This project adds clear, in-context reporting to Vitally's Playbooks by making run behavior and action outcomes visible directly in the editor. It introduces aggregate node-level metrics and detailed per-run states, alongside tables that show all objects created or updated by each action with filtering by action and run. Together, this gives teams a fast way to validate performance, spot errors, understand outcomes, and refine Playbooks without leaving the workflow or relying on support.

Every new customer faced the same paradox: Vitally was powerful but complex. Setting up a basic playbook required configuring triggers, stages, templates, dashboards, and automation. Most took 6-8 weeks to build their first workflow, while 850+ customers had already solved these problems with no way to share solutions.

Blueprints reframes setup as curation, not creation. Instead of empty Views and Playbooks, customers browse 35+ expert-recommended templates organized by use case and feature. Each shows a preview of what will be created before committing.

Applying a Blueprint triggers a configuration flow under 60 seconds. Everything is created in Draft mode with automation off, giving customers control to customize before activating. Customers can also publish their own Blueprints, creating a virtuous cycle.

Formula Traits introduce a flexible way for teams to calculate and store derived values directly on Vitally objects using math, logic, and aggregations. Users can build custom traits from existing properties and related data, preview and validate results, apply consistent formatting, and use those values everywhere traits are supported. This enables more powerful reporting, forecasting, health scoring, and analysis without external tools or manual workarounds.

The original onboarding asked developers to invest time before demonstrating value. Users faced a 4-step wizard requiring codebase connection, IDE settings, and permissions before seeing what the AI could do. Research showed 67% abandoned during setup.

The new flow inverts this. It starts with a 90-second interactive demo showcasing the AI refactoring a complex function into clean code. Users interact with the demo and see how it thinks. Only after this 'aha moment' do we suggest connecting their codebase.

Setup is optional and progressive. Users can skip it and explore with sample projects, connecting later when convinced. This 'value first, commitment later' approach respects that developers want to quickly assess if a tool is worth their time.

The original mobile check-in assumed patients would complete it at home the night before. Reality: 73% of attempts happened within 15 minutes of the appointment, in the parking lot, with patients stressed. The flow demanded account creation, 12 questions, insurance verification, and consent forms. Only 12% finished.

The redesign reframes check-in around one goal: get patients from car to exam room fast. The new flow asks three questions: 'Any changes since your last visit?' (defaulting to no), 'What brings you in today?', and 'Confirm insurance' (single tap). Everything else is pre-filled or moved to post-check-in.

The key innovation is async verification: insurance validation happens in the background after check-in, flagging issues to staff before the appointment. This removes friction from the patient's critical path.

Traders were drowning in undifferentiated signals. The new dashboard introduced confidence scoring, asset class filtering, and a personalized 'watchlist' view that surfaced only the signals relevant to each trader's strategy.

The tool consolidated 7 separate applications into one integrated workflow. Analysts could now pull live market data, insert auto-updating charts, apply brand-consistent templates, and publish directly to the client portal without leaving the application.