Erfassen Sie Ihre Fortschritte
Trailhead-Startseite
Trailhead-Startseite

See How Reply Recommendations Work

Learning Objectives

After completing this unit, you’ll be able to:

  • Learn how agents use Einstein Reply Recommendations. 
  • Improve Einstein Reply Recommendations as needed.

See Einstein Reply Recommendations in Action

Sita is ready to see Einstein Reply Recommendations in action. Let’s follow Sita and Maria’s journey using Einstein Reply Recommendations for the first time as they shadow product support specialist Ada Belewa, who spends her day helping customers. 

Ada receives a new chat from a customer about an installation date mixup. As soon as the first message arrives, several reply recommendations appear in the Einstein Replies component. The first reply is just what she needs, so she clicks Post to send it as a chat response.

An agent’s chat window alongside the Einstein Replies component with recommended replies.

The customer responds with a follow-up question, and the recommended replies automatically refresh. Ada clicks Edit on the second reply to add an extra detail about their installation policy, then sends the chat.

An agent editing a reply from the Service Console before sending it to a customer.

Ada has quickly resolved the customer’s issue. The customer thanks her, and new recommended replies appear. Ada clicks Post on the top reply to send it in the chat. Because Maria added the contact name into the reply using a merge field when she was publishing it to quick text, the customer sees their name in the farewell chat. Ada ends the chat and the customer leaves happy.

Completed chat with Einstein Reply Recommendations.

Customize Your Replies 

Einstein Reply Recommendations turns previous replies into templates for your agents to use. You can customize these templates from the Quick Text tab at any time. For instance, let’s say you wanted to incorporate a more friendly tone and increase customer engagement. You can change a simple “Hello” into a “Good morning, how may I help you?”

The quick text record for “Hello” being updated to “Good morning, how may I help you?”

Improve Reply Recommendations 

While it’s a smart tool, Einstein’s AI power can only go so far without your help. Einstein recommends replies from your previous chats, and it’s especially loyal to your quick text record. If a customer’s message is too unique, Einstein can be at a loss for words.

Luckily, there are things you can do to help Einstein guide your agents in the right direction.  

Improve Einstein Reply Recommendations by reviewing your quick text often. You can create new replies in quick text and include them in Einstein’s recommendations by selecting Include in reply recommendations on the record. If your agents find better ways to phrase a specific reply, you can always designate agents to keep your replies current and consistent with your company’s voice and tone by editing published replies in quick text. 

Also, your agents’ actions in real time help Einstein learn which replies to recommend. When a reply recommendation isn’t right for the situation, click Not Helpful underneath the reply. Einstein gathers this feedback for future suggestions.

Not Helpful button highlighted beneath a reply recommendation.


Einstein Reply Recommendations is a great way to connect with your customers using the most up-to-date technology out there. This technology has helped Ursa Major Solar close more cases with its amazing products and equally amazing support team.

Einstein Reply Recommendations is also just one of the many tools offered by Einstein for Service. Close more cases faster, easier, and smarter by becoming familiar with the rest of the Einstein for Service features, and learning more about how these features can work together to improve your business. To learn more, check out the Make Service Cloud Smarter trail on Trailhead. 

Resources