Anthropic
Overview
The Anthropic Claude Analytics dashboard provides a centralized view of LLM usage, cost, and performance across projects using Claude models. It enables teams to monitor adoption, control costs, and optimize usage of LLM-powered workflows.Navigation
To access this module:- Go to LLM Metrics from the sidebar.
- Select Anthropic.
Time Filter
What is it?
A filter that allows users to view metrics over different time ranges.Why is it important?
Different time views help in short-term monitoring, trend analysis, and reporting.Where does it exist?
Top-right section of the dashboard (Daily, Weekly, Monthly toggle).Executive Summary
What is it?
A high-level summary of key LLM usage metrics and insights.Why is it important?
Provides quick insights for stakeholders and reduces the need to analyze raw data manually.Where does it exist?
Top-right section as a button labeled Executive Summary.Export Report
What is it?
A feature to download LLM usage data.Why is it important?
Enables external analysis, supports audits, and helps in sharing insights across teams.Where does it exist?
Top-right section as a button labeled Export Report.Metrics Overview
Total Cost
What is it?
Total expenditure on Claude usage over a defined period such as the last 30 days.Why is it important?
Helps track spending, manage budgets, and identify cost-heavy projects.Where does it exist?
Top metrics section of the dashboard.Total Tokens
What is it?
Total number of tokens consumed across all projects.Why is it important?
Indicates overall LLM usage and helps track adoption across teams.Where does it exist?
Top metrics section of the dashboard.Total Cache Tokens
What is it?
Number of tokens served from cache instead of fresh API calls.Why is it important?
Reduces cost, improves performance, and shows efficiency of caching.Where does it exist?
Top metrics section of the dashboard.Tabs
API Keys
What is it?
A configuration section to manage Anthropic API keys.Why is it important?
Required to connect to Anthropic services and ensures secure access.Where does it exist?
Tab under the Anthropic dashboard.Projects
What is it?
A view that displays LLM usage segmented by project.Why is it important?
Helps identify high-usage projects and enables cost allocation per team or project.Where does it exist?
Active tab under Anthropic dashboard.Models
What is it?
A view that shows usage distribution across different Claude models.Why is it important?
Helps compare cost versus performance and optimize model selection.Where does it exist?
Tab under the Anthropic dashboard.Organizations
What is it?
A view that aggregates usage at the organization level.Why is it important?
Useful for multi-team environments and provides leadership-level insights.Where does it exist?
Tab under the Anthropic dashboard.API Keys
Overview
The API Keys section provides insights into how individual Anthropic API keys are being used across the system. It helps track token consumption, efficiency, and usage patterns at a granular level. This view is essential for monitoring usage per application, service, or team that is tied to a specific API key.Token Consumption by API Key
What is it?
A time-series graph showing the number of tokens consumed by a selected API key over a defined period.Why is it important?
Helps identify usage trends, spikes, and irregular activity for a specific API key.Where does it exist?
First graph under the API Keys tab.How to Use
- Select an API key from the dropdown.
- Use the Daily, Weekly, or Monthly filter.
- Analyze spikes or drops in token usage over time.
Use Cases
- Detect abnormal usage spikes
- Monitor usage of a specific service or integration
- Debug sudden increases in cost
Cache Efficiency by API Key
What is it?
A graph showing how many tokens are served from cache versus newly generated tokens for a selected API key.Why is it important?
Indicates how efficiently caching is being used.Where does it exist?
Second graph under the API Keys tab.How to Use
- Select an API key.
- Observe trends in cache usage.
- Compare with total token consumption.
Use Cases
- Optimize cost through caching
- Identify redundant API calls
- Improve system performance
Input vs Output Token Ratio by API Key
What is it?
A graph showing the ratio between input tokens and output tokens for a selected API key.Why is it important?
Helps understand how efficiently prompts are designed.Where does it exist?
Third graph under the API Keys tab.How to Use
- Select an API key.
- Analyze ratio trends over time.
- Identify inefficiencies.
Use Cases
- Optimize prompt engineering
- Reduce unnecessary token usage
- Improve response efficiency
Token Usage by API Key
What is it?
A comparative bar chart showing total input tokens and output tokens consumed by each API key.Why is it important?
Provides a quick comparison of usage across different API keys.Where does it exist?
Below the ratio graph.How to Use
- Compare token usage across API keys
- Identify top-consuming keys
- Investigate disproportionate usage
Use Cases
- Cost allocation by service or team
- Identify heavy usage patterns
- Detect inefficient integrations
ROI & Business Impact Summary
What is it?
A summary section that translates LLM usage into business impact metrics such as time saved, cost savings, and ROI.Why is it important?
Helps stakeholders understand the value generated by LLM usage in business terms.Where does it exist?
Bottom section of the API Keys dashboard.Metrics Included
- Total Hours Saved
- Estimated Cost Saved
- Monthly ROI
- Annual ROI
- ROI Multiplier
Use Cases
- Executive reporting
- Justifying LLM adoption
- Measuring business impact
- Decision-making for scaling usage
Key Use Cases
Usage Monitoring
Track how each API key is consuming tokens over time.Cost Optimization
Identify inefficient usage patterns and reduce unnecessary costs.Performance Improvement
Use cache and ratio insights to improve system efficiency.Debugging & Issue Detection
Detect abnormal spikes or drops in usage.Business Impact Tracking
Translate technical usage into measurable ROI.Models
Overview
The Models view provides insights into how different Claude models are being used across the system. It helps analyze token consumption, cost distribution, efficiency, and performance at the model level. This view is essential for optimizing model selection and controlling costs based on usage patterns.Token Consumption by Model
What is it?
A time-series graph showing the number of tokens consumed by a selected Claude model over a defined period.Why is it important?
Helps identify which models are being used the most and how usage changes over time.Where does it exist?
First graph under the Models tab.Use Cases
- Identify most-used models
- Monitor adoption of newer models
- Detect abnormal usage patterns
Cost Usage by Model
What is it?
A graph showing the cost incurred by a selected model over time.Why is it important?
Helps track which models are driving costs.Where does it exist?
Second graph under the Models tab.Use Cases
- Identify high-cost models
- Optimize model selection
- Improve budget efficiency
Cache Efficiency by Model
What is it?
A graph showing how efficiently caching is used for a selected model.Why is it important?
Efficient caching reduces cost and improves response time.Where does it exist?
Third graph under the Models tab.Use Cases
- Improve caching strategies
- Reduce repeated computations
- Optimize performance
Input vs Output Token Ratio by Model
What is it?
A graph showing the ratio between input tokens and output tokens for a selected model.Why is it important?
Helps evaluate prompt efficiency.Where does it exist?
Fourth graph under the Models tab.Use Cases
- Optimize prompt engineering
- Reduce token usage
- Improve response efficiency
Token Usage by Model
What is it?
A comparative chart showing total input and output token consumption by model.Why is it important?
Provides a quick comparison across models.Where does it exist?
Below the ratio graph.Use Cases
- Model-level cost allocation
- Identify inefficient models
- Support optimization decisions
ROI & Business Impact Summary
Provides business impact metrics based on model usage, including ROI, cost savings, and productivity gains.Organizations
Overview
The Organizations view provides aggregated insights into LLM usage across different organizations or teams. It enables high-level monitoring of token consumption, cost distribution, and efficiency at an organizational level. This view is particularly useful for enterprises managing multiple teams or business units using LLM services.Token Consumption by Organization
What is it?
A time-series graph showing total token consumption by organization.Why is it important?
Helps track how different organizations utilize LLM resources.Where does it exist?
First graph under the Organizations tab.Use Cases
- Compare usage across teams
- Detect abnormal spikes
- Monitor company-wide adoption
Cost Usage by Organization
What is it?
A graph showing costs incurred by each organization.Why is it important?
Provides visibility into spending distribution.Where does it exist?
Second graph under the Organizations tab.Use Cases
- Budget allocation
- Identify cost-heavy departments
- Optimize spending
Cache Efficiency by Organization
What is it?
A graph showing cache utilization across organizations.Why is it important?
Efficient caching improves performance and reduces costs.Where does it exist?
Third graph under the Organizations tab.Use Cases
- Improve caching strategies
- Reduce redundant API calls
- Enhance performance
Input vs Output Token Ratio by Organization
What is it?
A graph showing the ratio between input and output tokens across organizations.Why is it important?
Provides insight into prompt efficiency across teams.Where does it exist?
Fourth graph under the Organizations tab.Use Cases
- Standardize prompt engineering
- Reduce token waste
- Improve efficiency
ROI & Business Impact Summary
Metrics Included
- Total Hours Saved
- Estimated Cost Saved
- Monthly ROI
- Annual ROI
- ROI Multiplier
Use Cases
- Executive reporting
- Organization-level performance evaluation
- Justifying LLM investments