Building Smart Customer Service using AI with Low Code

Admit it, customer service is always a headache for most of the company, especially when dealing with nasty customer, it is not only taking time, but also mentally demoralizing. Let’s list down some of the pain points, we may not addressed all, but most of it

  1. Nasty customers’ emails are demoralizing, and need to take time to write a proper response, even for a experienced customer service personnel
  2. It take time to train the junior staff on how to response to the email in proper manner, be it complain or compliment
  3. Different personnel might handling different issue, example, James, as public relationship to handle customers’ compliment, and Joe, 1st level technical support to handle customer complain as soon as it get

With the help of AI, what we can do to help by designing a flow are

  1. Categorize the email by asking AI to read for us first, be it Spam, Compliment, or Complain
  2. Based on the category, the emails are going to different route, for different personnel
  3. All emails, will attached with a suggested response by AI, to ease the time to author an email response.

So we've got a problem

Customer service is always not an easy task, and some company handle with complicated flow as well. For now, we will create a simpler flow, for the simulation of customers’ email escalation with low code.

Here are the criteria

  • Complain email to log a ticket in customer support platform, usually 1st level of the support will get it and take a look first
  • Compliment email will send to another personnel to handle, usually public relationship manager
  • Spam email, we will not handle for now – of course, you can always build another flow to handle that
  • Each email will be having a suggested reply

Business Flow

We have 3 sections here

  1. Incoming Email: Entry point where the process started
  2. AI: Engine behind the scene which analyze incoming email
  3. Support: The real human who being read and process the problem / complain stated on the email
  4. Public Relation: The real human who being read and response to the compliment nicely

 

Below is a simple flow chart to help you understand better.

Test Data

We will need some test data to start with, let’s us simulate the 2 type of customers for a Beauty Shop

The Solutions

At lowcode101.com, our aim is to reduce the time to coding as much as possible, so that we have time for our video games instead of pathetic troubleshooting. To achieve this, some time we have to mix different tools together.

In this case, mixing different tools together is can’t avoidable.

At the point of making the tutorial, OpenAI just released the new engine, GPT-3.5 Turbo, with slightly different API parameters, but with better form of response
  • OpenAI GPT-3.5 Turbo x Email x make x Zendesk

    OpenAI GPT-3.5 Turbo will be the AI brain, Email will be the customer service entry point, Zendesk will be the helpdesk platform, and make will be the platform to integrate all 3 systems.

  • OpenAI GPT-3.5 Turbo x Email x make x NocoDB

    OpenAI GPT-3.5 Turbo will be the AI brain, Email will be the customer entry point, and NocoDB to become the helpdesk platform, and make will be the platform to integrate all 3 systems.

Leave a Reply

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

Prev
OpenAI GPT3 x Airtable x Make: Low Code Smart Contact Form

OpenAI GPT3 x Airtable x Make: Low Code Smart Contact Form

AI no longer a rocket science, with the help of Make, we are now able to make

Next
OpenAI GPT-3.5 Turbo x Email x make x Zendesk: Low Code Smart Customer Service

OpenAI GPT-3.5 Turbo x Email x make x Zendesk: Low Code Smart Customer Service

Having different scenarios to handle in your customer service?

You May Also Like
Total
0
Share