Chapter 3: Salesforce AI for Service

This chapter will cover the core service-related AI options in Salesforce. We will systematically look at Einstein Bots, Einstein Case Classification, Einstein Article Recommendations, and Einstein Reply Recommendations. As we do, we will cover the main features and configuration options available. For each topic, we will also discuss the pros and cons and what things an architect has to think about when evaluating the feature. We will also extend our Pickled Plastics scenario by building a bot that answers reseller questions.

In this chapter, we're going to cover the following main topics:

• Introducing Service Cloud Einstein
• Deploying Einstein Bots
• Optimizing support with Einstein Article Recommendations
• Speeding up chat with Einstein Reply Recommendations
• Alleviating manual data entry with Einstein Case Classification

After completing this chapter, you will know how to architect solutions using Service Cloud...

Technical requirements

To follow along with the hands-on parts of this chapter, you will need to register for an org with Einstein Bots enabled, which includes a standard Salesforce developer org. This must also include the Sales Cloud Einstein Dev Org, which you may have created in the previous chapter. If you need a new one, you can sign up for one here: https://developer.salesforce.com/signup.

The Code in Action (CiA) video for the chapter can be found at https://bit.ly/2WJhqbJ.

Introducing Service Cloud Einstein

Service Cloud Einstein is the customer service offering of the Einstein Platform. It consists of four core offerings:

• Einstein Bots, a complete framework for building and deploying bots across multiple channels while leveraging Salesforce's outstanding NLP capabilities
• Einstein Article Recommendations, a pre-built machine learning model that makes suggestions for relevant articles based on the text in a case and past case article attachments
• Einstein Reply Recommendations, another pre-built model that is applied on top of your historical chat transcripts to generate suggested replies based on the context of a chat conversation
• Einstein Case Classification, a model that is built on top of your case history and generates suggestions

As can be surmised from these features, the primary purpose of Service Cloud Einstein is to make your customer service agents more efficient and accurate. By deploying bots, you can free up...

Deploying Einstein Bots

Einstein Bots is a framework for creating chatbots directly on the Salesforce platform. In the following sections, we will explore how and why we should do this.

Einstein Bots basics

A chatbot or bot, at the most basic level of understanding, is a piece of software that conducts a conversation either in text or in speech. Odds are you have interacted with a bot recently. Maybe you use Siri for voice search or have Alexa read you the news. Or, you might have encountered one when reaching out to customer service or when trying to book a ticket for some upcoming event. It is fair to say that conversational user interfaces are growing faster than any other type of human-computer interaction at present.

More and more companies are jumping on the bot bandwagon and for good reason. Deployed skillfully, they can be powerful allies in the quest to improve customer service without going bust. In particular, they can reduce the length of time your customer service...

Speeding up chat with Einstein Reply Recommendations

Einstein Reply Recommendations gives AI-powered suggested replies to agents using Salesforce Chat. In the following sections, you will learn how this works and what you need to adopt it successfully.

Einstein Reply Recommendations basics

Einstein Reply Recommendations is a way for you to leverage your old chat transcripts to generate suggested replies to common customer queries using a machine learning model. It works similarly to standard Quick Text, which you would write manually in so far as the agent is concerned, but the actual text is generated from historical conversations. Before you start getting concerned about inappropriate suggestions surfacing from some quirk of the AI compromising your customer conversations, it is worth noting that only suggestions that you have reviewed and approved will be suggested.

When you turn on this feature, Salesforce will build a model for leveraging your historical chat transcripts...

Summary

In this chapter, we looked at the four parts of the Service Cloud Einstein offering: Einstein Bots, Einstein Article Recommendations, Einstein Reply recommendations, and Einstein Case Classification. We spent the most time and energy on Einstein Bots, learning how to design and configure conversational agents. Introducing bots can transform an enterprise and spending the time to deeply understand how and why will pay dividends in years to come.

We then moved on to Article Recommendations, a handy way to aid your agents in locating knowledge base articles. If you have a big article database on Salesforce and are using it to resolve cases regularly, this is the offering for you. We learned that Reply Recommendations is probably the weakest of the Service Cloud Einstein pack because while it seems highly useful to be able to generate replies based on past conversations, the fact that this does not happen continuously makes the feature likely to be time-limited. Finally, we...

Questions

1. What feature could you use to optimize your knowledge management workflow?
2. What are the four key elements you use to define a dialog in Einstein Bots?
3. When are Einstein Reply Recommendations refreshed?