A Simple Formula for High Converting Upsell Offers
Hey I’m Shane, welcome to another issue of The CRO Weekly where each week I explore how to build a high converting Ecommerce store. If you’re not subscribed you join 977(!) people that are right here:
If you’re on Twitter, you can follow me for bite-sized content on design, ecommerce, and CRO.
Upsells are nothing new.
Retailers have been using them for decades to add incremental revenue to every purchase.
Ever been to Dunkin donuts? Notice how they always ask if you want to add a donut (or more recently hashbrowns) to your order?
Or any shoe store that offers you some sneaker cleaner when you get a new pair of shoes.
Or famously, an offer that worked so well it was contributing to America’s obesity epidemic, McDonalds would ask if you wanted to ‘supersize’ your order.
Things are no different in Ecommerce. A good upsell or cross-sell offer can add small amounts of revenue to every order that can really add up.
And while plenty of apps will have you believe you need fancy “AI” (quotes used deliberately) product recommendations, it doesn’t have to be complicated.
In today’s post we’ll talk about how you can create a simple offer that adds incremental revenue to every sale.
How to Think About Upsells and Cross-sells
First, let’s start with some definitions.
An upsell offer is when you sell someone on the same product, but an enhanced version. Think McDonalds supersizing. They don’t add anything to your order, they just take you from Medium Fries -> Supersize Fries.
A cross-sell offer is when you sell someone a new, related product. Think of Dunkin and their offer to add a donut to your order.
Honestly, the lines get pretty blurry here so you’ll often see people (myself included) use the terms interchangeably. For the rest of this post I’m just going to say ‘offer’ when referring to both.
Before we talk about creating an offer, we need to talk about goals.
Adding Incremental Revenue
Repeat after me: the goal of an offer is to add incremental revenue. What does that mean? I think it’s best explained with an example.
Imagine you sell Shampoo. 9/10 times a customer that has bought your shampoo more than once is probably going to buy it again next month. So, whether you email them, advertise to them, send them a text, it probably has nothing to do with them buying your shampoo again.
If you send out an email campaign to that customer, and they end up buying more shampoo, you did not add any *incremental* revenue with the email even though that revenue will be attributed to email. They were going to buy regardless.
Incremental basically means, how much revenue did we earn on top of what we would have made without doing said thing.
It’s the same thing with offers. Imagine someone buys a shirt on your store. Before they go back to the collection you show them an offer for the matching pants and they add them to their cart. Sure, you might have improved the UX but there’s a good chance they would have gone back to the collection and added the pants anyway.
The question then is, how many additional people did we get to buy the pants with this widget. Otherwise we’re going to over attribute the success of any given tool.
Make sense?
What Makes a Great Offer?
Great offers are crafted, not found.
I mean that in a very literal sense because the best offers have a few traits that none of your products might have:
No variants
Cheap
Valuable to customer regardless of what they’re buying
Can be heavily discounted
Not available in the regular shopping experience
Let’s talk about each of these.
No Variants
We’re going for impulse buys here. When Foot Locker offers you shoe cleaner, there aren’t 15 different kinds. You’re buying basketball shoes and they have shoe cleaner for basketball shoes. A simple option like color, pattern, etc is fine but try to minimize decision making as much as possible.
Cheap
You want your upsell offer to be in the $10-$30 range depending on the cost of your average product.
Valuable regardless of what they’re buying
Regardless of what is in their cart, this offer should be relevant to them. Again, no matter what you get at Dunkin you can always add hash browns or a donut.
Can be heavily discounted
There’s plenty of debate around the efficacy of discounting. But, very smart people like Dave Rekuc believe in heavy discounts for adding incremental revenue. Modern Fuel sells $200 pens and usually not at a discount. BUT if you buy a second super fancy pen (something most people wouldn’t think of doing) they’ll give it to you for 50% off.
Discounting incremental offers does not turn people into coupon shoppers or hurt the value of your brand.
Not available in the regular shopping experience
Ideally your offer isn’t something people would usually be shopping for. If you sell shoes, most people shop your store for shoes. The shoe cleaner, even if people could find it on your store, isn’t what people are there for. If you start offering products that people would be shopping for anyway, then you’re getting into non-incremental territory.
Looking at an example
A great example of cross-sells that hit all of these points is Homesick candles.
The tissue paper
✅ No Variants
✅ Cheap
✅ Valuable regardless of what you’re buying
🛑 Can be heavily discounted
✅ Not available in the regular shopping experience
As you can see, not every offer needs to hit all of the points. But tissue paper is a great example of an offer that plenty of brands can add without much extra work. Similar things you can try are:
“Mystery gifts” - offer people 20-40% off a random accessory from your store
Fancy gift messages - similar to the gift wrap idea you can sell an upgraded version of the gift note
Package insurance - you’re probably covering the cost of products being lost, stolen, or damaged in transit so you might as well create it as a real product
The Wick Trimmer
✅ No Variants
✅ Cheap
✅ Valuable regardless of what you’re buying
✅ Can be heavily discounted
✅ Not available in the regular shopping experience
This is the poster-child of cross-sell offers. If you like candles you probably have heard that you should ‘trim your wicks’. So when you’re buying a new candle, you see this wick trimmer… and it’s on sale!? In your head you’re like “You know… I should already be doing this and I haven’t. I kinda feel bad not buying this for my poor untrimmed candles. And it’s on a discount!”
How well does it work? Taylor Sicard (worth a follow on Twitter) runs Homesick and says that 20-40% of customers take at least one of these offers.
The inevitable question of “AI”
When you start thinking about creating an offer you will inevitably hear someone say “just use AI!”
There are a bunch of tools offering some version of smart product recommendations. They look at your order history, etc and try to recommend a product that other people who ordered similar things have added to their cart.
I don’t know about you, but that whole value proposition doesn’t sound very incremental to me. These sorts of recommendations are great while someone is actively shopping but once you’ve got them on the hook (in the cart, checkout, or post-purchase) we’re trying to increase AOV not just get people to equilibrium
Side note: if people are using your cart to shop your store… that’s probably not ideal.
There are other downsides to dynamic recommendations too:
A lot of products on your store probably have variants
Most recommended products probably aren’t cheap
It’s really hard to properly discount a dynamically recommended product
All of these products are available in the regular shopping experience
Basically… it breaks all of our rules for a good offer.
That’s all for today’s issue! In other somewhat exciting news, I’m thinking of bringing on my first sponsor for The CRO Weekly. I will probably be doing something where I give them a shout out at the top of the newsletter for 4 weeks in a row. If you know any companies that might want to sponsor, please tell them to reach out on Twitter!
Thanks and hope you have a great weekend :)