A report is the core unit of work in LLM Insights. It defines which brand you're measuring, which competitors you're benchmarking against, which AI platforms you're tracking, and which questions are being asked. Every metric you'll later analyze rolls up from how you set up this report, so it's worth getting the setup right. 

This article walks through the full setup flow, from choosing a use case to launching your first report and managing it after it goes live. 

Creating a new report 

Navigate to the Record Manager and click + Create Report in the top right. This launches the report creation flow, starting with use case selection.

Select a Use Case

LLM Insights offers two use cases. Choose the one that aligns with your goal you can always build separate reports for the other later.

Brand Visibility: Tracks how your brand is represented, mentioned, and perceived across AI platforms. Best suited when your focus is your own brand's presence and sentiment, without formal competitor benchmarking.

Competitor Visibility: Compares your brand's AI visibility and sentiment against specific competitors. Best suited when your priority is share-of-voice analysis and identifying where competitors are outperforming you.

Once you've made your selection, click Submit to move into the three-step configuration form.

​

Configure the Report 

This step sets the boundaries of your report. Required fields are marked with an asterisk.

• Report Name: Give your report a clear, descriptive title (e.g., Acme_Competitor Benchmarking_Q2) that will be meaningful to teammates and easy to identify later.

• Brand: The primary brand this report will monitor.

• Competitor Brands: Up to five competitors to benchmark against. This field appears only for the Competitor Visibility use case.

• Additional Context: Optional. Use this free-text field to steer prompt generation toward specific themes, such as pricing, product quality, or customer experience. Greater specificity here produces more targeted prompts.

• AI Tools: Choose the platforms you want to track. Prompts will only run on the tools selected here.

• Report Refresh Frequency: Set how often the report reruns: Daily, Weekly, or another available cadence. The report will refresh automatically until the End Date is reached.

• End Date: The date on which automatic refreshes stop. After this point, the report moves to an Expired state and can be viewed but not edited.

• Note on prompt consumption: Each refresh cycle draws from your contracted prompt volume. Pricing is based on 50 active prompts at any time, regardless of how they're spread across reports or how frequently the report runs. While daily refresh is recommended for the most current insights, choose the cadence that matches how often you actually use the data.

• Click Next to continue.

Review Prompt Pillars 

• ​Prompt Pillars are the thematic categories that structure your report. They determine how prompts are organized and distributed across topics.

• Sprinklr AI suggests an initial set of pillars based on your selected brands, any additional context provided, and Social Listening signals from the past 60 days. Because these pillars are grounded in real customer conversations, they reflect what people are actually talking about rather than generic categories.

• Five pillars are suggested by default, each up to 80 characters. Review them with a critical eye: does each pillar represent a theme that genuinely matters to your customers and where AI visibility has business impact? You can keep, edit, remove, or add pillars as needed. At least one pillar is required to proceed.

• Click Next once your pillar list is finalized.

Review Prompts 

• Once your pillars are finalized, Sprinklr AI generates the actual prompts within each pillar. These are the real questions that will be sent to AI tools to test whether your brand is mentioned, how it's described, and what sources AI tools cite when answering. 

• By default, LLM Insights generates 50 prompts, distributed evenly across your pillars. If you've configured more than five pillars, distribution may vary. 

• Prompts are derived from real social listening signals, recent news coverage, and how AI tools tend to interpret and respond to your selected inputs. 

• Review each prompt critically. Keep the prompts that reflect how real customers ask about your category. Edit any that are close but not quite right. Add custom prompts to cover specific themes, product lines, or scenarios the AI didn't generate. Remove prompts that don't fit your business. You need at least one prompt selected to continue. 

• When you're happy with the list, click Generate Report. 

After Submittiing

Your report enters a generation queue. Initial runs typically complete within one to two hours, depending on the number of prompts and platforms selected.

An in-app notification will alert you when the report is ready. Once complete, a View Report button will appear next to the report name in the Record Manager.

There's no need to stay on the page generation happens in the background while you continue working in Sprinklr.