Some things you’ll need to guess in the beginning. But *your* guess in *your* industry should be better than anyone else’s guess, so you’re in good hands. After sending your first test campaigns, data will start to roll in and you’ll need to nail it every time. Half a per cent here and there can make the difference between historic failure and superstar winner. The maths of direct mail are your friend.

Here are a few metrics that you must understand to play this game properly:

## Cost Per Response (CPR)

This is an easy one – it’s simply the cost of the promotion divided by the number of responses. How much you include in the promotion cost is up to you – it may be as simple as the cost of print and fulfilment; but could also include design, copy, image licenses, time etc. Make sure you’re consistent with your maths.

## Allowable Margin (AM)

Dead simple – the amount of money left for advertising and profit after everything else is spent (i.e. revenue minus the *incremental* cost of selling one more Thing)

## The Maths of Churn .. bleargh

Churn rate is the percentage of customers who give up on being customers for whatever reason after a month expressed as a percentage. It’s not directly useful for DM other than telling you how many new customers you need to refill the bucket, but you will need it to have a stab at CLV.

So.. if you had 1000 paying folks on Jan 1st and by Feb 1st 75 of them leave, your churn rate is:

(75/1000)*100 = 7.5% per period (in this case a month)

From the churn rate, you can calculate how long an average customer stays with you:

(1/Period Churn Rate)*100 = (1/7.5%)*100 = 13.3 months

Doh! Don’t forget … only count *existing* accounts, not new ones joining in the period or your maths is mangled.

## Customer Lifetime Value (CLV)

CLV is simply how much margin a new customer is worth to you over their entire life with you.

New customer acquisition is expensive. First, you need to find a prospect, then you need to convince them you’re more worthy of their money than any of the other thousands of ways they could invest it.

Thus, we need to keep a close handle on not only acquisition cost (number of actions/spend to get those actions), but what those things we’re acquiring are worth…

The simplest way to estimate CLV is to base a future projection on historic performance. So, figure out how much margin a customer segment generates in a specific period, and multiply that by the expected number of periods they’ll be spending with you.

That will give you the amount of AM you can expect to earn from one new customer.

If you have a 3 year customer lifespan, and spending is £500 per year with a gross margin of 30% , then the maths for one incremental customer is £1500×30%=£450 profit.

Armed with this knowledge, you can work backwards and know that for a pure new business acquisition direct mail campaign costing £1 per piece, you are breaking even at a 0.22% conversion rate (NB assuming you haven’t already included your marketing costs inside of your cost of goods sold).

Target Conversion Rate to Break Even = (Cost Per Piece/CLV)*100

In other words, the maths says spend £20 per mail piece and you’ll need direct mail to convert 4.44% to new customers just to break even.

Make your model as complicated as you like for CLV, but the bottom line is without a number, you won’t know how to measure the success of a direct mail test campaign for new acquisitions. Simple as. You don’t need much maths, but you must track some.

**Planning from the maths**

Now you know that there’s a target conversion rate across the board of 0.22% for example, you can reality check the segments you are sending to.

If you find out that management consultants spend twice as much with you than IT consultants, you can now many an informed decision about what kind of campaigns to run against either, honing the messaging and behavioural triggers to maximise returns or even dropping certain segments completely. A grasp of the maths of direct mail will always reality check and guide decisions.

The money isn’t only in the list, it’s in how clever you are with data-driven decision making.