Types of Content Curation Tools

A way to obtain relevant information with a reduced level of effort is to rely on experts. Various platforms provide content targeted to specific industriy segments, markets or topics. The user can subscribe to specific feeds based on the relevance of the topic and the trust the user has of the expert who is curating the topic. Experts can adapt the content and present it in a way that is adapted to its audience. Experts need to allocate a good amount of resources to make sure they don't miss a piece of important information.

Pros: High understanding of audience needs, adaptability

Cons: Long process, dependence on specific people

Examples: Scoop.it, Thomson Reuters, Industry Publications

Most content curation tools use a series of algorithms to determine which content is popular and make it more visible to users. Popularity ranking assumes that the more a Web page is shared, the higher its “quality”. Inbound links, “Likes”, “+1’s” and tweets are some of the indicators used to determine that Internet users found a piece of content valuable. Focus on popularity reduces the noise surrounding an item. The limitation of popularity ranking is that it requires gathering users’ "votes" and therefore delays the discovery process. Popularity can also be manipulated by promotional techniques such as SEO and SMO.

Pros: Focus on quality

Cons: Easily manipulated, long process

Examples: Google, StumbleUpon, Digg, Reddit, Delicious, FlipBoard

Content personalization involves using technology to accommodate the different information needs between individuals. It surfaces content that is assumed to be more relevant for a specific user. There are two approaches and they might be combined:

Machine learning: Monitoring the browsing activity of the user to identify which content is more liked. For example, the types of links that are clicked more often and the time spent on specific pages are used to identify which content is more valuable to a specific user. One of the reasons Google built Gmail is to have users logged in when they perform a search and use this data to filter information on their behalf.

Signal sensing: Even when the user is not logged in, a large number of signals are used to segment users in order to filter the information for them. Such signals include location, browser, computer, screen resolution, and others.

Both approaches are based on assumptions. The big challenge is to understand the users' information needs. It requires understanding the user’s context. Technology still struggles to know when the users' needs have changed.

Pros: Individual experience

Cons: Slow response to change, based on assumptions

Examples: Facebook, Google, Amazon, Trap.It

The social graph is an increasingly used approach to curate information. The user’s connections act as filters by only sharing the information they find useful. This represents an improvement over popularity because users are more likely to trust what their friends are sharing. Some tools, such as Twitter or Facebook, provide ways to create lists that allows users to group particular connections together. The limitation of people-centric curation is that what interests users might not necessarily interest their social graph. There is some overlap but users still get information overload due to the high amount of unrelated information.

Pros: In real-time, high engagement, trust

Cons: Noise, dependence on connections

Examples: Facebook, Twitter, TweetDeck

This approach consists of representing the emergence of content over the Web. The massive growth in the volume of published content allows the identification emerging patterns. This allows users to identify specific aspects of the events that are of higher interest to them. Content curation tools using this method are valuable for users needing to know “what is going on” with topics that are evolving quickly over time. This is the approach we follow at Darwin Ecosystem.

Pros: In real-time, dynamic, user selection

Cons: Requires high volume of information

Examples: Darwin Ecosystem, SkyGrid