Counsel
Counsel is an open-source AI financial advisor designed for small and medium businesses, providing a complete C-suite experience in a single application. It uses eight on-device statistical engines to analyze a business's financial data and deliver insights that are transparent and auditable. Unlike conventional AI copilots that generate polished summaries, Counsel attaches a 'receipt' to every claim, showing the exact method used, the confidence level, and the completeness of the underlying data. The tool emphasizes honesty by refusing to fabricate numbers when the math cannot support a claim, instead clearly stating its limitations. The platform covers essential CFO responsibilities, including trailing 30-day revenue analysis, net margin computation, and 13-week cash-flow simulations. It also includes advanced features such as customer survival analysis to detect quiet churn, structural break detection for revenue shifts, and graded decision tracking for price experiments. Every calculation is performed on the user's device, ensuring privacy and enabling offline operation, even in airplane mode. The tool is built for businesses that typically lack a finance department, such as food trucks, salons, and small firms, and it integrates data from multiple sources into a unified ledger. The project is open source, with the core application and its underlying Aurora engine available on GitHub. It offers a live demo for evaluation and is positioned as an honest, math-driven complement to existing accounting and analytics platforms. Counsel aims to replace anecdotal advice with evidence-based guidance, making it a practical alternative to fractional CFOs for cost-conscious small business owners.
- Category
- Database
- Pricing
- Free
Problem solved
Small business owners often cannot afford a full-time CFO or finance team to interpret their financial data. Counsel solves this by providing AI-powered CFO, COO, and CEO-level analysis with transparent statistical methods, on-device privacy, and a refusal to guess. It transforms raw ledger data into actionable, auditable insights, tracks decisions over time, and gives owners confidence to make data-driven choices without needing a data analyst or expensive consultant.
Use cases
- Get a daily morning brief of revenue, transactions, and customer metrics
- Run a 13-week cash-flow simulation to determine a safe owner's salary
- Evaluate a price increase with expected margin and confidence intervals
- Detect structural breaks in revenue with statistical significance
- Analyze customer churn and repeat-purchase survival patterns
- Monitor daily metrics against weekday patterns for anomalies
- Make tracked, graded business decisions instead of one-time insights
Audience
- Business Users
- Startups
- Developers
Pros
- Every claim includes a transparent 'receipt' with method, confidence, and data completeness
- Runs statistical engines locally, preserving privacy and working offline
- Refuses to make claims when math can't support them; no fabricated numbers
- Includes decision tracking and graded experiments (e.g., price tests)
- Free and open source with live demo
Cons
- Not a replacement for a certified accountant or tax professional
- Requires connecting financial data from multiple sources
- Early-stage project with potential integration limitations
- Statistical outputs may be difficult for non-analysts to interpret
- Self-hosting or technical setup may be needed for full control
Official website: https://counsel-site.pages.dev/
Canonical LambdaBase page: https://www.lambdabase.com/tools/counsel