Getting the Most Out of Cobalt 

Cobalt from Badger Meter is an analytics tool that helps utilities streamline operations through the use of statistical methods, machine learning, large language models (LLMs), natural language processing (NLP), and generative AI. It is integrated across BEACON®, AquaCUE®, BEACON Field™ (iOS and Android), EyeOnWater®, and BadgerMeter.com to support documentation search and data analysis. 

How Cobalt Works 

Cobalt enables users to ask questions (prompts) related to documentation or their utility’s BEACON data and receive context-based answers. 

  • Only the user’s question (prompt) is sent to the AI engine. 
  • No customer or utility data is transmitted outside the system. 
  • Documentation questions are matched with internal technical sources (from Badgermeter.com, BEACON Help, etc.) and summarized. 
  • Data queries are processed within the scope of the user’s BEACON environment—ensuring that data remains within that boundary. 

Access Points 

On BadgerMeter.com 

Users can enter documentation-related questions into the website’s main search box. Summarized answers are provided with links to more detailed resources. 

Within BEACON (and AquaCUE) 

With the Help drawer open, you can get answers to any question you might have about BEACON.

Select the Cobalt icon in the upper right to begin a query. Cobalt uses the user’s access level to determine if documentation, BEACON data, or both can be queried. 

  • On Monitor pages, prompts are available in the upper right corner
    and on meter cards for quick access to usage pattern analysis and issue investigation.
     
  • On the System Sync Data Exchange page, prompts provide one-click insights tied to specific file activities. 


On the iOS BEACON Field App
 

Three main functions are supported: 

  • Documentation search (manuals, help content). 
  • Data analysis within your utility’s BEACON environment. 
  • Actions in the field, including GPS updates, meter tagging, and image uploads.  

 

Tips for Effective Use 

  1. Write Clear, Specific Prompts.
  • Be direct and include necessary context.
    Example: “Show daily usage for meter 3481 over the past 30 days.” 
  • Use terminology found in our documentation (e.g., refer to “endpoints” instead of “radios”). 
  • Be specific.
    Example: Ask “How to repair a tamper on an ORION Cellular endpoint,” rather than “How to repair tampers.” 
  1. Refine as Needed.
  • Narrow broad queries.
    Example: “Leak” → “Configure leak alerts for residential meters.” 
  • Rephrase if results aren’t useful.
    Example: Try “consumption trends” if “usage data” yields incomplete results. 
  • Ask follow-up questions to continue refining. 
  1. Request Examples and Guides.
  • You can ask for step-by-step walkthroughs or example scenarios.
    Example: “Example of setting up a high-usage Event alert.” 
  1. Field Support.
  • Field staff can use the Cobalt-enhanced BEACON Field app to update location data, swap equipment, and upload images with minimal manual input. Such updates create a downloadable CSV file that can be used to update your CIS. 
  1. Error Handling
  • Include exact error messages when asking for troubleshooting help.
    Example: “Endpoint serial number ‘1562967666’ is invalid for endpoint type ‘J’.” 
  • You can also request common solutions or known fixes. 

 

Access Control and Privacy 

Cobalt follows role-based access controls: 

  • Utility administrators assign the Cobalt user role and manage permissions. 
  • Documentation and BEACON data access are scoped to the user’s assigned role. 
  • Access can be updated or revoked at any time. 

Privacy Features 

  • No cross-utility data access. 
  • All data analysis is performed within BEACON.
  • All queries and results are scoped to the user’s organization. 

 

History

  • Previous queries can be found in the History section for reference. 

 

 

Tagged with: , , , , ,
Knowledge Base

User Guide
Follow via Email

Enter your email address to get new content notifications by email.