You are using an older browser that might negatively affect how this site is displayed. Please update to a modern browser to have a better experience. Sorry for the inconvenience!

Einstein Article Recommendations


Definition 

Einstein Article Recommendations recommends relevant knowledge articles to solve customers’ cases. 

Einstein builds a model using data from the past closed cases. The model: 

  • Proactively examines new cases as they come in 
  • Identifies key words, phrases, and text field values within the case 
  • Suggests articles that are most likely to solve the problem 

These recommendations appear right where agents work— in the Knowledge component of the Lightning Service Console. There’s no need to manually search for or sift through long lists of articles; agents can simply select the most relevant articles from a short list to attach to cases and send to customers. 

Prerequisites 

Salesforce Org with Spring’20  

Determining Article Relevance 

To determine an article’s relevance to an open case, Einstein Article Recommendations considers a number of factors. These factors include: 

  • Term overlap: How closely the case’s language matches the article’s language 
  • Attachments: How often agents attached the article to similar cases 
  • Dismissals: How often agents dismissed the article as not helpful on similar cases 
  • Term span: The distance between certain case terms in the article 
  • Longest common subsequence: The length of the longest common text sequence between the case and article 

Set Up Einstein Article Recommendations 

  1.  From Setup, in the Quick Find box, enter Einstein Article Recommendations and select Einstein Article Recommendations. 
  2. Click Review Terms to review and accept the Master Service Agreement. 
  3. Click the toggle to turn on Einstein Article Recommendations. It is now time to set up  model! 

Add the Knowledge Component to the Lightning Service Console 

  1. From Setup, in the Quick Find box, enter App Manager, and select App Manager. 
  2. Click the dropdown next to the Lightning Service Console app where you want to add the Knowledge component, and click Edit | Navigation Items. 
  3. Select Knowledge from the Available Items list, and add it to the Selected Items list. 

       4. Click Save. 

Select the Fields for Your Article Recommendation Model 

  1. From Setup, in the Quick Find box, enter Einstein Article Recommendations, and select Einstein Article Recommendations. 
  2. To create model, under Select Fields, click Select | Next. 
  3. Select the fields from cases that you want to incorporate into your model, and click Next. 

        4. Select the fields from Knowledge articles that you want to incorporate into your model, and click Save. 

Build and Activate Your Article Recommendation Model 

      1. From Setup, in the Quick Find box, enter Einstein Article Recommendations, and select Einstein Article Recommendations. 

      2. Make sure that you selected the case and knowledge article fields for your model. Then, under Build Model, click Build | Build Model. 

      Sit tight while the system builds your article recommendation model. This process can take up to 48 hours. 

     3. After model is complete, click Activate. When the model is active, agents with permission see article recommendations in the Knowledge component of the Lightning Service Console.  

Give Agents Access to Einstein Article Recommendations 

Add Einstein Article Recommendations to a Permission Set 

  1. From Setup, in the Quick Find box, enter Permission Sets, and select Permission Sets. 
  2. Click the name of the permission set where you want to enable Einstein Article Recommendations. 
  3. Click System Permissions | Edit. 
  4. Select View and Act on Einstein Article Recommendations. 
  5. Click Save. 

Add Einstein Article Recommendations to a User Profile 

  1. From Setup, in the Quick Find box, enter Profiles, and select Profiles. 
  2. Click Edit next to the name of the profile that needs access to Einstein Article Recommendations. 
  3. Under System Permissions, select View and Act on Einstein Article Recommendations. 
  4. Click Save. 

Einstein Article Recommendations in the Lightning Service Console 

  1. To see how relevant an article is to a case, look at the relevance score at the top of the article recommendation.  
  2. A higher relevance score indicates that the article’s content matches more terms that are found in the case’s content. 

  • To attach a recommended article to a case, click the dropdown next to the article, and then click Attach Article. 
  • To edit a recommended article, click the dropdown next to the article, and then click Edit as Draft. 
  • If an article isn’t relevant to the case, click Not Helpful next the article.