Estimation

Estimation is the practice used to size total delivery complexity consistently across teams.

Outcomes

Typical outcomes from estimation:

• Shared complexity language across the delivery team
• Better sprint planning and sequencing decisions
• Earlier visibility of delivery risk
• More predictable path to Done

T-shirt estimation model

To keep planning predictable and consistent across all Releaf projects, we use a unified T-shirt estimation system.

This estimate represents total complexity: Development, QA testing, and retesting or bug fixing.

Who sets the estimate and when

• QA adds the initial T-shirt estimate during Requirements Analysis.
• At this stage, a dev-only estimate may be misleading (for example, low implementation effort but high QA scope).
• QA already sees scenario complexity, logic risk, data touchpoints, and regression surface.
• The initial estimate is based on overall issue scope and QA complexity.

Dev may adjust the estimate when taking the issue into work (Todo to In Progress) if implementation effort differs, using the same sizing scale.

Unified T-shirt sizing

• XS: 1 scenario, minimal changes, almost no logic modification
• S: 2-3 scenarios, small logic updates, no impactful data changes
• M: 3-6 scenarios, medium logic complexity, data involved
• L: complex logic, multiple dependencies, several user flows
• XL: full features, integrations, major logic changes, many scenarios and edge cases

What the estimate includes

• Development work
• QA testing
• Smoke/regression testing before production
• Bug fixing
• Retesting
• Work with external dependencies (APIs, Stripe, DHL, ID verification, etc.)
• Clarification and communication loops

This is not just development time. It reflects total effort required to deliver an issue to done following our Definition of Done.