WBS output format

Every Ptah project can export a Work Breakdown Structure (WBS) as an Excel workbook. The output is designed to be a first draft that a PM can review and hand directly to a client. This page documents the structure and the estimation rules so you know what to expect.

Workbook sheets

The exported file contains five sheets.

1. Module breakdown

One row per task, grouped by module. Columns:

| Column | Description | |---|---| | Module | Name of the module (e.g., User Authentication) | | Task | Specific task (e.g., Login API + UI) | | BE (h) | Backend hours | | FE (h) | Frontend hours | | QC (h) | QC / testing hours | | Design (h) | UI / UX design hours | | DevOps (h) | Infrastructure and pipeline hours | | Total (h) | Sum across all roles | | AI Support | Whether AI can assist this task (Yes / No) | | Assumption | Notes for this specific task |

If a task only needs one role, the other role columns are zero. Tasks are never duplicated per role.

2. Summary by module (with delivery tiers)

One row per module with role totals and three delivery tiers:

Comfortable = Base × 1.30 — generous pace, thorough code review, full QC coverage.
Standard = Base × 1.15 — normal pace, enough buffer to handle unexpected issues.
Tight = Base × 1.00 — no buffer, continuous work, acceptable quality but no safety margin.

The tiers let a PM present the client with options rather than a single number.

3. Summary by role

Total hours per role, multiplied by your organization's rate card to produce a cost per role. The bottom of the sheet shows the total cost for each of the three delivery tiers.

Rate cards are configured per organization under Settings → Rate card. Ptah ships with a default rate card that you can override.

4. Timeline

A Gantt-style timeline estimate based on team size and module dependencies. Three variants are shown, one per delivery tier. The timeline is illustrative — it assumes a fixed team composition and does not account for vacation, sick days, or other real-world variables.

5. Assumptions and risks

Captures:

All assumptions applied during estimation
Items explicitly marked as out of scope
Identified risks and suggested mitigations
Notes on which tasks AI assistance reduced effort for, and by how much

Estimation rules

Ptah's WBS generator follows a set of rules that match what experienced outsourcing teams use in practice:

Minimum task size is two hours. Even a "trivial" task carries setup, review, and deploy overhead.
CRUD baseline is 44 hours per entity (16 BE + 20 FE + 8 QC). More for entities with complex relationships.
Auth module is always at least 40 hours total to cover login, registration, password reset, session management, and middleware.
QC ratio is 30–40% of development effort.
DevOps ratio is 10–15% of total effort.
PM and BA overhead is 10–15% of total effort.
Integration buffer adds 20–30% when a module depends on another.
Risk buffer adds 15–20% to the final number.
Design effort is 4–8 hours per UI screen if the project needs design work.

These rules are documented in .claude/rules/wbs-format.md inside your project repository and can be adjusted per organization.

AI-assisted discount

Ptah tracks which tasks can be accelerated with AI assistance and applies a discount to the final hours. Typical reductions:

| Work type | AI effort reduction | |---|---| | Boilerplate (CRUD, API routes, schemas) | 30–40% | | Unit tests | 40–50% | | Documentation | 50–60% | | Code review | 20–30% | | Bug fixing | 10–20% | | Complex business logic | 0–10% | | UI / UX design | 0–10% |

The full estimate is always shown alongside the discounted estimate, so the PM can decide which number to quote.

Historical baselines

Ptah ships with baseline hours for common module types — authentication, CRUD entities, payment integration, file upload, notifications, dashboards, search, RBAC, email integration, real-time features. Your organization can override each baseline with historical data from past projects.

BA workflow — how the spec feeds into the WBS.
FAQ — common questions about the estimates.