Emma gives instant and precise answers to your customers’ common questions, so your team can focus on conversations where a human touch is needed.

By instantly answering common queries, Emma will resolve 32% of your customers’ common questions (on average), create a better customer experience, and save your team hours each week.

How does it work?

Emma understand your customer's questions by comparing them to a list of sample questions you curate. We call these sample questions the Training Questions. Each Answer has its own unique list of Training Questions.

When your customers ask questions in Messenger, the bot does more than find exact matches or keywords when comparing - it uses an algorithm to analyze the answer training as a whole, looking for patterns. Things like synonyms and phrasing variations are all accounted for.

When a customer asks a question in a conversation, the bot references its training before it decides how to answer.

  1. If the bot is confident that it has a match, it will respond with the answer bot flow
  2. If the bot is not confident, it will not respond and log the question as an Unanswered question

Training best practices

As the bot’s understanding of a particular answer is dependent on the Training Questions, the larger the sample of Training Questions, the higher likelihood that the bot will understand your customer's question. By effectively training Emma, you’ll increase its resolution and coverage rates and save your team and customers more time.

There are three key ways to train Emma to provide more resolutions: 

  • Train Emma to recognize each question - add at least 10 example questions to each answer.
  • Broaden Emma’s coverage - create answers that cover many different topics.
  • Deepen Emma’s coverage - create multiple answers that cover a single topic.

Read more about training Emma here.

What’s next?

Need Help?

If you have any further questions, please start a Live Chat. Just "Click" on the Chat Icon in the lower right corner to talk with our support team.