Meanwhile, discover strategies to boost conversion rates, transaction size, and customer engagement in our Idea Guide.
Personalization in ecommerce makes total sense: The more a customer is served relevant content, the more likely they are to spend more time on your site - and ideally, convert.
Apps like LimeSpot also make personalization a snap, providing data-driven product recommendations and offers on autopilot. Even with these resources available, many people still don’t get personalization quite right.
As the personalization engine behind thousands of ecommerce sites, we’ve rounded up some of the most common personalization missteps we’ve successfully guided our clients away from.
Get ready to level up your personalization prowess with this helpful guide to avoiding personalization fails.
If you’re going to do personalization, you really have to be ready to do personalization. This means actively spotlighting personalized product recommendations and offers in prominent places - including product detail pages (PDPs), the homepage, and collections pages.
A/B tests run by LimeSpot clients have proven time and again that the more exposure customers have to personalization, the more likely they are to act on it.
In short: Don’t relegate a single carousel of product recommendations or recent views to the very bottom of a mile-long PDP. Instead, put recommendations front and center.
Here’s a few of our favorite placement ideas:
Look, our personalization engine is super sophisticated, accurately linking products your customers are going to be more interested in buying based on the products they’re showing interest in or have already bought.
We’ve built LimeSpot to give our merchants the flexibility to let things run with hands-off continuous adjustment, or to select their own options for bundles, upsells and cross-sells, or even entire featured collections.
Let’s talk use case. If you’re a site that sells matching sets, in many cases, a shopper may want to buy matching items. However, recommended products could exist in multiple ways. If a luggage and bag store had a pink paisley print, and a shopper is looking at a tote bag in that print, the algorithm may surface the same tote bag in other prints, other tote bags in varying prints, or the same print on different styles and sizes of bag altogether.
A store owner looking for more control can actually run an A/B test using featured collections or rules to see which type of cross-sell works the best. From there, they can manually set the type of products that appear on each product page.
We find manual intervention especially valuable for items that are meant to be purchased in a limited matching set of two, or possibly three items, - for example, a scarf / beanie / mittens set, or a coordinating sports bra and shorts.
Remember: Sometimes personalization is less about serving up a product that a customer might be interested in but doesn’t ‘go’ with what they’re looking at. Instead, it might be more about upselling the customer to buy more based on spotlighting products that really do belong together based on specific criteria you, as the merchant, are uniquely aware of.
The customer journey rarely goes like this: See a product on a website, and buy it. Social media, online communities, reviews and recommendations, videos, and more can all factor into where, when, and why a customer is finally ready to hit the checkout.
So why should your approach to personalization be limited to just your site? Luckily, there are many easy ways to offer personalized product recommendations, messaging, and offers that go well beyond the ecommerce site experience.
Where to offer personalization beyond ecommerce sites:
If every customer behaved the same, had the same interests, and shopped for the same items, life would sure be a lot simpler. But that’s just not how it works. Yet many ecommerce sites offer a pretty universal experience for 100% of their shoppers.
Enter segmentation. The first step to segmentation, using a powerful engine like LimeSpot’s Segmented Experiences, is setting up types of shoppers (or segments) based on criteria you set. From there, customers will automatically be sorted into these unique segments. How you use them is up to you. For example, you may run an email campaign tailored to a specific segment of high order value shoppers. Or you may want to offer a special banner or pop up for customers that have been on your site before and left before checking out.
The types of segments you create and use cases to market to them are pretty limitless - and incredibly easy to act on once you’ve started building segments. Want the ultimate guide to segmentation? Check out our personalization ebook here.
Here’s a few ideas for how to segment customers:
You can even layer segments together to create an even more tailored experience. For example, a wine boutique may have a segment for customers that buy higher-end red wines. When the customer goes to visit the red wine section of the site, the collection may automatically bring red wines $50+ to the top of the collection page. Learn more about curated collections.
Tools like LimeSpot’s personalized recommendations have tons of different recommendation types, and styles. For example, you could create a bundle that pulls Frequently Bought Together items, or a carousel that spotlights Bestsellers.
The trick is recognizing that simply choosing one type of recommendation and placing it universally isn’t necessarily going to work. Different box types work best on different pages, at varying points in the online shopper journey. Bestsellers, Trending, or New Arrivals are great to spotlight on the home page, for example, while Frequently Bought Together or Related Items work great on a PDP. Don’t forget about Upsells or Cross-sells when a shopper’s heading toward checkout (via pop up, in the cart, or at checkout).
It’s not just about varying the types of boxes you place, but stacking the number of recommendations you provide in a single place. For example, the home page can display multiple types of boxes - highlighting Bestsellers at the top, Trending halfway down, and Recent Views at the bottom. PDPs can also have multiple (or stacked rows) of recommendations as well. More recommendations = more chances to connect with and delight customers.
When you read stats like 75% of consumers find personalization to be somewhat ‘creepy’, it’s time to tread carefully.
The main thing to consider when personalizing is to aim to delight customers and have them thinking ‘that was easy’ - not, ‘wait, how did they know that?’
To use an example, even if data pointed to a customer likely expecting a baby, the last thing you’d want to do is start sending them a ‘congrats’ message or inviting them to set up a baby registry the second they land on your site. That shopper may be struggling with infertility, dealing with a miscarriage, or simply browsing for a gift for someone else and not interested in children of their own.
What is fair game is promoting ‘related items’ in the baby category, when that customer makes a visit to a baby-oriented product.
Consider timing as well. A customer doesn’t necessarily want or need to be segmented the first time they arrive on your site (although real-time personalization is possible with LimeSpot). While your site experience should adapt to what the customer is interested in during their session, maybe wait till they’ve made a few return visits before you mark them as ‘Big Spending Sneakerheads’ or ‘Bargain-Hunting Yoga Fans’.
When in doubt - ask around. If someone would be unsettled to see an incredibly on the nose or assumptive offer, it might be time to reconsider.
This is a big one: There are a ton of ways to implement personalization, and not every strategy is going to work equally well for every business. But how could you possibly know that, unless you test?
LimeSpot’s personalization platform includes robust A/B/n testing capabilities that lets you set up unique site experiences and monitor them over time to see which ones are driving more conversions (or other metrics as desired - like time on site or clicks).
This is an essential way to understand what strategies work best, without going in 100% on just one of them.
Here’s a few ideas of things you can A/B/n test with personalization:
You set the parameters and depth of your experiment - as well as the length. And just because you’ve got one experiment mastered, don’t leave things there. Remember a golden rule of commerce: Always. Be. Testing. That includes going back on things you thought you knew to be true - revisit experiments regularly to ensure they’re still driving the best results possible.
Personalization may seem easy but there are some nuances to it that can take it from good to great. Luckily, platforms like LimeSpot do a lot of the heavy lifting when it comes to automating personalization. But like any tool, LimeSpot is only as powerful as the intent you put behind it.
Follow this guide to advance your approach to personalizing online shopping. Want a helping hand along the way? We work closely with our clients to build successful, continually optimized full-site (and beyond!) personalization strategies. Contact our sales team to learn more.
