Scoring frameworks

The three frameworks that power Dacard.ai scoring

Scoring frameworks

Dacard.ai scores products across three frameworks. Each measures a different surface (team, process, product) and together they form one read.

Three frameworks, one read

Team Operations

How capable is your team at building intelligent products? 27 dimensions across 6 functions. Composite 27-135.

Development Lifecycle

How mature is your build process? 6 stages, 36 tasks, from spec to compound learning.

Product Assessment

How deep is intelligence in your product? 27 dimensions across UX, architecture, data flywheel, and business model.

---

Team Operations

Team Operations Framework

The flagship framework. Evaluates a product team’s operational capability across 27 dimensions organized into 6 functions, each scored 1 to 5, for a composite score of 27 to 135.

Five maturity stages

Stage	Score	What it means
Foundation	27-48	Basic or absent capabilities. Most work is manual and ad hoc.
Building	49-70	Emerging practices, inconsistently applied across functions.
Scaling	71-91	Systematic processes in place with measurable outcomes.
Leading	92-113	Practices deeply embedded in the team’s operating model.
Compounding	114-135	Self-improving systems. Every cycle makes the team better.

Six functions, 27 dimensions

Strategy (4 dimensions)

Market Intelligence, Decision Quality, Roadmap Discipline, Competitive Positioning

Design (4 dimensions)

Research and Discovery, Prototyping Speed, Experience Design, Design-Dev Handoff

Development (4 dimensions)

Architecture and Systems, Spec and Context Quality, Build vs Buy, Delivery Velocity

Operations (4 dimensions)

Customer Signal Synthesis, Product Analytics, Data Strategy and Flywheel, Feedback Loop Quality

GTM (4 dimensions)

Positioning and Messaging, Launch Execution, Adoption and Expansion, Pricing and Packaging

Intelligence (4 dimensions)

Quality and Experimentation, Team Orchestration, Process Iteration, Cost and Token Economics

Deep dive: Team Operations

Full 27-dimension definitions, scoring criteria, and stage progression guide.

Development Lifecycle

Development Lifecycle Framework

Maps the six stages of building intelligent products with 36 tasks across stages. Development Lifecycle uses a self-assessment model rather than automated URL scoring.

Six lifecycle stages

1. Specify

   Define the problem with clear constraints. Establish what the model can and cannot do before building.

2. Context

   Assemble the knowledge layer the model needs. RAG pipelines, embeddings, grounding data.

3. Orchestrate

   Wire up models, agents, and pipelines. Connect inference to product surfaces.

4. Validate

   Test quality at machine speed. Automated evals, human review, and prompt tuning.

5. Ship

   Deploy and manage inference costs. Token budgets, latency targets, and model selection.

6. Compound

   Close the feedback loop. Usage data flows back into model and product improvement.

Cross-cutting concerns

Three forces span all stages and must be managed continuously:

Token Economics

The cost per inference. Every stage has a token budget impact.

Role Fluidity

The shift from specialist to generalist as model capability collapses role boundaries.

Cognitive Debt

Accumulated prompt complexity, model coupling, and evaluation gaps.

Deep dive: Development Lifecycle

Full stage definitions, 36 tasks, and operations stack categories.

Product Assessment

Product Assessment Framework

Evaluates how deeply intelligence is embedded in the product itself across 27 dimensions. Where Team Operations measures the team, Product Assessment measures the product: architecture, UX, data flywheel, and business model.

Key assessment areas

Area	What it evaluates
Integration Depth	How central machine intelligence is to the product’s core value
Personalization	Adaptive experiences that learn from user behavior
Automation	Intelligent workflows that reduce manual effort
Data Flywheel	Whether usage data improves the product over time
Interaction Patterns	Interface patterns designed for model-led interactions
Cost Architecture	Token economics and inference cost management

Product Assessment scoring

Product Assessment uses the same 1-5 dimension scoring as Team Operations. It runs as a separate assessment via the /api/score/product endpoint or through the platform UI on any scored product’s Product Assessment tab.

Product Assessment API reference

Score a product using the Product Assessment framework.

---

How scoring works

When you submit a URL to the scoring API (POST /api/score), Dacard.ai:

1. Crawl

   The platform visits the URL and reads up to 6 pages, extracting signals from the product’s public-facing presence, documentation, and UX patterns.

2. Evaluate

   Each of the 27 Team Operations dimensions is assessed on a 1-5 scale based on observable signals.

3. Generate

   A composite score (27-135), maturity stage, dimension-level insights, and coaching moves are produced.

4. Store

   The result is saved with a shareable link at /r/{id}.

Scores can also be generated anonymously without authentication, subject to rate limits. See Rate Limits.

Reading the three frameworks together

When all three frameworks score well, the team, the build process, and the product reinforce each other. Common mismatches:

Scenario	Interpretation
High Team, low Lifecycle	Strong team, weak process. Talented team shipping inconsistently.
High Product, low Team	Capable product, weak ops. Impressive surface, unsustainable operations.
High across all three	Aligned. The target state.

Learn more

Score a URL

API reference for the Team Operations scoring endpoint.

Get a result

Retrieve stored scoring results.

DAC chat

Ask DAC for moves based on your scores.