Complete Guide to Salesforce Einstein Bots

Share this article...

Einstein Bots are a powerful tool that use Artificial Intelligence (AI) to improve the customer service experience. Einstein Bots can live on many channels, including SMS, Chat, Slack, Facebook Messenger, and WhatsApp – and they support conversations in multiple languages.

The main benefits of bots are that they can answer routine queries without an agent being required, capture the necessary information before transferring to an agent and be served up to customers on their preferred channel.

The Value of Salesforce Einstein Bots

The value of Einstein Bots is that they are native to Salesforce, so you can use other customer information to automate the process. In addition, you can join these together with Einstein Discovery, Next Best Action, Prediction Builder, and Service Cloud Einstein.

READ MORE: Get Started With AI: 3 Freemium Salesforce Einstein Features

If you’d like to demonstrate the value of a bot, you can use the Einstein Bots Assessor (from Salesforce Setup) to work out ROI.

Screenshot from Salesforce Setup showing the Einstein Bots ROI calculator

Salesforce Einstein Bot Licensing

In order to use Einstein Bots, you need to be on Enterprise, Performance, Unlimited, or Developer edition.

As a feature, Einstein Bots is available with Service Cloud Unlimited Edition or Digital Engagement user licenses. Each applicable license is provided with 25 Einstein Bots conversations per month.

Recently announced as part of Sales Cloud Genie, Salesforce will shortly be bringing Einstein Bots enhancements to Sales Cloud from the Summer ‘23 release.

Key Terms

If you are new to Einstein Bots, some of the terms in this article will likely be new to you.

The “Messaging Channels” are currently:

  • WhatsApp
  • Facebook Messenger
  • SMS

The “Einstein Bots API” is used for your systems that are not natively supported, for example, Twitter.

In the context of bots, when referring to “Chat”, this is talking about whether you are within your app or on the web via Web Chat.

“Natural Language Processing (NLP)” uses a model set up to help your bot understand what your customer is saying and respond appropriately.

“Entities” ensure that the information provided is in the right format, e.g. postcodes.

Use Cases

The aim of a Salesforce Einstein Bot is to serve up what your customers need without needing a human agent (or Salesforce User) to be involved.

Some use cases I’ve seen are:

  • Providing answers via Knowledge Articles
  • Creating new inquiries as Leads
  • Creating new Cases
  • Searching for Records and providing updates on these Records (e.g. Orders, Cases)
  • Templated document collection
  • Collecting feedback
Screenshot of a Bot creating a case for a customer with an issue via Chat.

As a bot can connect to Flow or Apex, there are numerous other ways to automate your processes using clicks or code, and information inside or outside of Salesforce.

Although a bot is geared towards Service Cloud based scenarios, there is nothing stopping these from being used for Sales Cloud based situations too.

You also have the option to use Natural Language Process (NLP) to understand free text input and use Entities to ensure information such as Email Addresses are provided in the right format.

Einstein Bot Types – The Battle of the Bots!

Back in Summer ‘22, Salesforce introduced a new type of bot: “Enhanced Bots”. For parity, when this was introduced, existing bots were re-labelled as “Standard Bots”. At the same time, Salesforce brought us a new Bots Builder to make building a bot even more initiative.

Over the last few releases, Salesforce has built on the foundations of the Enhanced Bot and provided additional functionality that was never part of a standard out-of-the-box bot – for example, sending files or images.

The main difference between an Enhanced Bot and a Standard Bot is feature parity. Right now, when using an Enhanced Bot, rich content preview is not available. This means your Enhanced Bot can only display text content when using Salesforce Chat.

Left: A bot conversation in text preview. Right: A bot conversation in rich content preview.

The introduction template provided for bots is not available with Enhanced Bots.

If you are providing Apex, Flow, or external services, context variables then work differently with Enhanced Bots, and the menu displayed to customers works differently too.

Both types of bots support the Einstein Bots API, Chat, and Messaging Channels.

Let’s Talk about Best Practices

As part of becoming accredited in Digital Engagement, Salesforce expects you to be aware of some Bot best practices, as this is a core part of the exam.

  • Be sure to include as many utterances as possible to avoid failed understanding.
  • Use verbs in your menu titles so customers know what that option does, e.g. Check the status of my order.
  • When handling errors, consider routing to an agent or searching for Knowledge.
Screenshot of a Chatbot that’s about to search the Knowledge Base as it didn’t understand
  • Measure your Bot Performance using the templated Dashboards, and in the initial phases (after going live) review the log for any errors.
  • When using the same bot across multiple channels, consider the user experience because, for example, you won’t be able to send files via SMS!

This exam is only open to Salesforce Partners, however, best practices apply to all.

For those eligible for the exam, more information can be found on Partner Learning Camp.

Summary

Bots are a very powerful tool that you can use to automate your customer experiences. They are a great way to expand your operations without needing to grow your team, and they can be configured in clicks, not code.

Be sure to adhere to the best practices in this article when trying this out for yourself!

Other Resources

6 thoughts on “Complete Guide to Salesforce Einstein Bots

  1. Hi!

    If I need to have a bot in both English and French, do I need to create two bots, (obviously two buttons) and two deployments (for both Chat and Embedded Service)?

    1. From what I know- you can only have Bots that use English. I am unsure of the Fact, that you can change the language to French. For “Dialog Intent” purposes, your Bot will only support English. This is regard to “NLP” and your Bots ability to learn from past interactions with your customers. Therefore, making your Bot more effective and you creating less Intent utterances to maintain it and keep your users happy. From the deployment stand point you would need to create a new one for each unique channel. So your Bots are accommodating your users as needed.
      (Bots have no Limitations…Essentially.)

      Thats from my understanding.
      Hopefully, I was able to help a bit.

      1. Thanks for the answers.

        I can see in one of your screenshots that you have an Email entity with a Regex pattern validation. Does that work for you? For some reason, in my case it either redirects to another dialog or ends the chat. Is there something I am missing?

Add Comment