog Channel.
  • 1/0 Indicator:  Sales Rep Channel.
  • 1/0 Indicator:  Prospect Catalog Channel.
  • 1/0 Indicator:  Search Channel.
  • 1/0 Indicator:  Print Ads Channel.
  • % of Demand from Product Category 01.
  • % of Demand from Product Category 02.
  • % of Demand from Product Category 03.
  • % of Demand from Product Category 04.
  • % of Demand from Product Category 05.
  • % of Demand from Product Category 06.
  • % of Demand from Product Category 07.
  • % of Demand from Product Category 08.
  • % of Demand from Product Category 09.
  • % of Demand from Product Category 10.
Place each variable into a Factor Analysis / Principal Components Analysis in your preferred software package.

Use a Varimax Rotation.

Plot the coefficients from the Rotated Component Matrix on an x/y grid.

It's that simple!

Now go try the methodology on your business, and plot your brand ecosystem.

1 comment:

Note: Only a member of this blog may post a comment.

End of the Line for the Orvis Catalog

When a traditional catalog brand makes changes, I get a head's up via blog / email subscriber metrics. Blog unsubscribes happen from the...