Setting up promotions and discounts without eroding margin

The most common mistake is to decide the discount first and then look for a purpose. Do the opposite: first set what the promotion has to achieve, then choose the right tool. A store's typical goals are few and different from one another, and each calls for a different kind of offer. A promotion that works for clearing stock near expiry is almost always wrong for attracting new customers.

• Clearing dead or near expiry stock, turning motionless goods into cash.
• Attracting new customers into the store, accepting a thin profit on the first purchase to start a relationship.
• Raising the average basket, nudging people already in the store to buy more.
• Rewarding loyal customers, to protect the regulars from online competition.
• Covering a slow time of year, bringing people back when footfall drops.

A discount is not measured against price, it is measured against margin, and this is where almost everyone fools themselves. If a product carries a thirty per cent margin and you take twenty per cent off the price, you are not giving up a fifth of the profit: you are burning two thirds of it. The reason is that the discount eats into the margin, not the cost. Grasping this changes how you decide every offer.

The table shows why a discount equal to the margin percentage wipes out the profit: you are selling at cost. That is why a sensible discount has to be offset by a rise in units sold, and the rise has to be more than proportional. If you halve the unit margin, you have to more than double the units sold just to break even. A sum almost nobody does beforehand, and many discover after the fact.

There are many ways to run a promotion, and most protect the margin better than a straight discount because they ask for something in return: an extra purchase, a full price item bundled in, a customer detail. Here is how to design the offer starting from the goal you chose.

1. 1

   For the average basket, tie the offer to a threshold

   A discount or a free gift that kicks in above a certain spend pushes people to buy more and concentrates the benefit on those bringing a higher basket. The threshold goes above your current average basket, not below.

2. 2

   For clearing stock, bundle instead of discounting

   A dead product bundled with a fast mover, at a package price, moves the stock without devaluing the item in the customer's eyes and without training them to expect straight discounts.

3. 3

   For new customers, offer on the first purchase and capture the contact

   A benefit reserved for first time visitors only makes sense if in return you get a way to bring them back, such as joining the loyalty scheme. Otherwise you are paying for a customer who disappears.

4. 4

   For loyalty, reward in points or deferred value

   A reward that builds up with purchases or a voucher for next time costs less than an immediate discount and brings the customer back into the store a second time.

5. 5

   Always set a time and quantity limit

   A promotion with no end date becomes the new list price and stops working. A short, clear window creates urgency and lets you measure results over a defined period.

A promotion with no measurement is a bet you will never know you won. You do not need a complex system: a few numbers compared with the previous period and with the same period a year earlier are enough. The question to answer is not how much did I sell, but how much margin did I keep and did I hit the goal I wrote at the start.

1. Compare the total margin for the period with the previous period, not just turnover.
2. See whether new customers came in, if that was the goal, and how many returned afterwards.
3. Check the average basket during the promotion against the normal level.
4. Verify whether the near expiry stock really moved or you just discounted goods you were selling anyway.
5. Write one line of conclusion, repeat or do not repeat, so the next decision starts from data, not memory.