Rate Content

Design Pattern

Let users rate content in order to democratically help decide what is of higher quality.

User ratings act as a mechanism to handle risk for your users: is something worthwhile to spend time or money on? Promote community participation by letting users democratically decide what is of higher quality. Consider accompanying quantitative ratings with qualitative comments or reviews.

This pattern is much like the Vote To Promote pattern. It differs from the Vote To Promote pattern by having different outcome. The outcome is to allow users to guide other users about what is good and bad rather than to promote what is interesting.

The pattern consists of a number of mechanisms that work together:

• Voting mechanism. Provide a mechanism for your users to rate an item on a numeric scale. The most popular scale is 1-5, where a rating of one is worse than a rating of 5. A user gets one vote and can possibly add an explanatory comment along with the rating. When a user rates an item, feedback should be given back to the user informing them that the rating has been recorded.
• Display the average rating an item has received. The average of all ratings an item has received shows the perceived quality of an item, and will guide new users in whether an item is worthwhile.
• Display explanatory comments from users rating an item. An item can often be rated either low or high for a number of different reasons all originating from different users’ subjective opinion. A perceived flaw by one user is not necessarily a perceived flaw of another. To add a more quality and depth to the given ratings, allow the users to review an item by letting them explain themselves in free text.
• Show the highest rated items. Sum up the highest rated items in lists on a main page.
• Favor quality items. Favor items rated high in search results, when browsing tags, and showing related information.
• Related items. When showing one item, display its rating. Additionally, use the people who also rated the item highly, to create a list of related items, by showing other similar items these people rated highly.

The Rate Content pattern promotes community participation and can assist you in separating good quality content from bad quality content. This is especially useful when your website relies on user submitted content.

Rating content is about handling risk from the user’s point of view. Will a user on eBay cheat me or is a book on amazon worthwhile my time and money?

The main reason to use the Rate Content pattern is assist the user in managing risk. Examples of such risk are spending money on buying a product or from investing time on another user.

When implementing a rating system, you need to consider a number of things in relation to what it should be used for and how it should be communicated:

1. Who is doing the rating
2. What is being rated
3. What behavious are we trying to encourage or discourage?

Who is doing the rating

When reading reviews about an item and seeing how other users rate a given item a number of concerns are bound to be raised by the user: “Is the rating and review honest and authentic?”, “Are the users who rated a given item like me?”, and “Do the reviewers posses the relevant competence to review the product or service in question?”.

Amazon and Netflix have solved this problem by letting people rate the ratings: was this review helpful for you?

What is being rated

Be clear when communicating what is being rated and make sure that the user knows exactly what he or she is rating. If what is being rated seems ambiguous, the value of the review for other users in turn diminishes. For instance, if a website collects good computer deals that can be rated by the website’s users, is it the trustworthiness of the seller, the products, or the price that is being rated?

Consider making your rating system multi-dimensional, or simply be clear about what – exactly what – is being rated.

What behavious are we trying to encourage or discourage?

Like with the similar community driven patterns, several pitfalls can make or break the success of the Rate Content pattern. Are some users trying to promote or demote certain items? – perhaps their own or their competitors product – maybe their friends or foes. You might want to setup a number of measures to prevent users from misusing the system, such as limiting the maximum activity of a user, looking out for malicious activity, or promote trusted key users to count more than others

Design your rating system so that you encourage the behavior you want. Making your rating system multi-dimensional by breaking up the full rating into sub-parts that by their title explain exactly what you want measured (rated), is one way of solving this problem.