How we turn vague project requirements into defensible, standardized workload estimates β no guessing required.
And how an ISO standard changes everything.
COSMIC (Common Software Measurement International Consortium) ISO 19761 is the international standard for measuring functional software size.
Instead of guessing how long a feature takes, COSMIC measures how much data movement a feature requires. Every software feature can be broken down into four primitive operations β each counted to produce a COSMIC Function Point (CFP) value. The CFP value is then combined with calibration data to produce hour estimates.
Data moving into the software from a user or external system
Data moving out of the software to a user or external system
Data retrieved from persistent storage within the software
Data stored to persistent storage within the software
Each data movement = 1 CFP (COSMIC Function Point). A typical login feature has 4β6 CFP. A payment flow has 8β14 CFP. A full e-commerce platform typically measures 80β200+ CFP total.
Scopeslab extends COSMIC with a two-stage workflow that separates business understanding from technical execution.
AI analyzes your requirements documents and produces a comprehensive business understanding of the project. This stage is for project owners β no technical knowledge required.
The technical estimation stage, driven by calibration data and project-specific factors. Produces the final role-based workload breakdown for development teams and agencies.
Raw CFP counts are adjusted for real-world conditions using calibrated multipliers. No two projects are identical.
Adjusts each feature's CFP based on its functional complexity. Simple CRUD features score lower; complex AI/ML or real-time features score higher. Applied per-feature before project-level factors.
Five project-level multipliers: industry domain complexity, technology stack maturity, uncertainty level, integration density, and team composition. Combined into a single project factor.
Accounts for AI-assisted coding tools. Expert AI users see genuine productivity amplification β but only where AI can actually help (frontend, backend). Design, QA, and DevOps use separate factors.
Our estimation accuracy comes from real-world data β not guesses.
Real-world project data contributed by developers and studios globally. Each record contains feature type, CFP, and actual hours spent per role.
Features are categorized into 79 distinct tiers across 14 groups: from simple static pages to complex AI/ML pipelines, each with calibrated productivity benchmarks.
The COSMIC ISO 19761 standard is publicly available. Our two-stage process is fully documented. You can verify any estimate using the same methodology independently.
Every estimation report shows exactly how the numbers were calculated β feature by feature, factor by factor. No black boxes, no hidden multipliers.
While our methodology is transparent, your project data is strictly confidential. We never share calibration contributions with identifying information. All aggregated data is anonymized.
Our calibration database grows with every completed project. As more real-world data is contributed, every estimate becomes more accurate for everyone.
See COSMIC methodology in action on your own project β free AI analysis, no credit card required.