## February 14, 2012

### Calculating Jasmine

There are two things that set Jasmine apart from Jennifer.
1. Jennifer is e-commerce old-school.  She'll buy on Amazon using Amazon Prime.  Jasmine says that "if it is important, it will find me".  Her social commerce purchases are influenced by her friends.
2. Jennifer is the "multichannel attribution" shopper, using search + email + affiliates and other channels to purchase.  Jasmine loves MyHabit, and isn't going to leverage Google's empire to buy, she'll leverage Facebook.  If you see Jasmine on your site, you're likely to see an e-commerce order that doesn't appear to be attributed to any marketing activities.
Remember our weighting strategy?
• Mail Orders = Weight of 0.00.
• Telephone Orders = Weight of 0.15.
• Online Orders Matched Back to a Catalog = Weight of 0.30.
• Search Orders Matched Back to a Catalog = Weight of 0.40.
• Email Orders Matched Back to a Catalog = Weight of 0.50.
• Pure Search Orders = Weight of 0.60.
• Pure Email Orders = Weight of 0.70.
• Online Advertising Orders, No Offline Interaction = Weight of 0.75.
• Pure Online Orders = Weight of 0.80.
• Mobile, Social, Flash Sales Orders = Weight of 1.00.
Well, here's what Jasmine looks like.  She placed an order 13-24 months ago, a pure email order for \$100, and then she purchased yesterday via a link from Facebook, spending \$100.

Net Weighted Value = (\$100 * 0.50 * 0.70 + \$100 * 1.00 * 1.00) / (\$100 * 0.50 + \$100 * 1.00) = 0.900.

Here's our segmentation strategy:
• Judy = Value of 0.000 to 0.333.
• Jennifer = Value of 0.334 to 0.667.
• Jasmine = Value of 0.668 to 1.000.
Clearly, this is Jasmine.

If you want to make this stuff even more actionable, you link your web analytics data to this solution ... referring URL becomes very important ... skew net weighted value based on visits from Jasmine-like sites.

Tomorrow, we'll talk about what we, as marketers, do to ruin the "Jasmine Experience".

#### 1 comment:

1. With reference to "referring URL becomes very important" --

Users of Google Analytics can use the Custom Channel Groupings within the Multi-channel Funnels reports to create groups of 'Jasmine-like' sites based on the referring URL in the source.

This might well be useful in judging which sites are particularly relevant.

The data is anonymous, of course, so it won't be any use for the actual targeting of individuals. But it's still worth investigating.

### Hillstrom's Targeting: Value of Primary / Secondary / Tertiary

Ok, let's look at a practical example of Primary / Secondary / Tertiary categorization of merchandise categories. In this case, we ...