How to secure your automatic generated content on twitter ?


In the last 24 months, a large amount of automated campaigns with customised content (customisation of pictures, videos, etc.) have been diverted by ill-intentionned “trolls”. The risk is that it may result in a negative image for brands.

How to avoid this problem? Like a Bird propose its customers to filter partcipations and anwsers thanks to a system that detects words from black-lists. These are substantiated and regularly updated to cover the maximum of possibilities.

What are these black lists and how to use it?

To avoid the deterioration of the brands’ image on Twitter by sending automatic responses to participants with offensive words or handle, we have developed a black-list that has many benefits.

1/ A complete list

Like a Bird provides its customers a list of words to ban (insults, infamous personalities, …). This list is not definitive, you can always : add a new one, modifiy it, update it, etc. It is composed of :

  • Our own list
  • Lists that were made and completed by Like a Bird customers before they started their campaigns
  • Regular updates to cover the latest “negative” trends on Twitter

The result : a complete list that ensures the security and the good execution of your campaigns

2/ How does it work?

Our bots are then able to detect those words in a tweet or a handle (name of a Twitter account), which is even more useful when the brand is automatically answering to its participants (automatic replies, automatic generated images, etc.). Therefore, before answering, the bot will make sure that there is no words from the black list in the tweet. If it detects one of these words, it won’t send any answer to the participant.

3/ Results

Thanks to these blacklists, regularly updated, none of our content customisation campaign made with our customers (betting, television network, car manufacturer, etc.) has been touched by a bad buzz via offensive / illegal content!

Go further?

It is also possible, for the customers who ask for it, to make available a human validation backoffice : the pre-generated answers go up to this backoffice, then our customers only have to quickly control and validate the tweet to post it!

Want to try?

Do you want to have more information about this operation? Do you have a specific need but did not found anything to match it in our proposed operations? Don’t hesitate to contact us and to test our chatbot on Twitter and Messenger to find the best operation that will meet all your expectations.

Leave a Reply

Your email address will not be published. Required fields are marked *