“Judging by Amazon’s success, the recommendation system works. The company reported a 29% sales increase to $12.83 billion during its second fiscal quarter, up from $9.9 billion during the same time last year. A lot of that growth arguably has to do with the way Amazon has integrated recommendations into nearly every part of the purchasing process…”
Amazon, being the the multi-billion dollar behemoth a lot of online retailers look up to makes them the perfect company to watch and learn from.
In 2015, they generated 107.1 billion in net revenue.
They have billions of data points to test so many different things to find out what works and what doesn’t, quickly. Over the years we’ve been watching the way Amazon does recommendations both on-site and off, via email. Plus we also found out that their recommendations via email convert better than their on-site recommendations…
And that bit of information got us very excited…
“Amazon also doles out recommendations to users via email… In fact, the conversion rate and efficiency of such emails are ‘very high,’ significantly more effective than on-site recommendations. According to Sucharita Mulpuru, a Forrester analyst, Amazon’s conversion to sales of on-site recommendations could be as high as 60% in some cases based off the performance of other e-commerce sites.”
35% of Amazon.com’s revenue is generated by its recommendation engine. (source) So what’s their strategy?
1) “Recommended for you, Thomas”
- Clicking on the “Your Recommendations” link on Amazon.com leads users to a page full of products recommended just for you. Amazon recommends a range of products from different categories you’ve been browsing, with the aim of putting products in front of you that you’re likely to click, learn more about and buy.
2) Frequently Bought Together
- This recommendation has one main goal: increase average order value. ‘Frequently bought together’ recommendations aim to up-sell and cross-sell customers by providing product suggestions based on the items in their shopping cart or below products they’re currently looking at on-site.
3) Your recently viewed items and featured recommendations – Inspired by your browsing history
- Here Amazon looks at products you’ve been browsing and recommends very similar products of different shapes, sizes and brands to help you find something very similar to a product you’ve already shown an interest in. They throw different brands, colours, shapes and sizes at you with the hope that they’ll place one product in front of you that you cannot resist.
4) Browsing History
- If you’ve already looked at a product, it means you were slightly interested and Amazon knows it, so they’ll show you your browsing history in case you want to quickly go back and buy something you previously showed an interest in.
5) Related to items you’ve viewed
- Just as it says, Related to items you’ve viewed displays similar products in different sizes, brands, etc., to products you’ve looked at in the past. This has the same goal as point #3 (to help you find a product you have to buy), but Amazon has labelled it differently on their site.
6) Customers who bought this item also bought
- Similar to #2 (frequently bought together), Amazon displays items that have been purchased together in the past, with the goal to increase average order value through up-sells and cross sells. My guess is that these items are purchased together a little less often than ‘frequently bought together’ and is a way for Amazon to sell items that are not as popular to help retailers move their inventory.
7) There is a newer version of this item
- People love to upgrade their gadgets to the latest version and this recommendation appeals to that need. If I look at the old Kindle I bought on Amazon.com, there is a recommendation underneath the invoice letting me know there is a newer version of the product that I can upgrade to. It’s almost like a replenishment campaign but for an electronic device… Smart.
8) Recommended for you based on a previous purchase
- After I purchased a Kindle from Amazon.com I was taken to an ‘order details’ section. Underneath it they recommend a variety of different cases for the exact Kindle I had just purchased in an attempt to encourage a second purchase with a highly relevant cross-sell offer.
9) Best-selling in “category” – ‘Discovery’
- Amazon.com recommends category top sellers for shoppers looking to try the new and latest products. ‘Best-selling’ adds a social proof element to the recommendation, that ‘other people are doing it and so should you’. Best sellers from a specific product category help people find popular products and buy from new categories they may never have purchased from before, which opens up someone to a whole new range of up-sell and cross-sell opportunities.
Off-site Recommendations with Email
Kwasi Studios wrote up a great post on the email follow up they received after browsing ‘Point and Shoot’ cameras on Amazon.com.
Here are examples of the recommendations Amazon proceeded to send them via email:
1) “This week’s best selling Canon models”
- The first email was a range of best selling models from a product category they visited. As only Canon models were showcased in this email, you can be sure they were browsing that brand of Camera on-site or even added a Canon camera to their shopping cart.
2) “This week’s best selling Kodak models”
- The following email was another ‘best selling’ recommendation from the same category (cameras), but showcased the Kodak camera brand the user would also have been browsing. Amazon’s is showing their most popular cameras as they know most people buy one of these Kodak’s and they think you will as well.
3) “Buy a camera package”
- This recommendation email contained items frequently purchased together, with the aim to get you to buy a camera and cross-sell its accessories in order to increase average order value and the amount of revenue they generate from each customer.
4) “Best sellers across the entire product category” – No specific brand
- This email contains top selling items from across the entire product category (digital cameras) the user was browsing. There is no focus on any specific brand, again they just display best sellers that most people end up buying. These items have the best reviews and conversions rates and will likely turn an interested browser into a customer.
Take notice – Amazon is only recommending products and brands that this person has viewed on their site or items they had added to their cart.
Highly relevant emails are critical for improving your click-through rate, conversion and revenue per email metrics.
If Amazon started sending this person discount offers for children’s books or outdoor tents when they hadn’t been looking at these things, there would be a disconnect and these emails would likely be marked as spam or they’d just unsubscribe from future mailings.
A perfect example of what not to do came from an email I received from iHerb.com. I purchased some fish oil and vitamin B supplements and they proceeded to send me this…
Lets just say I didn’t take them up on their offer.
Sending the right message to the right person at the right time is such a common phrase these days, but it’s true, and it’ll help you increase the ROI from your email marketing efforts.
How Amazon’s Recommendation Engine Works
There are in-depth discussions about how Amazon’s recommendations engine works.
Existing recommendation algorithms couldn’t scale to Amazon’s tens of millions of customers and products, so they decided to develop their own.
Amazon currently uses item-to-item collaborative filtering, which scales to massive data sets and produces high-quality recommendations in real time.
This type of filtering matches each of the user’s purchased and rated items to similar items, then combines those similar items into a recommendation list for the user.
Their recommendation algorithm is an effective way of creating a personalized shopping experience for each customer which helps Amazon increase average order value and the amount of revenue generated from each customer.
And That’s Why Rejoiner Has Created Its Own Recommendation Engine
We’ve been so fascinated by Amazon over the years (and so are a lot of our customers).
After we saw them rolling out their recommendations engine, we knew we wanted something similar incorporated into our software so that we could help online retailers increase increase sales by intelligently offering up and predicting what customers are likely to buy next and then dynamically serving those products into our customer’s lifecycle email campaigns.
How Recommendations Helps Increase Average Order Value, Click-Through and Conversions From Email by Intelligently Predicting What Your Customers Are Likely To Buy Next, And Serving Those Products In Your Rejoiner Email Campaigns
When a subscriber or customer receives an email from you, the goal is to get them back to your site to make a purchase.
Now with Rejoiner’s recommendations engine, you can intelligently serve people, top selling items or products that are frequently purchased together, inside your emails to increase engagement and click-through rate back to your online store.
Serving other products that are frequently purchased together in your emails, giving people to the chance to add more products to a cart they previously abandoned, which helps increase average order value.
Sending an 3-step abandoned cart email series with the last product someone left in their cart is a great way to get them back to your site, but if they’re not interested in that specific product anymore they’re more than likely just going to delete the email.
So if a product they left in their cart isn’t going to peak their interest, the recommendations engine can act as a another way to encourage every person reading your emails to come back to your store.
Blue Nile do this perfectly with their ‘cart abandonment email sandwich’.
The first email is a standard abandoned cart email – ‘you left this Diamond Stud Earing in your cart’.
The second email recommends different earrings from the same diamond stud category.
The third and final email again shows the original item the person left in their cart, combined with other recommended items from the same category. The goal being to get the person to either buy the product they originally showed an interest in, or to get them back into the buying process again with a different product from the same category.
Emails 2 and 3 do a great job of giving the person multiple options to come back to the store and to either buy or start the buying process again.
Here are a few more examples of how recommendations can be used:
- Recommend other top selling products in the same category along with items your customers have abandoned
- Recommend top selling products from across your entire catalog that most customers buy
- Recommend products that are frequently purchased together along with items your customers have abandoned to increase average order value
- After a customer buys something, follow them up post-purchase with products that are frequently purchased together to support their first purchase
How Rejoiner’s Recommendation Engine Works
The Rejoiner recommendations engine mines data from:
- Purchased shopping cart data
- Items added to carts but abandoned
All this data is fed into Rejoiner’s recommendations engine to help predict what your customers are most likely to buy next.
Over time as you ‘feed’ the engine more data, it gets smarter and smarter with its recommendations so that your email subscribers and customers are more likely to engage, click and buy.
Just like the nerdy kid at school that keeps studying more and more every day to get better grades, the more data you feed the engine, the more personalized the recommendations will be to your customers, and the more sales you’ll be able to generate.
What To Do Now
If you enjoyed this post you can subscribe here for future post alerts.
If you’re an online retailer that would like to start using advanced features like recommendations in your email campaigns, then I suggest you request an ROI report to calculate how your revenue from email after switching to Rejoiner.