Loop Release Notes: Screening Inappropriate Messages

Applicable Product(s): Pulse, Messenger

Although Loop asks guests to input their names and contact information when they submit feedback, there are some who will nonetheless provide it anonymously — and then use that anonymity to include inappropriate or abusive language in messages to front-line staff. Understandably, many managers will want to protect their team from such messages. That is the purpose of Loop’s masked and screened comments feature.

Loop draws from English- and French-language dictionaries to automatically mask or screen comments that include inappropriate words and phrases. When creating the dictionary, the Benbria team based it on the repository of words that Google Search blacklisted from its autocomplete algorithms.

Terms that are blacklisted for masking are replaced by stars within the comments. Screening, on the other hand, removes the comments entirely rather than just the offending words. The comments do not appear in the employee inbox and do not trigger notifications. From the perspective of the front-line employee, they simply do not exist.

Whether the client chooses to use comment masking or screening will depend on their sensitivity to the blacklisted terms. In cases where comments are borderline abusive, it is often easier to remove the comment altogether.

Administrators have some control over which terms trigger the screening and which do not. Administrators can access these dictionaries to manually add or remove terms as they wish. Comments containing words removed by admins will then begin to appear in the employee inbox.

Even when comments are screened, administrators can still have access to them. There are several reasons administrators might need to do so. For one, a manager might need to respond to a comment regardless of any inappropriate language it may contain. For another, a comment might have been mistakenly screened due to term ambiguity.

For example, a Benbria client once received a message that was flagged as inappropriate because of the word “toque”, which is another word for “hat” in English. After some investigation, it was found that “toque” also matches a word from the French dictionary of screened terms. It’s an obscure colloquial term for “crazy or strange person.” In this case, the Benbria client decided to whitelist the term.

Because of the need to access screened comments, it is now possible for administrators to export them in CSV reports. This offers managers the ability to address guest issues while also protecting their front-line staff from demoralizing or hurtful messages.

What do you think of our new features? Let us know in the comments, or on LinkedIn, Facebook, or Twitter!

Leave a Reply