Don't give coffee drinkers free coffee

With footfall declining and wallets shrinking, retailers desperate to make a sale – any sale – could be forgiven for thinking a price cut promo is the best way to go. After all, everyone loves a bargain, and more people in store buying cheap stuff = a better return than nobody in store buying expensive stuff. Right? The truth is heavy discounting can easily backfire. We all know how margin works; if you sell lots of high margin items you make lots of money. Price discounts gouge a chunk out of that margin, and you're going to have to sell more of your discounted tchotchkes to make the same profit. The bigger the discount, the more you need to sell – and hey, it doesn't always work that way. You may actually be cannibalizing your own sales by encouraging people who would have bought a high margin product at full price to come in earlier (or wait til later) and buy it cheaper. Oops.

In NZ, camping & outdoors retailer Kathmandu found this out the hard way. After years of declining profits despite seemingly constant discounting and clearance sales, the company's administration is on a mission to turn the brand around. How? With more responsive promotions and better personalization – getting people in store by giving them what they want when they want it, and not inadvertently encouraging them to hold off until the next big sale.

If people want to pay full price, let them

If Sarah comes in every morning to get a coffee, you don't need to offer Sarah a discount on coffee. She's already coming in and buying it at full price; any discount you offer is going to needlessly erode your margin. What you should be doing is aiming for a different kind of behavior change – encourage Sarah to come in more often, or encourage her to spend more money by introducing her to more stuff she likes.

Behavior change 1: get Sarah in store more often

You know Sarah comes in between 8-9 on weekday mornings for her coffee fix, and you know she always orders an Americano. You don't see her at all for the rest of the day, but you know from her phone's location that she's usually nearby between 3-4pm. So since she's near, and since she likes coffee, and since it seems (in the absence of conflicting data at this point) she hasn't had a decent coffee since this morning, you push her a promo for a free drink when she's close to the store one afternoon. If she takes up the offer, you can push her a 50% discount a couple of afternoons later. Then a smaller discount after that. Once you've successfully introduced the new coffee-in-the-afternoon behavior, you can stop promoting it because you know she'll buy at full price.

Behavior change 2: encourage Sarah to spend more

Since Sarah's in-store regularly, you can take the opportunity to promote other products – probably something that goes well with coffee. By encouraging her to pick up a free (then discounted, then less discounted etc) croissant with her coffee, you can introduce a new, regular croissant-with-coffee behavior that increases Sarah's basket value and therefore her value as a customer. Of course if Sarah never takes up the croissant offer, or stops after the first free croissant, you'll need to work with your data to get a better idea of what Sarah will respond best to. Or make better croissants.

You can totally still give your Sarahs a free coffee now and then

Because unexpected free stuff is both surprising and delightful, and who doesn't love that.