Skip to content

Big Tech Best Practices

The Big Tech profiles combine proven product development practices from 10 leading methodologies into a comprehensive "best of all worlds" framework. They merge the best of Amazon, Google, Stripe, Netflix, Spotify, Meta, Apple, Microsoft, Basecamp (Shape Up), and Teresa Torres (Continuous Discovery).

Profile Variants

Profile Starting Document Best For
big-tech-product MRD + OKRs New product lines, major initiatives
big-tech-feature OpportunitySpec + OKRs Features on existing products

Both variants share the same methodology, principles, and practices. They differ only in the starting document.

When to Use

Use the Big Tech profiles when you want:

  • A comprehensive, battle-tested methodology
  • The rigor of multiple complementary approaches
  • High standards without dogmatic adherence to one company's way
  • Flexibility to emphasize different practices based on context

Big Tech Essentials

For teams that want the rigor of big-tech methodologies without the full 10-company synthesis, Big Tech Essentials provides a streamlined alternative focusing on three foundational methodologies:

  • Amazon: Customer obsession, PR/FAQ, 6-pager narratives
  • Google: OKRs, design docs, explicit tradeoffs
  • Stripe: API-first design, developer experience

Essentials Profile Variants

Profile Starting Document Pass Threshold
big-tech-essentials-product MRD + OKRs 70%
big-tech-essentials-feature OpportunitySpec + OKRs 70%

Essentials Product Flow

Big Tech Essentials Product Flow

Essentials Feature Flow

Big Tech Essentials Feature Flow

When to Use Essentials vs Full Big Tech

Scenario Use Essentials Use Full Big Tech
Team new to structured specs Yes No
Startup / small team Yes No
Enterprise with established practices Consider Yes
Need comprehensive evaluation No Yes
Training teams on fundamentals Yes No
Want Netflix/Spotify/Meta practices No Yes
Focus on customer + API + technical Yes Consider

Using the Essentials Profiles

# Initialize with essentials product profile
multispec init my-product --profile big-tech-essentials-product

# Initialize with essentials feature profile
multispec init my-feature --profile big-tech-essentials-feature

The essentials profiles include the same specs as the full profiles but with simplified rubrics that focus on the three core methodologies.

Company Contributions

Amazon - Customer Obsession & Narratives

Practice Description
Working Backwards Write the press release first, then build
Customer Obsession Start with the customer and work backwards
PR/FAQ Press release + FAQ to challenge assumptions
6-Pager Narrative documents over slides
Ownership Think long-term, end-to-end responsibility
Bias for Action Two-way door decisions should be fast
Disagree and Commit Truth-seeking, then alignment

Google - Engineering Rigor & Data

Practice Description
Design Docs Technical specifications with explicit tradeoffs
RFCs Peer review ensures diverse perspectives
OKRs Objectives and measurable key results
10x Thinking Aim for 10x improvement, not 10%
Alternatives Considered Always evaluate multiple approaches
Explicit Tradeoffs Every decision has costs; make them visible
Non-Goals What we won't do is as important as what we will
Launch and Iterate Ship, measure, learn, improve

Stripe - Developer Experience & APIs

Practice Description
API-First Design the interface before the implementation
Developer Experience Optimize for developer productivity
Documentation Quality Docs are part of the product
Incremental Delivery Ship small, ship often, ship safely
Error Messages Matter Helpful errors are a feature
Consistency APIs should be predictable

Netflix - Autonomy & Alignment

Practice Description
Freedom & Responsibility Give talented people freedom, expect accountability
Context Not Control Provide context for decisions, not commands
Highly Aligned, Loosely Coupled Align on strategy, decouple execution
Radical Transparency Share information broadly, default to open

Spotify - Agile at Scale

Practice Description
Bets Not Projects Frame work as bets with explicit risk
Autonomous Squads Small teams with end-to-end ownership
Fail Fast, Learn Fast Quick experiments over long planning

Meta - Speed & Boldness

Practice Description
Move Fast Speed is a competitive advantage
Bold Bets Take big swings on transformative ideas
Hackathon Culture Rapid prototyping to validate ideas

Apple - Focus & Responsibility

Practice Description
DRI Directly Responsible Individual for every decision
Say No Focus means saying no to good ideas
Deep Integration Own the full stack when it matters

Microsoft - Growth & Inclusion

Practice Description
Growth Mindset Learn from failure, embrace challenges
Accessibility First Design for everyone from the start

Basecamp - Shape Up

Practice Description
Fixed Time, Variable Scope Appetite sets time; scope flexes to fit
Shaping Define problems at the right abstraction level
Betting Not Planning No backlogs; bet on shaped pitches
Circuit Breaker Stop if it's not working; don't drag out failures
Appetite Not Estimates How much time is this worth, not how long
Hill Charts Track uncertainty (uphill) and execution (downhill)
Rabbit Holes Identify and avoid scope traps before starting
Cool-Down Two-week break between cycles for recovery

Teresa Torres - Continuous Discovery

Practice Description
Weekly Touchpoints Talk to customers every week, not just during research
Story-Based Interviews Collect stories of past behavior, not opinions
Opportunity Solution Trees Visualize path from outcome to solutions
Assumption Testing Test riskiest assumptions before building
Compare and Contrast Test multiple solutions in parallel
Product Trio PM, Design, Engineering collaborate on discovery
Outcome-Driven Start with measurable outcome, not feature requests
Small Experiments Run small, fast experiments to reduce risk

When to Use Which Practice

The Big Tech profile includes practices for different situations. Choose based on context:

Use Shape Up Practices When

  • Fixed timeline is more important than fixed scope
  • Estimating is wasteful; appetite makes more sense
  • Teams need autonomy to figure things out
  • Project needs clear circuit breaker for failure
  • Work requires exploration with bounded time

Use Continuous Discovery Practices When

  • Problem space is uncertain and needs exploration
  • Building features without recent customer input
  • Multiple solutions could work; need to validate
  • Assumptions are risky; need to test before building
  • Research needs to be integrated with delivery (not separate phase)

Use Working Backwards When

  • New product or major initiative
  • Need to align stakeholders on vision
  • Customer benefit needs to be crystal clear
  • FAQ will expose hidden concerns

Use Design Docs/RFC When

  • Architectural decisions have significant tradeoffs
  • Multiple alternatives need to be evaluated
  • Reversibility is uncertain (one-way door)
  • Peer review will add valuable perspectives

The Big Tech Flows

Big Tech Product Flow (MRD Start)

Big Tech Product Flow

Big Tech Feature Flow (OpportunitySpec Start)

Big Tech Feature Flow

Using the Big Tech Profiles

Initialize a Product Project

multispec init my-product --profile big-tech-product

Initialize a Feature Project

multispec init my-feature --profile big-tech-feature

Create MRD with OKRs (big-tech-product)

multispec draft mrd -p my-product

The MRD template includes:

  • Customer segment and pain points (Amazon)
  • Objectives and Key Results (Google)
  • Market opportunity (standard)

Create OpportunitySpec with OKRs (big-tech-feature)

multispec draft opportunity-spec -p my-feature

The OpportunitySpec template includes:

  • 12-box canvas (Patton + Cagan)
  • OKR alignment (Google)
  • Risk and assumptions validation

Working Backwards Phase

multispec synthesize press -p my-product
multispec synthesize faq -p my-product

The FAQ includes:

  • Customer FAQ (Amazon)
  • Stakeholder FAQ (Amazon)
  • Developer/API FAQ (Stripe)

Create PRD with API Contracts

multispec synthesize prd -p my-product

The PRD template includes:

  • User stories and requirements
  • API contracts (Stripe)
  • Non-goals (Google)
  • Alternatives considered (Google)
  • Success metrics (Google OKRs)

Create UXD with DX

multispec draft uxd -p my-product

The UXD template includes:

  • User journeys
  • Developer experience (Stripe)
  • Accessibility considerations (Microsoft)

Synthesize Design Doc

multispec synthesize trd -p my-product

The TRD/Design Doc includes:

  • Context and scope
  • Goals and non-goals
  • Alternatives analysis
  • Explicit tradeoffs
  • Reversibility assessment (two-way vs one-way door)
  • Scalability considerations (10x)

Unified Principles

The Big Tech profile distills 30+ practices into these core principles:

Customer & User Focus

  1. Customer Obsession - Start with the customer (Amazon)
  2. Developer Experience - Optimize for developers (Stripe)
  3. Accessibility First - Design for everyone (Microsoft)

Vision & Ambition

  1. 10x Thinking - Aim for transformative improvement (Google)
  2. Working Backwards - Start with the end state (Amazon)
  3. Bets Not Projects - Frame with explicit risk (Spotify)

Decision Making

  1. Explicit Tradeoffs - Make costs visible (Google)
  2. Alternatives Considered - Evaluate options (Google)
  3. Data-Driven - Back decisions with data (Google)
  4. Bias for Action - Move fast on reversible decisions (Amazon)
  5. DRI - Clear ownership (Apple)

Quality & Standards

  1. Highest Standards - Never settle (Amazon)
  2. API-First - Design interface first (Stripe)
  3. Documentation Quality - Docs are product (Stripe)
  4. Non-Goals - Be explicit about scope (Google)
  5. Say No - Focus through exclusion (Apple)

Collaboration & Review

  1. Peer Review - Diverse perspectives (Google)
  2. Disagree and Commit - Debate, then align (Amazon)
  3. Context Not Control - Empower, don't command (Netflix)
  4. Radical Transparency - Default to open (Netflix)

Execution & Delivery

  1. Ownership - End-to-end responsibility (Amazon)
  2. Incremental Delivery - Ship small and often (Stripe)
  3. Launch and Iterate - Ship, measure, improve (Google)
  4. Move Fast - Speed is advantage (Meta)
  5. Fail Fast, Learn Fast - Quick experiments (Spotify)

Team & Culture

  1. Freedom & Responsibility - Trust with accountability (Netflix)
  2. Highly Aligned, Loosely Coupled - Strategy + autonomy (Netflix)
  3. Autonomous Squads - Small empowered teams (Spotify)
  4. Growth Mindset - Learn from failure (Microsoft)
  5. Frugality - Constraints breed resourcefulness (Amazon)

Scoping & Planning (Shape Up)

  1. Fixed Time, Variable Scope - Appetite sets time; scope flexes (Basecamp)
  2. Appetite Not Estimates - How much is it worth, not how long (Basecamp)
  3. Shaping - Define at right abstraction level (Basecamp)
  4. Betting Not Planning - No backlogs; bet on pitches (Basecamp)
  5. Circuit Breaker - Stop failures early, don't drag out (Basecamp)
  6. Rabbit Holes - Identify scope traps upfront (Basecamp)

Discovery & Validation (Continuous Discovery)

  1. Weekly Touchpoints - Talk to customers every week (Torres)
  2. Story-Based Interviews - Collect stories, not opinions (Torres)
  3. Opportunity Solution Trees - Visualize outcome to solution path (Torres)
  4. Assumption Testing - Test riskiest assumptions first (Torres)
  5. Compare and Contrast - Test multiple solutions in parallel (Torres)
  6. Product Trio - PM, Design, Engineering collaborate (Torres)

Rubric Extensions

The Big Tech profile adds evaluation criteria from each methodology:

MRD Evaluation

Category Weight Source
OKR Quality 15% Google
Customer Clarity 15% Amazon
Standard MRD 70% Base

PRD Evaluation

Category Weight Source
API Contract 10% Stripe
Non-Goals 10% Google
Alternatives 10% Google
Standard PRD 70% Base

TRD Evaluation

Category Weight Source
Explicit Tradeoffs 15% Google
Reversibility 10% Google/Amazon
Scalability (10x) 10% Google
Standard TRD 65% Base

UXD Evaluation

Category Weight Source
Developer Experience 10% Stripe
Accessibility 10% Microsoft
Standard UXD 80% Base

Shape Up Pitch Evaluation

Category Weight Source
Problem Clarity 20% Basecamp
Appetite Reasoning 20% Basecamp
Solution Abstraction 25% Basecamp
Rabbit Holes 20% Basecamp
No-Gos 15% Basecamp

OST Evaluation

Category Weight Source
Outcome Measurable 25% Torres
Opportunities Researched 25% Torres
Solutions Multiple 25% Torres
Experiments Defined 25% Torres

Discovery Snapshot Evaluation

Category Weight Source
Weekly Cadence 30% Torres
Story Quality 35% Torres
Learnings Actionable 35% Torres

Assumption Map Evaluation

Category Weight Source
Type Coverage (DVFUE) 30% Torres
Prioritization 35% Torres
Tests Designed 35% Torres

Example Workflows

Big Tech Product Workflow

# 1. Initialize with big-tech-product profile
multispec init checkout-v2 --profile big-tech-product

# 2. Draft MRD with OKRs
multispec draft mrd -p checkout-v2
# Edit to add customer pain points and OKRs
multispec eval mrd -p checkout-v2
multispec approve mrd -p checkout-v2

# 3. Working Backwards (Amazon)
multispec synthesize press -p checkout-v2
multispec eval press -p checkout-v2
multispec approve press -p checkout-v2

multispec synthesize faq -p checkout-v2
# Add developer FAQ section (Stripe)
multispec eval faq -p checkout-v2
multispec approve faq -p checkout-v2

# 4. 6-Pager narrative (Amazon)
multispec synthesize narrative-6p -p checkout-v2
multispec eval narrative-6p -p checkout-v2
multispec approve narrative-6p -p checkout-v2

# 5. PRD with API contracts (Stripe + Google)
multispec synthesize prd -p checkout-v2
# Add API contract, non-goals, alternatives
multispec eval prd -p checkout-v2
multispec approve prd -p checkout-v2

# 6. UXD with DX (Stripe + Microsoft)
multispec draft uxd -p checkout-v2
# Include developer experience and accessibility
multispec eval uxd -p checkout-v2
multispec approve uxd -p checkout-v2

# 7. Design Doc (Google)
multispec synthesize trd -p checkout-v2
# Ensure tradeoffs, alternatives, scalability
multispec eval trd -p checkout-v2
multispec approve trd -p checkout-v2

# 8. Test Plan with Experiments
multispec synthesize tpd -p checkout-v2
multispec eval tpd -p checkout-v2
multispec approve tpd -p checkout-v2

# 9. Final reconciliation
multispec reconcile -p checkout-v2

# 10. Check status
multispec status -p checkout-v2

Big Tech Feature Workflow

# 1. Initialize with big-tech-feature profile
multispec init payment-retry --profile big-tech-feature

# 2. Draft OpportunitySpec with OKRs
multispec draft opportunity-spec -p payment-retry
# Fill in 12-box canvas with OKR alignment
multispec eval opportunity-spec -p payment-retry
multispec approve opportunity-spec -p payment-retry

# 3. Working Backwards (Amazon)
multispec synthesize press -p payment-retry
multispec eval press -p payment-retry
multispec approve press -p payment-retry

multispec synthesize faq -p payment-retry
multispec eval faq -p payment-retry
multispec approve faq -p payment-retry

# 4. PRD with API contracts (Stripe + Google)
multispec synthesize prd -p payment-retry
multispec eval prd -p payment-retry
multispec approve prd -p payment-retry

# 5. UXD with DX (Stripe + Microsoft)
multispec draft uxd -p payment-retry
multispec eval uxd -p payment-retry
multispec approve uxd -p payment-retry

# 6. Design Doc (Google)
multispec synthesize trd -p payment-retry
multispec eval trd -p payment-retry
multispec approve trd -p payment-retry

# 7. Test Plan with Experiments
multispec synthesize tpd -p payment-retry
multispec eval tpd -p payment-retry
multispec approve tpd -p payment-retry

# 8. Final reconciliation
multispec reconcile -p payment-retry

# 9. Check status
multispec status -p payment-retry

When to Use Other Profiles

Scenario Consider Instead
Pure API/platform product Stripe profile (more focused)
Greenfield consumer app AWS profile (narrative focus)
Engineering-heavy infrastructure Google profile (design doc focus)
Early-stage validation Lean Startup profile (hypothesis focus)
Feature on existing product AWS Feature profile (OpportunitySpec)
Fixed-time projects only Shape Up profile (dedicated)
Research-heavy, continuous Continuous Discovery profile (dedicated)

Optional Shape Up and Continuous Discovery Artifacts

The Big Tech profile includes optional artifacts from Shape Up and Continuous Discovery that can be used when appropriate:

Shape Up Artifacts

Artifact When to Use
shapeup-pitch Major features with fixed time budget
shapeup-scope Track progress during fixed-time execution 
# Optional: Create a shaped pitch for a major feature
multispec draft shapeup-pitch -p my-feature

# Optional: Track scopes during execution
multispec draft shapeup-scope -p my-feature

Continuous Discovery Artifacts

Artifact When to Use
ost Map opportunities to solutions with experiments
discovery-snapshot Track weekly discovery activities
assumption-map Prioritize assumptions for testing 
# Optional: Create Opportunity Solution Tree
multispec draft ost -p my-feature

# Optional: Track weekly discovery
multispec draft discovery-snapshot -p my-feature

# Optional: Map and prioritize assumptions
multispec draft assumption-map -p my-feature

References

Amazon

Google

Stripe

Netflix

Spotify

Meta

Apple

Microsoft

Basecamp (Shape Up)

Teresa Torres (Continuous Discovery)