Naver has added new functions to its auto suggestions via a new technology dubbed, “entity linking”.This new update allows users to browse 15 results, to which Google’s 5 pale by comparison. In addition to this, additional search suggestions are also provided to users in a second column.This new feature is particularly useful for more ambiguous search terms, such as those with the same keyword but with different search intentions.For instance, the example screenshot below demonstrates the search response from Naver when the term “Moulin Rouge” is searched for.

This is a more ambiguous search term as it could refer to either the famous Parisian cabaret, or the movie based around it:


Another search that could pose problems due to having multiple potential search intentions is “Facebook”. The user could be looking for several things from this search, be it requiring general information on the social network or searching for the direct link to the social networking site.

If the link shown below in green is clicked, the user will be directly referred to the Facebook homepage. This result is demonstrative of Naver’s new autosuggestion tool providing to users the most likely desired destination of their search.


Naver’s new autosuggestion tool is not useful just for navigating search intent of the user more accurately, but for providing users with many different new options to glean more information from their search.

For example, when using the platform to search for the Fast and Furious movie, it displays several different elements of the film with a short information snippet on the right. This function allows users to directly select the part of the movie they are looking for.

A demonstration of this can be seen in the screenshot below.


Another similar example is a title of a story for which you can find different kinds of media. For the title searched for here, you can directly see results for the movie, the comic and the TV-broadcast of the release:


The main goal of this enhancement is to understand and foresee the user’s search intent and give them exactly the results they are looking for in just one click.

Whether adding additional rows and columns into the autosuggest section makes searches simpler or more complicated for users is debatable. After all, Google has a reason for showing only 5 suggestions from a usability point of view.

However, considering Korean websites and SERPs are designed in a much ‘busier’ way, whereby a multitude of information is compressed on a single page, stuffing the autosuggest section tool to the platform may prove to be a better, more localized update in terms of perceived usability in the Korean market.

The autosuggest function is likely to be further developed in future, potentially also depending on usage data.