Website Personalization Best Practices


A high percentage of the popular web services and many smaller sites change their content according to the different needs of their visitors. We collected some well-known examples of on-site personalization, that could help you decide, how much your site could gain from displaying altered versions to its users.

This article is the fourth part of a post series about personalization in marketing. Click here if you would like to read it from the start, or here to download as an ebook.

You probably don’t need to go any further then the next open tab in your browser if you’re looking for examples of on-site personalization.

The majority of the most popular web services use personalization tactics to some degree. Almost everything that is displayed to you on Facebook is affected by your previous interactions and preferences. Of course, among the 100.000 factors that define the exact order of the items on your wall, there are ones, which reflect on other users’ actions, or the freshness of a post. Still it’s safe to assume that personalization is at a central stage at a site, which carried out a psychological experiment in secret to see, what happens when they hide words describing negative or positive emotions from the posts of 700.000 users.

It’s lesser known that the other internet giant, Google also uses personalization, when displaying search results. A company’s algorithm takes into consideration your location, browser type, Google + interactions and search history among other factors when selecting the most relevant and useful results. If you visit a site regularly it’s link might be placed higher when you search for a term, that the can be associated with it. You can also find the reviews of people you added to your circles on Google +, among the search results. These factors affect around one or two percent of the queries.

netflix homepage good example for personalization

Recommendation engines

Many sites use personalized recommendations to enhance the visitor’s experience. As you will see, this method can serve fundamentally different purposes.

On two well-known sites, Amazon and Netflix a very high percent of the user activity is driven by a recommendation system. Amazon built its algorithm to battle with the familiarity of brick and mortar stores. They use behavioral data and compare your search and purchase history to other similar user’s decisions to suggest items that you are most likely to buy.

The goal of the Netflix recommendation system is to come up the films that you are presumably going to spend the longest time with and to encourage you to discover more from their thousands of movies. In order to narrow down the selection for the visitors, the company developed a very complex tagging system for their content. The algorithms of Netflix try to find similar viewing patterns between users, since they are very likely to represent similar taste, and decide which categories are going to pop up when you visit the site.

While the systems used by Netflix and Amazon are developed by hundreds of engineers, it is possible to start to use one without the resources these sites have, as many e-commerce platforms offer plugins that personalize the shopping experience.

Content personalization

The contents of a news site or a blog can also be modified using behavioral data, and there are services, which provide recommendations based the user’s previous interactions with a site’s articles, and the trending topics. Our company offers a deeper level of content personalization for publishers. We aim to provide different versions of a website to a visitors based on their Facebook likes and activities.

We find it very likely that content personalization is going to have a bigger role in the future, and fewer visitors will see the same versions of websites.

recommendation engines

Website personalization is easier than you think

While it might seem more difficult to personalize a website than emails, this doesn’t mean it’s not worth doing. Most website owners could benefit from a certain level of personalization, and it’s possible that modifying some parts of your site could also enhance the experience of users significantly.

Many smaller sites have already developed unique applications to provide a more tailored version of their contents to the visitors. You can find fashion sites, that offer clothes to their visitors based on their physical attributes, and the brands they like on Facebook, and there are furniture shops which narrow their selection, after asking a few questions about your preferred style.

Our experiences show that personalization can lead to a 10-50% uplift in conversion, depending on how well set up the strategy is. If you have a very large collection of products, or your blog covers a wide array of topics content personalization will almost certainly have great results!

The next part of the series is about digital footprints in brick and mortar stores. Click here to continue reading!

by Sáfár Attila
on Tuesday, March 29, 2016

Be the first who reads our next article!

Weekly selection of the best personalization, business intelligence and conversion optimization articles.

It's time to check your inbox! ;)

In order to continue, you have to accept our legal conditions.

If you want to sign up without facebook, you have to provide an email address.

With subscribing on our weekly newsletter you are agreeing to the Terms of Use and Privacy Policy.