BrandPilot: connecting brand strategy and product planning

Early-stage teams often write a brand positioning document and, separately, a feature backlog. The two rarely reference each other: positioning claims are not reflected in the MVP, and roadmap trade-offs are made without the customer problems in view.

Strategy consultants and product managers bridge this gap manually through workshops and templates. The concept question was: what would a lightweight, self-serve tool for that bridging work look like?

BrandPilot asks a single set of discovery questions and derives every downstream artefact from them — brand statements, personas, problem-to-solution mapping, a feature backlog, a roadmap and a visual direction. Because everything shares one data model, a change to a customer problem flows through to the mapping, the journey and the exported document.

A deliberate decision was not to use generative AI. Outputs come from documented templates and lookup tables, so they are predictable, explainable and free to run. Each statement exposes the rule that produced it, and the product never claims to be “intelligent”.

• Short, validated steps. Discovery is split into six sections with inline validation, helpful examples of the expected phrasing and a review step that links back to each section.
• Nothing is lost. Answers autosave on every change; regenerating the strategy preserves statements the user edited and features they added.
• Drafts, clearly labelled. Generated text carries a “Rule-based draft” badge until edited; personas are marked illustrative; visual directions carry a trademark disclaimer.
• Keyboard parity. Drag-and-drop on the roadmap has arrow-button equivalents; the matrix has a full list view; every dialog is focus-trapped and dismissible with Escape.
• Calm visual language. Deep navy, off-white and a restrained teal accent, editorial serif headings and generous whitespace — closer to a premium B2B tool than an admin template. Light and dark themes are both designed.

Below is the actual matrix component from the studio, rendered with the fictional demo project's backlog. Hover or select points to inspect features.

Captured from the running application with the fictional “Orchard Ledger” demo project.

src/
├─ app/                 routes: /, /case-study, /studio/*
├─ components/
│  ├─ marketing/        landing & case-study sections
│  ├─ studio/           shell, wizard, strategy, product, features, roadmap, identity, export
│  └─ ui/               shadcn/ui primitives
├─ lib/
│  ├─ domain/           types, options, Zod schemas
│  ├─ generators/       brand, product, identity, next-steps (pure)
│  ├─ pdf/              jsPDF document builder
│  └─ prioritization.ts scoring, quadrants, scope & progress
└─ store/               external store, pure mutations, React hook

• Landing page with a live, interactive preview of the engine
• Six-step discovery wizard with validation, autosave and review
• Brand strategy with eight editable statements and revert-to-draft
• Vision canvas, personas, problem–solution map and user journey
• Custom features with impact/effort scoring and MoSCoW
• Prioritisation matrix and MVP capacity guardrails
• Phased roadmap with drag-and-drop, keyboard moves and status
• Visual directions with WCAG contrast checks and UI moodboard
• PDF strategy export and JSON backup / import
• Fictional demo project, workspace reset, light & dark themes

• Accounts and cloud sync so teams can collaborate on one project, with comments and version history.
• Optional, clearly labelled AI-assisted rewriting layered on top of the deterministic drafts — never replacing them.
• Customer-interview capture that links evidence to problems, personas and features.
• Exports to slide decks and to issue trackers so the roadmap becomes tickets.
• Usability testing with founders to validate the discovery questions and terminology.

No usage, revenue or client outcomes are reported here because none exist — this is a concept built to demonstrate approach and craft.