This demo shows three real organisations using Seed with different domain packs and org customisation layers. The data is synthesised from actual tool stacks used by each company.
Dyson uses the Hardware pack with an exponential cost model. Data sources: PTC Windchill (DW-* parts), Altium (PCB metadata), Octopart (component supply chain), Confluence (engineering specs), internal test systems.
GDS (UK Gov) uses the Service Design pack with a custom public sector org layer. Data sources: Jira (GOV-* tickets), Confluence, Miro (journey maps), GOV.UK Design System, accessibility audit reports.
Click any node to see its provenance — which tool it came from, the exact text that generated it, and the confidence score. Use "Advance stage" to move the program forward and watch tensions escalate as irreversibility increases.
Tension detected
seed
Process intelligence platform
Sign in to Seed
Your data stays in your workspace. Sign in to save connectors, process graphs, and workspace settings.
Email
Password
Email
Password
or
Step 1 of 3 — Workspace
Set up your workspace
This helps Seed calibrate the right process vocabulary, level taxonomy, and tension types for your context.
Workspace name
Your role
Team size
Step 2 of 3 — Domain pack
What are you building?
Pick the domain pack that best matches your work. This sets the process levels and cost model.
⬡
Software product
SaaS · API · mobile Platform → Feature
◈
Hardware product
Electronics · medical Program → Component
◎
Service design
Customer journeys Policy → Channel
◌
R&D / Research
Hypothesis driven Programme → Protocol
Step 3 of 3 — Project brief
What are you working on?
Optional: paste a brief or PRD. The engine will extract your first process graph live.
Connector registry
Credentials stored locally only · never sent anywhere except the tool's own API
Connect
Organisation canvas
Cross-team process view · intent coherence by squad