Automation

Adding an automated reply

You can add automated replies simply by pressing “Add reply”. Adding automation saves you time because it answers incoming messages from customers for you.

First you give your reply a name, let’s say “Opening hours Easter”. Then you select how incoming messages are matched and analyzed. You can choose between Natural Language Processed (NLP) or keyword based. When selecting NLP, you will get a list of all predefined NLP’s in the system.

For keyword based replies you can set the keywords in the next step. You can also add keywords to NLP replies to make them more specific. Finally you fill in your reply and you’re good to go. Your reply can contain text but also hyperlinks to important pages on your website.

You will get a notification when an incoming customer request is not recognized by the automation. When this happens you have to answer the customer manually. Learn more about setting notifications here.

Example conversation:

  • Customer says “Hi”
  • Automation replies “Hi”
  • Customer says “What are your opening hours?
  • Automation replies “We’re open from Monday till Friday between 9AM and 6PM”

Difference between NLP and keyword based replies

There are two types of automated replies to set up:

  • Natural Language Processed (NLP) replies: Incoming messages are analyzed by artificial intelligence to understand the intent of a sentence. E.g.” What are your opening hours next Friday?”. It’s the most advanced way of recognizing what a customer request is all about.
  • Keyword based replies: With a keyword-based reply you can set specific keywords that a sentence needs to contain. E.g. “Opening hours” + “Easter”

NLP automations can be used as much as you like. You can also add keywords to an NLP automation to make it more specific. Let’s say you’ve selected the Opening hours NLP. By adding the keyword ” Easter” to it you can create your own “Opening hours Easter” NLP automation.

How to prioritize automated replies

Automated replies are handled from top to bottom in your list. When an incoming message can be answered by multiple automated replies, the one most on top will be triggered.  Let’s give an example:

Customer says “What are your opening hours?

You have two automated replies that can answer this request, one NLP “Opening hours” reply and one keyword based reply with “opening hours” as a keyword. The one most on top of the list will be triggered and will reply to the customer.