Get Insights with Search Analytics
Learning Objectives
After completing this unit, you’ll be able to:
- Explain how Search Analytics provides insights with analytics dashboards.
- View trends in search data with built-in dashboards for AI-Generated Search Answers.
How Does Search Analytics Work?
Search Analytics works with Data Cloud using data lake objects (DLOs), data model objects (DMOs), and calculated insights objects (CIOs). When you enable Search Analytics in orgs with CRM Analytics, you can access dashboards to view trends in your search data. Use these trends and insights to change the search engine, optimize content, and train users. To enable Search Analytics for Einstein Search, see Enable Search Analytics.
Einstein Search records any search query, search response, or user interaction with a search result in search logs. With Search Analytics, you get standard DMOs and CIOs to store search-related events and metrics. Metrics are aggregated over events and computed by using formulas in the CIOs.
Additionally, include user data in your Search Analytics metrics to get targeted insights. Learn more about Search Analytics DMOs and CIOs in Search Analytics Reference.
To use Search Analytics, you need a Data Cloud license and Data Cloud credits. For more information, see the help article, Billing Considerations for Search Analytics and check out the Data Cloud Credit Consumption: Quick Look badge.
You access Search Analytics dashboards from Analytics Studio and view out-of-the-box dashboards for various search events, including Einstein AI-Generated Search Answers.
Awinha Oahu, the admin at Get Cloudy Consulting, a high-tech consulting firm, turned on AI-Generated Search Answers in Get Cloudy’s agent help portal. AI-Generated Search Answers uses large language models (LLMs) to generate responses to user questions from its knowledge base articles stored in the standard Knowledge object or in Data Cloud data model objects. Get Cloudy’s executive management wants to understand how well the Search Answers feature and its own knowledge base content help agents respond to customer questions. Agents give feedback on responses using the thumbs-up and thumbs-down feature in the response.
Einstein Search Analytics already provides out-of-the-box dashboards for various Search Answers components with built-in CIOs. As the system admin, Awhina Oahu can view the metrics computed for AI-generated Search Answers components using the CIOs and show the metrics to the executives.
Before going to the dashboards, Awhina Oahu, who also has Data Cloud admin permissions, can view the metrics aggregated in the related Search Analytics CIO using Data Cloud’s Data Explorer.
Once you enable Search Analytics and start collecting search-related data, you can view the built-in dashboards in Analytics Studio.
- From App Launcher, select Analytics Studio.
- Select Search Analytics Dashboard.
- Click Show More to explore the various components built-in for Search Answers.
Awhina Oahu can show the dashboards for the following components to view insights about AI-Generated Search Answer queries and responses.
-
Queries per Day: A chart that shows the total number of Search Answers queries per day.
-
Number of Interactions by Type per Day: A chart that shows the total number of interactions by search feature in Search Answers, such as copy answers, show more, or thumbs-up.
-
Precision: A chart that shows precision over time. Precision evaluates the accuracy of a recommendation or search system and tells you how many results are relevant.
-
Percent Positive and Negative Reactions to the Search Answers per Day: A chart shows the percentage of positive and negative reactions to Search Answers per day. A query reaction is positive if the user clicks or hovers over the result, gives the result a thumbs up, or clicks Show More. A query reaction is negative if the user gives the result a thumbs down. If the user doesn’t click or hover in the result, give the result a thumbs up or down, or click Show More, the query isn’t given a reaction rating.
For all the components and dashboards available for Search Answers, check out, Understand the Search Analytics Dashboard.
Understand how the search is performing in your Salesforce org based on these dashboards. Monitor and assess content performance. Use this information to customize search, optimize content, and train users.
- Customize your search by creating synonym groups, promoting articles in search results and search terms in knowledge articles.
- Optimize content by adding, changing, or removing content such that the content performs better, serving agent needs.
- Train users to address search behaviors such as providing feedback.
In conclusion, Einstein Search Analytics provides out-of-the-box dashboards for various search components with built-in CIOs. In the next unit, learn how to discover more insights from generative AI feedback and audit data by building custom CIOs