An Example Of The Direct Marketing Customer Continuum

This one comes up all the time. Take a look at catalog performance, over a four year period of time.

Evolution Of Segment Performance
		HHs	Phone	Online	Total	Profit	Change
2004	Catalog Only	10,000	$5.00	$0.20	$5.20	$0.81
	Catalog + Online	3,000	$2.00	$2.00	$4.00	$0.45
	Online Only	2,000	$0.05	$1.75	$1.80	($0.21)
	Total Segment	15,000	$3.74	$0.77	$4.51	$0.60
		HHs	Phone	Online	Total	Profit
2005	Catalog Only	9,000	$5.00	$0.20	$5.20	$0.81
	Catalog + Online	3,500	$2.00	$2.00	$4.00	$0.45
	Online Only	2,500	$0.05	$1.75	$1.80	($0.21)
	Total Segment	15,000	$3.48	$0.88	$4.35	$0.56	-3.4%
		HHs	Phone	Online	Total	Profit
2006	Catalog Only	8,000	$5.00	$0.20	$5.20	$0.81
	Catalog + Online	4,000	$2.00	$2.00	$4.00	$0.45
	Online Only	3,000	$0.05	$1.75	$1.80	($0.21)
	Total Segment	15,000	$3.21	$0.99	$4.20	$0.51	-3.5%
		HHs	Phone	Online	Total	Profit
2007	Catalog Only	6,000	$5.00	$0.20	$5.20	$0.81
	Catalog + Online	4,000	$2.00	$2.00	$4.00	$0.45
	Online Only	5,000	$0.05	$1.75	$1.80	($0.21)
	Total Segment	15,000	$2.55	$1.20	$3.75	$0.37	-10.8%

Again, I see this one all the time. Traditional RFM performance illustrates a segment that is "dying", performing progressively worse over time.

What is actually happening is quite different. Customers are shifting their status along the Direct Marketing Customer Continuum.

Catalog-Only customers require advertising. Notice that their performance hasn't changed over time.

Catalog + Online (those vaunted Multichannel Customers) are in the middle of our continuum, using advertising and search and word of mouth to buy merchandise. Notice that their performance hasn't changed over time.

Online-Only (customers that are self-serve customers, not needing advertising, have not changed their performance over time.

So the three key segments that multichannel brands track are all performing the same, over time. Yet in total, the performance of the total segment is dropping like a rock.

What is happening is that customers are moving along the continuum. Look at the number of households in each segment, from 2004 to 2007. Customers are shifting from needing advertising to not needing advertising.

Yet, while customers shift their behavior, the multichannel brand continues to do the same thing (mail catalogs), then wonders why productivity is dropping.

Simple solution: Take your RFM segment, and split them out by recent (not lifetime) purchase behavior --- catalog-only, catalog+online, online-only. If you see this phenomenon occurring, your problem is solved. Mail fewer catalogs to the online-only and multichannel audience, and improve the profitability of your brand.

Honestly --- for most multichannel brands, it isn't more challenging than that. And for those of you who are more advanced, go ahead and do the incremental mail/holdout tests we talk about all the time, you'll see different results than you see in your matchback reporting.

Modified RFM For E-Mail Targeting

RFM is great for targeting one catalog to one customer. However, RFM is tough to manage in a multichannel environment.

This becomes clear in e-mail targeting. Say you have a Mens version of an e-mail campaign, and a Womens version of an e-mail campaign --- a customer could receive either version on the same date. Use this customer as an example:

• Customer spent $100 on Mens merchandise in the past three months.
• This customer also spent $200 on Womens merchandise 7-12 months ago, and spent $100 on Womens merchandise 13-24 months ago.

Which version of the e-mail campaign do you send to a customer? You could use RFM --- your customer is a 0-3 month $100 mens buyer, and is simultaneously a 7-12 month $300 womens buyer. Which "segment" carries more "weight".

This is where we apply "Modified RFM".

Have your statistician build a regression model one time --- and use the "weights" or "coefficients" for your modified RFM scheme. I realize this is statistical blasphemy, however, we aren't managing clinical trials for cancer drugs, we're deciding which version of an e-mail campaign a customer receives.

Step 1: Pick a "dependent" variable for "Mens". I like to look at the past twelve months.

Step 2: Create a series of "independent" variables:

• Dollars spent on Mens in past three months (prior to the dependent time period).
• Dollars spent on Mens 4-6 months ago (prior to the dependent time period).
• Dollars spent on Mens 7-12 months ago.
• Dollars spent on Mens 13-24 months ago.
• Dollars spent on Mens 25+ months ago.

Step 3: Regress these five variables against your dependent variable. The "coefficients" become "weights" for e-mail targeting, as you'll see soon.

Step 4: Repeat Steps 1-3 for Womens merchandise.

Now, we can evaluate which version of an e-mail campaign a customer should receive. Let's look at our example:

E-Mail Targeting Strategy: Mens Weights
	Spend	Factor	Weight
00 to 03 Months	$100.00	1.600	160.0
04 to 06 Months	$0.00	0.600	0.0
07 to 12 Months	$0.00	0.300	0.0
13 to 24 Months	$0.00	0.150	0.0
25 to 99 Months	$0.00	0.050	0.0
Total Weight			160.0

E-Mail Targeting Strategy: Womens Weights
	Spend	Factor	Weight
00 to 03 Months	$0.00	1.600	0.0
04 to 06 Months	$0.00	0.600	0.0
07 to 12 Months	$200.00	0.300	60.0
13 to 24 Months	$100.00	0.150	15.0
25 to 99 Months	$0.00	0.050	0.0
Total Weight			75.0

For the Mens version of the e-mail campaign, the customer receives a "weight" of 160.

For the Womens version of the e-mail campaign, the customer receives a "weight" of 75.

So, you should send the customer the Mens version of the e-mail.

For your next campaign, you don't have to build models again --- remember, we're not trying to cure cancer, we're just figuring out which version of an e-mail campaign will improve response a bit. Just apply the same weights built in your prior modeling process, and decide who gets which version.

The key here is to not build separate RFM schemes. Instead, you build variables in your database that summarize purchases by 0-3 month, 4-6 month, 7-12 month, 13-24 month, and 25+ month time periods. Then you "weight" those purchases based on importance. This gives you a good targeting strategy.

Statistical purists will blast me for misuse of appropriate statistical techniques. That's fine. We're just trying improve e-mail marketing performance, while minimizing use of internal resources, or minimize expense incurred when hiring consulting statisticians. This gets you 80% of the benefit for about 5% of the work.

Multichannel Forensics And Information Value

Friend of MineThatData Alan Rimm-Kaufman shares that Google will retain about eighteen months of clickstream data, going forward.

Multichannel Retail faces similar challenges, when looking at the value of customer information within the context of multichannel forensics.

Customer information ages differently in multichannel retail. While the relationships are different for each business, the following example helps illustrate the point.

• Catalog: A purchase twelve months ago is worth about 1/2 of what a purchase that occurred today is worth.
  
• E-Commerce: A purchase twelve months ago is worth about 1/4th of what a purchase that occurred today is worth.
• Retail: A purchase twelve months ago is worth about 1/8th of what a purchase that occurred today is worth.
• Clickstream: A visit twelve months ago is worth about 1/32nd of what a visit that occurred today is worth.

This topic becomes important when evaluating actual customer behavior. Most multichannel retailers would consider a customer who purchased via catalog twenty-four months ago, and purchased online today, to be a "multichannel" customer.

The reality is that this customer is heavily skewed toward the online channel.

Multichannel marketers have an opportunity to run a regression-style analysis, to determine the appropriate weight to use with older purchase information. The weights determine how customers are segmented, and consequently, determine how the multichannel retailer markets to the customer.

Multichannel CEOs and CMOs: On Friday morning, talk to your analytics staff about segmenting customers on the basis of the value of older purchase information. Have your staff apply a new technique that ultimately mimics the time-honored system of "RFM --- Recency, Frequency and Monetary".