How do you know when something works in direct marketing? The suits probably never explain this to you, so I’ll give it a shot, using a simple case study.

When we were in New York, we worked with a big brokerage house. They wanted to attract investors with liquid (i.e. moveable) assets of at least $500,000. They’d tried cold calling, advertising, referrals – everything but direct mail so they asked us to give it a shot.

They were nice people but didn’t know anything about DM, so we had a lot of initial chitchat. They were surprised when I asked them how much of their investors’ money they got to keep. They asked “What difference does that make?” and I explained that’s how we figure out if the program works.

They get an annual fee of 1.5%.

Then I asked how long the average high end investor stays with them. Again with the “What difference does that make?” We explained that’s how we figure out the Lifetime Value of a customer.

Three years seemed about right to them.

1.5% of $500,000 is $7,500 a year. Soooo, over three years, a high end client will generate $22,500 dollars in fees. You subtract the cost of getting that client and servicing the account from that number.

We put together a program with a lot of tests because nothing anyone else had done for them ever worked, mostly because the very rich are a tough audience.

In the meantime, while our tests were in the mail, the company hired a gung ho advertising manager.

When the results came in we did the numbers. There were about a dozen test cells and their average response was 0.7% (seven tenths of one percent) with an average conversion rate of 55%. That means that 0.385% of the people we mailed invested $500,000 or more with the company.

Then we extrapolated from all the test results to estimate what the result would be if we used only the winning elements of each test. We’re very conservative when we do this. No matter how we ran the numbers, we figured that the simplest package with the best performing letter, the best offer, first class stamp (rather than preprinted postage paid indicia) on the reply envelope, mailed to the top 5 lists in the top 6 markets would generate a basic response of 3.2% with a conversion rate of 60% which would result in 1.92% of the people we mailed to investing $500,000 or more with our client.

We had found a gold mine! The mailings and the offer were expensive, about $2.50 each, including all costs right up to the fulfillment of the expensive premium.

So, for every 10,000 pieces we mailed, we’d spend $25,000 and get 192 people to invest at least $500,000 for a total of $96,000,000. Revenue (1.5% x 3 years) for the company would be $4,320,000. If we mailed 100,000 names, we’d generate over $40,000,000 for a cost of $250,000. That’s an ROI of about 160 to 1. We’d created a money-making machine that for every dollar we put in the top spits out $160 over three years. Factor in the cost of servicing those accounts and the ROI is around 120 to 1. We were very excited.

We presented these numbers to the new advertising manager at our first meeting. She barely listened then looked at us with a steely eye and said “You didn’t get 1% from your tests. Everyone knows that’s the standard in direct response.” And the company never did another mailing.

Lessons? 1) A little knowledge is a (very) dangerous thing. 2) Advertising people should never be involved in direct marketing.