Remember the last time you bought a present for a close friend’s birthday? You probably chose the gift based on what you thought they would like – maybe she showed you earrings that she would just love to have, or he told you that he would like to go for an advanced driving course. The wrapping you used was in their favorite color, and you chose a card that made you think of them. You wrote their name on the envelope, and the message you scribbled inside was personal and came from your heart. You also gave the gift on the right date – their birthday. You still smile when you think of the joyous expression on their faces when they opened their gift.

This is personalization. You went out of your way to carefully choose the right gift because you knew exactly what they wanted.

Online personalization is similar. But, since you don’t know the person on the other side of the website or email, the challenges are a bit different.

Online personalization occurs when the look, choices, functionality, or experiences are tailored to an individual’s needs to offer targeted, one-to-one content or products. Personalization is used to improve user experience and customer satisfaction.

Here are some examples of how personalization is applied:

• Newsletters or emails: The receiver is addressed by name, and the content is unique to that person’s interests or where they are in the buyer cycle. For instance, an email could ask a person whether they are still interested in products from a discarded shopping cart, and different emails would need to be sent to someone who just signed up for a newsletter compared to a loyal customer.
• Recommender software: Articles to read (news sites) or products to buy (e-commerce) are recommended according to the behavior and characteristics of the site’s visitor – “because you read this, you might enjoy this” or “others who bought this, also added this to their cart.”
• Mobile apps: App users can choose their likes and dislikes, and recommendations keep their current location in mind. For instance, dining apps let you choose your dietary preferences and recommend restaurants that are in your area.

Personalization Statistics

Now that you have a better idea of what personalization is, let’s look at some personalization statistics related to how consumers react to unique recommendations and offers:

• 88% of consumers are more likely to make a purchase if the retailer delivers personal products and connected experiences across different channels
• 74% of online consumers get frustrated with websites that feature content or offers that are irrelevant to them

It’s no surprise, then, that marketers who implement personalization see great results:

• Personalized emails deliver six times higher transaction rates
• Marketers who personalize web experiences see an increase of 19% in sales

Personalization is not without its struggles, though:

• 60% of marketers struggle to deliver real-time personalization
• 66% of marketers don’t have the necessary internal resources to personalize their marketing programs

Personalization Challenges and How to Address Them

It’s easy to search for a personalized gift for a close friend, but if you send out hundreds of emails or you have thousands of visitors to your site each day, there is just no way you can personalize these experiences and offers manually, especially since you don’t know these people personally. We need the help from technology.

Not all personalization technologies are created equal, though. Here are some of the tactics that are used:

• Visitors are added into groups or buckets according to demographics such as age, gender, and location. The same offers are then showed to the whole group. This is not always very accurate – you certainly don’t like all the stuff that your neighbor, of the same age and gender, likes.
• Apps let people choose what they like or don’t like. Though this can tailor experiences and offers considerably, your user doesn’t necessarily have the time or inclination to fill in your little questionnaire.
• When someone signs up for a newsletter, as a minimum their name is requested. More information, like gender, age, and profession can be asked. But, as in the first point, these demographics are still too broad.

For personalization to be really successful, you need to dig deeper. You need to be able to tailor your content according to the likes and needs of individual users. Personalization is best done with machine learning.

Here’s how LiftIgniter’s machine learning personalization works:

1. First, data on a visitor’s behavior, on and off-site, are gathered in real time.
2. The data is then broken down and categorized.
3. The data is then analyzed by machine learning algorithms.
4. The algorithms learn more about the person and can then predict what he or she will do or would like to see next.
5. Recommendations are offered to the user.

All of this happens in real-time. The visitor is surprised by accurate recommendations as they have no idea what has been going on in the background.

Personalization Examples in Everyday Life

Personalization is becoming more common by the day, and every time you are online, you are exposed to it. Here are some real-life examples you have probably already encountered:

• Amazon: A great example of a company that is doing it right, this e-commerce store analyses your current and previous behavior on and off-site with machine learning, and tailors its recommendations according to this.
• Netflix and Hulu: These sites recommend new content based on what you watched before and the ratings you have given shows. It will also suggest a new episode or season that is out from a series that you have devoured.
• Spotify, Google Play Music, and Pandora: Songs are picked according to what you have listened to before, what is in your playlist, or songs to which you gave a thumbs up or down.
• Bombfell and Stitchfix: These two sites let you answer a style survey which they compare to data from other subscribers. With the help of a computer, a human personal stylist picks outfits personalized to your taste, shape, and size, and you receive a neat package on your doorstep containing clothes you are bound to like.
• Social Media: Facebook, LinkedIn, and Pinterest are all masters of personalization. They will show you a personalized feed with content from the people that you most often interact with or pins similar to those that you have pinned before. They will suggest connections to people that you may know or with whom you have something in common. You will also see personalized ads.
• Google’s mobile app: This app gives you reminders from your to-do lists, weather and traffic updates for your area, and suggests articles to read according to your interests and recent views.
• Re-targeting advertising: You may have noticed that a few hours or days after you looked at a product on one site, an ad advertising that product will pop up on another site.
• Interest-based advertising: These ads will show content according to your interests. For instance, if you often read about machine learning, an article about the latest machine learning news may be advertised on another site you visit.

If you’re interested in learning more about personalization, or you want to experience what LiftIgniter’s personalization algorithms fuelled by machine learning can do for your site, get in touch.