Let’s start with some facts. First up – product recommendations are not a new technology. For years, Amazon has managed to create bigger orders by suggesting items to complement what’s already in the customer’s cart.
Product recommendations are also hugely effective at improving conversion rates and driving up your order values. Recent studies show that browsers who engage with recommended products have a 70% higher conversion rate. When you combine this with the impact on order values from customers putting additional products in their cart, it’s plain to see how recommendations can grow an eCommerce brand.
Product recommendations are applicable to lots of industries – from fashion to consumer electronics. But for beauty brands in particular, they’re becoming an essential part of their online experience.
Product recommendations: the beauty case
There are many reasons why beauty brands should make product recommendations a priority in their tech roadmap.
Here are some of the main ones:
- Discoverability: Beauty brands have extensive product catalogs. A recommendation engine can effectively pluck suggestions from the depths of a site. Without them, it can be difficult for customers to find the right products for them.
- Complimentary products: Beauty sites tend to have lots of products that work well with each other. For instance, if a customer is looking into a new razor, they might need shaving foam to go with it.
- Dual strategy: Beauty brands have the added benefit of being able to cross-sell customers by suggesting additional products or upsell them by recommending a more premium choice. This creates multiple opportunities to increase revenue.
- Assistance: Beauty customers occasionally need guidance when it comes to making purchases. When programmed correctly, a product recommendation can provide assistance by making suitable recommendations, much like an in-store expert.
So, we’ve established that beauty brands and product recommendations are an ideal match. But the technology has been around for years – why are we talking about them now?
Why it’s time for product recommendations to shine
Many beauty brands already have a product recommendation engine to upsell and cross-sell customers. However, many of these lack refinement in terms of relevancy, positioning, and effectiveness.
Only a small percentage can suggest products throughout the customer’s journey, rather than as a one-off. Furthermore, with concepts like AI making it easier to factor in historic data to create true, 1-2-1 recommendations, there is huge scope for innovation.
We at RevLifter have a few thoughts on why now is a great time for your beauty brand to get a recommendation engine.
Improvements in technology
Platforms like RevLifter make it possible to launch recommendations through a plug-and-play setup. This leans on the use of a tag placed on the retailer’s site, which enables the serving of recommendations, based on the customer’s on-site behavior, via overlays or ‘stickier’ tools like the Offers Wallet.
Before the dawn of tags, beauty brands would have to commit a serious amount of time and resources into building their recommendation engine in-house. The tag-based route to implementation is designed to be rapid. Seriously – you could be serving your first recommendations within a few weeks.
Improvements in relevancy (sometimes…)
Product recommendation engines are also a lot more intelligent these days.
I’m sure we’ve all laughed at some of the ridiculous suggestions we’ve received while browsing an eCommerce site. With several years of testing, launching, and optimizing under our belts, we’ve seen a vast improvement in the relevancy of recommendations.
One of the key developments is the use of real-time customer behavior, like exit signals and interactions with specific pages, to personalize recommendations. Using this information as the base means you’re instantly reacting to the customer’s every move.
What the future holds
It’s also interesting to factor in exciting concepts like artificial intelligence.
That’s where product recommendations go up a level by considering past purchases and interactions. The result – being able to predict what the customer might be interested in based on lookalike journeys.
How to create a beauty product recommendation engine
It’s not particularly difficult to launch product recommendations on your beauty site. However, there are a few questions you’ll want to ask yourself before getting started.
First, you’ll have to consider whether you’re recommending products as part of an overlay system or through a stickier tool.
Overlays
Often triggered when a customer adds something to their cart, overlays immediately present a recommendation to consider.
These are best served either on the product page as soon as the item is added to the cart, or on the checkout as the final chance to stretch the order value. It’s important to note that overlays are designed to be interruptive. You have to nail the execution, otherwise they provide a terrible experience for customers.
To show you a well-executed overlay, here’s what I see on the checkout page at men’s grooming brand Harry’s after adding a razor to my cart.
Sticky tool
The more conventional product recommendation engine stays with the customer as they progress on their journey. It dynamically changes its suggestions depending on the customer’s actions, and gives the experience of a personal shopper providing guidance.
Here’s a great example from one of RevLifter’s brands, Face the Future, which serves recommendations in their offers wallet.
Types of beauty product recommendations
Next, you’ll need to decide which types of recommendations to serve. Our favorite options for beauty brands include:
- Best sellers (based on what’s currently being purchased)
- Trending (based on what’s being viewed)
- Sale items (recommends items from your sale category)
- Frequently bought with (based on cart contents and purchase data)
- Previously viewed (for users that viewed a product without adding it to their cart)
- Margin booster (recommends high-margin products/categories to customers with low-margin products in their cart)
- Recommendation deal (supplies a deal on purchases of additional items)
These recommendations are perfect for beauty brands. Results aren’t guaranteed though, and you can launch terrible (or incredible) campaigns with each.
The difference between a good recommendation and a bad one tends to come down to just a few factors:
- Relevancy in the context of the customer’s cart or interests.
- Timing the recommendation to appear precisely when a customer might consider it.
- Positioning the recommendation in the right place to increase engagement.
It goes without saying, but you should also keep everything as personalized as possible. Broad recommendations serve no purpose and are swiftly ignored. Good recommendations drive benefits to you and the customer.
Final tips for launching product recommendations
Now you’ve nailed your solution, implementation, and recommendations, we’ll leave you with a few pointers for launching a beauty product recommendation engine:
1. Tag your data properly
It won’t be possible to serve many (if any) of the recommendations listed above without tagging your data in the right way.
Categorising your products according to sales and margin broadens the scope of your recommendations. At a minimum, you’ll need to tag them in a way that creates links throughout your catalog. That way, you can suggest a product that compliments the customer’s existing cart.
2. Consider past purchases
There are very few golden rules regarding product recommendations. One is that no customer wants to receive a recommendation for something they’ve already bought…
Unless you’re a beauty customer.
This is one of few industries where customers are often receptive to buying items they’ve tried and tested. Encouraging the customer to ‘buy it again’ when they return to your site is a great idea provided you’re working with accurate purchase data.
3. Consult with an expert
Product recommendations create tons of opportunities to grow your business in ways you might not have considered. However, you might need an expert to spot them.
RevLifter has lots of suggestions for beauty brands in particular. For example, you could try recommending a gift set from a specific brand when the customer already has one of its contents in their cart. The best technologies can even show how much the customer saves by purchasing the set vs a single item.
Fancy another suggestion from an expert? We’ve seen a growing trend in beauty resellers creating an extra stream of revenue by running supplier-funded promotions via their recommendation engine. This sees the brand pushing relevant suggestions from brands that have paid for extra promotion.
