If you've been following this blog, you can tell I believe the "do-not-mail" movement will soon become a very significant issue for multichannel and catalog marketers.
In one conversation with Chuck, he shared his belief that CatalogChoice and other do-not-mail efforts would help catalogers avoid mailing individuals who didn't want their books, thus helping catalogers be more profitable.
I wasn't so sure about that, so I built a small mathematical model to help myself understand the relationship between mailing costs, response rates, never-mails, and the bottom line.
Consider your holiday mailings in aggregate as one campaign.
Let n_b and n_p represent the number of buyers and prospects mailed, respectively.
Let r_b and r_p represent your buyer and prospect response rate, and let theta_b and theta_p represent the fraction of buyers and prospects who put themselves on the never-mail-me list.
Let c denote the cost of a catalog, and alpha represent contribution per order, defined as average order less average COGS less average variable order costs (credit card fees, pick-pack costs, etc).
We'll assume theta_b and theta_p are independent of r_b and r_p. That says the typical buyer never-mail-me optout responds no better or worse than the average buyer, and the typical prospect never-mail-me optout responds no better or worse than the average prospect.
If anything, that assumption is conservative: I've had conversations with catalogers indicating that the buyers choosing to opt-out through Catalog Choice are typically far better than average buyers. More on that later.
OK. Without the never-mail effect, you'll earn
dollars on the mailing.
With the never-mail opt outs, you'll earn
dollars on the mailing.
So when would never-mail opt-outs increase profits?
Using high-school algebra, we can rearrange terms like this.
For the cataloger concerned with profit, the never-mail-services increase your bottom line (in the short term, more on that later) if they cull more low-responding low-profit prospects relative to their removal of high-responding high-profit buyers.
How much bigger does the prospect optout rate, theta_p, have to be relative to the buyer opt-out rate, theta_b, to make it work out in the mailer's favor?
The critical ratio depends on several factors
- c, the cost of the book. The more expensive your book, the larger your buyer opt-out rate can be and you'll still be OK.
- n_p / n_b , the ratio of prospect to buyer circ. The more you mail prospects, the larger your buyer opt-out rate can be and you'll still be OK.
- r_b alpha - c , the contribution per buyer book. The more money you make from a buyer, the small your buyer-opt out rate has to be.
Here's a spreadsheet you can use to play with some of these numbers.
In this case, I set the buyer opt-out rate, theta_b, at 2%. Ouch.
I dialed in a cost-per-book of 50c, an average order of $150, COGS at 50%, variable costs at 5%, buyer circ at 1000k with a 2% response, and prospect circ at 1000K at a 0.5% response rate.
With these numbers and a 2% buyer do-not-mail me rate, the prospect opt-out rate has to be at least 3.4%, so that the money saved from losing low-responding prospects covers the income loss from the high-responding buyers who opt-out.
This model only considers a single mailing.
By ignoring life-time value and the need for fresh buyers to keep a house-file healthy, it underestimates the impact of the opt-outs.
Time's, They Are A'Changing
The catalog world is changing, and changing quickly. The do-not-mail movement is one of the small-but-soon-to-be-seismic shifts hitting the industry.
The question isn't whether or not you participate in the do-not-mail collectives. You're going to have to, by consumer sentiment and perhaps even by law in some states.
The question isn't whether or not you are going to see some of your best buyers raising their hands to never receive your book. You will see this, and their online sales won't make up what you made when you pushed (far too many, likely) books at them.
The question is: how do catalogers revise their business models in 2008 and 2009 to reach 2010?
Spreadsheet: Catalog opt-out model