Back to Blog

Content Recommendation Engine for Digital Publishing

Content Recommendation Engine for Digital Publishing_1 (1)

As a digital publisher, you constantly face the challenge of attracting a loyal online audience to build a successful brand. The best way to address this problem is by providing impactful and personalized content that keeps the user engaged. Most publishers use content recommendation systems to enhance user experience, increase page views, and improve site ranking. 

The following digital publishers successfully implemented recommendation engines in their business to derive value:

a) The New York Times pioneered in implementing different personalization strategies on the basis of the reader’s interests, geographical location, etc.

b) Hearst Newspapers increased readership by offering smart article recommendations around trending topics and user interests.

c) The Boston Globe implemented personalization technology to customize marketing as well as editorial content.

In this article, we talk about content recommendation engines, how they work and influence the digital publishing industry.

Table of Contents

What is content recommendation?

Content recommendation is a tool used to suggest informative content that grabs the attention of online audience. When exposed to relevant and good quality content, the reader obviously keeps coming back for more. Hence, content recommendation helps businesses to enhance overall user engagement and brand loyalty. It also eliminates the bounce rate and increases website traffic.

How does content recommendation engine work in digital publishing?

To understand how a content recommendation engine works in the publishing industry, here is a quick example. In 2015, the engineering team at The New York Times had built an in-house content recommendation engine, which suggested articles to users by drawing insights from user preferences, interests, and reading history.

The screen grab below shows that whenever you subscribe to an email newsletter, you can choose the stories you want to read. You’ll get personalized news feeds based on your favorite picks ‘Morning Briefing’, ‘Today’s Headlines’, etc. delivered directly to your inbox. Further, you can opt for a daily, weekly, weekday, or ‘thrice-a-week’ newsletter based on your news reading appetite and subscription.

New York Times - border(Image courtesy: The New York Times)

You can create your personal newsstand by managing your topic feed as shown below. So, while on-the-go, you can skim through latest stories from ‘Business & Tech’, ‘Lifestyle’, ‘Arts & Culture'.

New York Times 2 - border(Image courtesy: The New York Times)

So, by now we are familiar with how recommendation engines work in digital publishing. Next, we take a look at some of the factors influencing content recommendation.

Related: Download this comprehensive cheat sheet to deploy machine learning on  time and on budget →

Factors influencing content recommendation

Recommendation engines use complex models to determine which portions of a story or article make the most impact. So, what are the factors influencing the functioning of a content recommender system.

  • Session attributes

  • Paid or unpaid subscriber

  • Engagement quality (time spent on a particular news story)

  • Topics and entities covered in the current story

  • Engagement Level with previous stories

  • User interests as per browsing history

  • Device used

  • Location

  • Demographics

  • User attributes

Some other factors impacting the quality of content recommendation, include:

  • Type of content - temporal or evergreen

  • Featured image selection

  • Widget placement

  • Frequency of updated news

A good quality content recommendation engine has the ability to map readers into different segments and deliver suggestions relevant to their interests. The next question is what are the benefits of having a good quality content recommendation. We discuss it in the following section.

Key Benefits of Content Recommendation

  • Drives traffic: A recommendation engine help in driving traffic to your site through personalized emails.

  • Increases visitor engagement: It engages customers by offering customized news stories based on their preferences and interests.

  • Enhances conversion rate: As viewers feel satisfied with the type of content suggestions, they are most likely to subscribe to the website for more blogs and articles.

  • Higher lead generation: If you provide high-value content to your readers, it increases your credibility and viewership. Targeting the right people by capturing their attention helps in lead generation.

The below bar chart gives an overview of the benefits derived from a content recommendation system: Increased Visitor Engagement by 73%, enhanced customer experience by 54%, improved conversion rates by 53%, and increased lead generation by 45%.

Canva(Image Source: Canva)

How the future looks?

As it’s rightly said, ‘Nothing comes for free’, so is the case with reading all your favorite stuff online. So when you start clicking on the recommended content you are prompted to opt for a subscription model. This method of restricting access to content is termed as a 'paywall'.

Paywall is an efficient method to limit access to the entire content thus facilitating paid subscriptions. It is a good way to derive profits for digital publishers and build a sustainable revenue model. Paywalls help in identifying and categorizing potential subscribers based on the content they like to read. It’s entirely the publisher’s discretion as to how much content should be available for free reading and the portion they want to keep behind paywalls.

The paywall model offers 3 different options:

Lead-In: Here, publishers offer a snippet or partial view of a news story and a reader needs to pay a subscription fee for accessing the complete article. A few top media giants following this model include The Wall Street Journal, The Times.

Metered: In this option, users are allowed to read 3-4 articles per month without any extra charges. After the stipulated limit expires, the user needs to subscribe and pay. Top websites, like Medium, The New York Times, etc. follow this model.

Here is an example of the message received from Medium after accessing 3-4 free articles in a month:

Medium membership 3-2(Image courtesy: Medium)

Hard Paywalls: This type of content is fully gated, and it does not crawl up in search results. A gated content doesn’t appear in any organic search results. Based on the first story you read, recommendation engines judge your behavior on the basis of 2 important parameters.

  1. What will be the reader’s next story?

  2. What is the reader’s impression of the stories offered by the digital news publisher?

Taking the above points into consideration, relevant and quality online content will be available at a certain price. Owing to the fast-paced lifestyle, digital publishers are searching for alternate avenues for generating revenue through online readership. Whether it's breaking news, new book release, or top global tracks, everyone wants instant updates at the tip of their fingers. Providing accurate content at the right time to the right audience is the need of the hour.

As a digital publisher, if you want to embrace AI and machine learning to stay updated with the latest trends, subscribe to Skyl.ai.

Check out the various solutions that can be built using Sky Platform here.

cheat-sheet

    

Comments