AGENCYSCRIPT
CoursesEnterpriseBlog
๐Ÿ‘‘FoundersSign inJoin Waitlist
AGENCYSCRIPT

Governed Certification Framework

The operating system for AI-enabled agency building. Certify judgment under constraint. Standards over scale. Governance over shortcuts.

Stay informed

Governance updates, certification insights, and industry standards.

Products

  • Platform
  • Certification
  • Launch Program
  • Vault
  • The Book

Certification

  • Foundation (AS-F)
  • Operator (AS-O)
  • Architect (AS-A)
  • Principal (AS-P)

Resources

  • Blog
  • Verify Credential
  • Enterprise
  • Partners
  • Pricing

Company

  • About
  • Contact
  • Careers
  • Press
ยฉ 2026 Agency Script, Inc.ยท
Privacy PolicyTerms of ServiceCertification AgreementSecurity

Standards over scale. Judgment over volume. Governance over shortcuts.

On This Page

What an Agency Playbook Is (And Is Not)Playbook ArchitectureSection 1: Who We AreSection 2: How We Win BusinessSection 3: How We DeliverSection 4: How We Manage Client RelationshipsSection 5: How We Manage Our PeopleSection 6: How We Run the BusinessWriting the Playbook: Practical ProcessStep 1: Inventory What Already ExistsStep 2: Identify the Critical GapsStep 3: Assign AuthorsStep 4: Write for the ReaderStep 5: Review and IterateStep 6: Publish and TrainMaintaining the PlaybookOwnershipUpdate TriggersQuarterly ReviewVersion HistoryCommon Playbook MistakesToo LongToo AspirationalNo EnforcementOne-Size-Fits-AllSiloed CreationYour Next Step
Home/Blog/Three Directors, Three Methods, One Confused Client Base
Operations

Three Directors, Three Methods, One Confused Client Base

A

Agency Script Editorial

Editorial Team

ยทMarch 20, 2026ยท13 min read
agency playbookoperations manualprocess documentationscalability

A 45-person AI agency in Denver had three delivery directors, each running their practice area differently. Director A required formal architecture reviews before any technical decision. Director B operated on trust, letting senior engineers make decisions independently. Director C used a committee model where technical decisions went through a panel. Clients receiving different experiences depending on which practice area their project landed in โ€” one got detailed documentation and formal governance, another got fast delivery with minimal paperwork, the third got thorough but slow decision-making. When a senior PM rotated from Director A's team to Director B's team, she spent three weeks confused about which processes applied. When a client's engagement expanded across two practice areas, the inconsistency became visible and embarrassing. The agency was, in effect, three separate agencies wearing the same brand.

The fix was not choosing one director's approach and forcing it on the others. The fix was a playbook โ€” a comprehensive document that defined the agency's standard operating model while allowing practice-level adaptation. Within six months of implementing the playbook, client experience became consistent, new hire ramp-up time dropped by 40%, and the agency could confidently scale because the operating model was documented rather than dependent on individual leaders.

What an Agency Playbook Is (And Is Not)

An agency playbook is a comprehensive operational guide that documents how your agency works. It covers everything from how you sell and onboard clients to how you deliver projects, manage people, and run the business. It is the single reference that any team member can consult to understand how things work.

A playbook is not:

  • A policy manual (rigid rules that feel bureaucratic)
  • A training document (step-by-step tutorials for specific tasks)
  • A strategy document (vision and goals)
  • A wiki (dynamic, frequently changing reference material)

A playbook is:

  • The operating philosophy and standards of your agency
  • The "how we do things here" guide
  • The bridge between strategy (what we want to achieve) and execution (how we actually do it)
  • A living document that evolves as your agency matures

Think of it as the difference between a recipe book and a cooking philosophy. A recipe tells you the exact steps to make a specific dish. A cooking philosophy teaches you how to approach any dish โ€” knife skills, flavor principles, heat management, plating. Your playbook teaches your team how to approach any situation using your agency's principles and frameworks.

Playbook Architecture

Section 1: Who We Are

This is the foundation โ€” your agency's identity, values, and operating philosophy.

Company Overview:

  • What we do (services, specializations, approach)
  • Who we serve (target clients, industries, company sizes)
  • How we are different (competitive positioning, unique strengths)
  • Where we are going (vision, growth direction)

Values and Principles:

  • Core values (3-5 values that genuinely guide decisions)
  • Operating principles (how values translate into day-to-day behavior)
  • Quality standards (what "excellent" means at your agency)

Why this matters for the playbook: Values and principles are the decision-making shortcuts your team uses when the playbook does not cover a specific situation. An engineer facing a trade-off between speed and thoroughness should be able to reference your principles and make the right call without asking a manager.

Section 2: How We Win Business

Document your sales and business development process.

Ideal Client Profile:

  • Company characteristics (size, industry, technical maturity)
  • Project characteristics (budget range, timeline, complexity)
  • Behavioral characteristics (how ideal clients engage, what they value)
  • Red flags (signals that a prospect will be a problematic client)

Sales Process:

  • Lead qualification criteria
  • Sales stages and expected activities at each stage
  • Proposal structure and content guidelines
  • Pricing methodology (how to estimate, what margin targets to hit)
  • Deal approval process (who approves deals at different sizes)
  • Handoff process from sales to delivery

Pricing Guidelines:

  • Standard rate ranges by role and seniority
  • Project types and typical pricing structures (fixed-price, T&M, retainer)
  • Discount approval authority
  • Minimum project size and engagement terms

Section 3: How We Deliver

This is the largest and most referenced section. It covers everything about project delivery.

Project Lifecycle:

Phase 1: Kickoff and Onboarding

  • Client onboarding checklist
  • Internal kickoff meeting template
  • Stakeholder mapping process
  • Data access and infrastructure setup procedures
  • Communication plan template

Phase 2: Discovery

  • Data assessment framework
  • Technical feasibility evaluation
  • Requirements clarification process
  • Discovery deliverables and templates
  • Discovery-to-development transition criteria

Phase 3: Development

  • Sprint cadence and ceremonies
  • Coding standards and conventions
  • Code review process and expectations
  • Testing requirements (unit, integration, performance)
  • Documentation requirements
  • Model development lifecycle (training, validation, testing)
  • Architecture decision process
  • Client approval and feedback loops

Phase 4: Deployment

  • Deployment checklist
  • Staging and production environment standards
  • Rollback procedures
  • Monitoring and alerting setup
  • Performance validation criteria
  • Client acceptance testing process

Phase 5: Handoff and Closure

  • Knowledge transfer to client team
  • Documentation deliverables
  • Project retrospective process
  • Client satisfaction survey
  • Contract renewal or completion process

Quality Standards:

  • Definition of done for different deliverable types
  • Peer review requirements
  • Testing coverage expectations
  • Performance benchmarks
  • Security standards

Project Management Standards:

  • Status reporting format and frequency
  • Risk management process
  • Scope change management
  • Escalation procedures
  • Resource request process

Section 4: How We Manage Client Relationships

Communication Standards:

  • Response time expectations (email within 4 hours, Slack within 1 hour during business hours)
  • Status update format and frequency
  • Escalation path for issues and complaints
  • Executive sponsor engagement cadence

Client Health Monitoring:

  • How to assess client satisfaction
  • Warning signs of relationship deterioration
  • Intervention playbook for at-risk accounts
  • Regular business review format

Account Growth:

  • How to identify expansion opportunities
  • Cross-sell and upsell approach
  • Contract renewal timeline and process
  • Account planning for strategic clients

Section 5: How We Manage Our People

Hiring:

  • Hiring process (sourcing, screening, interviewing, offering)
  • Interview rubric and evaluation criteria
  • Compensation philosophy and bands
  • Offer approval process

Onboarding:

  • First-day checklist
  • First-week schedule
  • 30/60/90 day milestones
  • Mentor/buddy assignment process

Performance Management:

  • Review cadence and format
  • Feedback framework (how to give and receive feedback)
  • Career tracks and promotion criteria
  • Performance improvement process

Development:

  • Training and conference budget
  • Learning time allocation
  • Certification support
  • Internal knowledge sharing expectations

Offboarding:

  • Notice period expectations
  • Knowledge transfer process
  • Client transition procedures
  • Exit interview process

Section 6: How We Run the Business

Financial Management:

  • Invoicing process and timing
  • Expense policy and approval process
  • Budget review cadence
  • Financial reporting schedule

Meeting Cadence:

  • Required recurring meetings by role
  • Meeting templates and agendas
  • Decision-making framework

Tools and Systems:

  • Tool stack overview with purpose of each tool
  • Access request process
  • Data and security policies
  • Acceptable use guidelines

Compliance and Legal:

  • Data handling requirements
  • Client confidentiality standards
  • IP ownership defaults
  • Regulatory obligations by industry

Writing the Playbook: Practical Process

Step 1: Inventory What Already Exists

Before writing anything new, gather everything your agency has already documented โ€” process docs, templates, checklists, policy documents, training materials. Spread it out (physically or digitally) and organize it into the playbook sections. You will likely find that 30-50% of your playbook content already exists in some form.

Step 2: Identify the Critical Gaps

Map your inventory against the playbook architecture. Where are the gaps? Prioritize filling gaps in the areas that cause the most pain:

  • Areas where inconsistency creates client experience problems
  • Areas where new hires take the longest to ramp up
  • Areas where mistakes happen most frequently
  • Areas most dependent on specific individuals

Step 3: Assign Authors

Each section should be written by the person who knows it best โ€” not necessarily the most senior person, but the person who actually does the work or manages the process. The sales section should be written by the sales lead, the delivery section by the delivery lead, the people section by the HR/operations person.

Set deadlines: Give each author 2-3 weeks to draft their section. Without deadlines, playbook creation takes forever.

Step 4: Write for the Reader

The playbook's audience is a competent professional who is new to your agency. They understand their craft but do not know how your agency specifically does things. Write at that level:

  • Explain the "why" before the "how." People follow processes more consistently when they understand the reasoning
  • Use concrete examples. "Our standard project estimate includes a 20% buffer for scope uncertainty" is more useful than "include appropriate contingency"
  • Be specific about expectations. "Respond to client emails within 4 business hours" is better than "respond promptly"
  • Include templates and examples. A sample status report is worth more than a paragraph describing what a status report should contain
  • Keep it scannable. Bullet points, headers, bold text, and short paragraphs. Nobody reads a wall of text

Step 5: Review and Iterate

Once drafts are complete, review as a leadership team:

  • Does each section accurately reflect how we actually work (not how we aspire to work)?
  • Are there contradictions between sections?
  • Is the language consistent in tone and terminology?
  • Would a new hire understand this without additional explanation?

Incorporate feedback and produce a version 1.0. It does not need to be perfect โ€” it needs to be useful. Perfecting it forever means nobody benefits from it.

Step 6: Publish and Train

Publish the playbook in your wiki or knowledge management system. Then train the team:

  • Announce the playbook at an all-hands meeting. Explain what it is, why it exists, and how to use it
  • Walk through key sections in team meetings over the following weeks
  • Include playbook review in your new hire onboarding process
  • Reference the playbook in real situations ("per our playbook, the escalation process is...")

Maintaining the Playbook

A playbook that is never updated becomes fiction. Outdated guidance is worse than no guidance because people follow it and get wrong answers.

Ownership

Assign a playbook owner โ€” someone responsible for the overall document's health, currency, and quality. This is usually the COO or operations lead. Each section should also have a subject matter owner who reviews and updates it.

Update Triggers

Update the playbook when:

  • A process changes
  • A retrospective identifies a gap
  • A new team member points out something confusing or inaccurate
  • A new tool or system is adopted
  • Company strategy or structure changes

Quarterly Review

Every quarter, the playbook owner reviews the document section by section. For each section, check with the subject matter owner: Is this still accurate? Does anything need to change? Have any gaps been identified?

Version History

Maintain a change log at the beginning of the playbook. Note what changed, when, and why. This helps team members understand what is new and indicates that the playbook is a living document, not a static relic.

Common Playbook Mistakes

Too Long

A 200-page playbook will not be read. Aim for 40-60 pages for the core document, with links to detailed templates and supplementary materials. The playbook should be a guide that points people to detail, not a compendium of every possible instruction.

Too Aspirational

Writing the playbook as you wish things worked rather than how they actually work creates a credibility gap. Document your real processes first. Then improve them. Then update the playbook to reflect the improvements. A playbook that describes a fantasy world gets dismissed as irrelevant.

No Enforcement

If the playbook says "all projects require a kickoff meeting" but nobody enforces it, the playbook loses credibility. Leadership must model playbook adherence and address non-compliance. This does not mean rigid enforcement โ€” it means asking "why did we skip the kickoff meeting?" and either reinforcing the process or updating the playbook based on the answer.

One-Size-Fits-All

Not every process applies to every project. A $50,000 two-month project does not need the same governance as a $500,000 year-long engagement. The playbook should specify when different levels of process apply. Use tiers: "For projects under $100K, the abbreviated kickoff process applies. For projects over $100K, the full kickoff process applies."

Siloed Creation

If the playbook is written by operations alone, it reflects an operations perspective. Sales, delivery, engineering, and people management all need to contribute to and review the sections relevant to them. Siloed creation produces a document that feels disconnected from how actual work happens.

Your Next Step

You do not need to write a complete playbook this month. Start with the section that would reduce the most pain if it existed today. If new hire ramp-up is your biggest challenge, write the delivery section first โ€” how projects run, what the standards are, what tools to use. If client experience inconsistency is the problem, write the client management section. If sales-to-delivery handoffs keep failing, write the sales process and kickoff sections.

Pick one section. Assign an author. Set a two-week deadline. Publish it and share it with the team. Then pick the next section. Within three months, you will have a functional playbook that meaningfully improves consistency and reduces the ramp-up time for every new person who joins your agency. The playbook compounds in value โ€” every section added makes the agency slightly more self-sufficient and slightly less dependent on tribal knowledge.

Search Articles

Categories

OperationsSalesDeliveryGovernance

Popular Tags

prompt engineeringai fundamentalsai toolsthe difference between AIMLagency operationsagency growthenterprise sales

Share Article

A

Agency Script Editorial

Editorial Team

The Agency Script editorial team delivers operational insights on AI delivery, certification, and governance for modern agency operators.

Related Articles

Operations

Understaffed or Overstaffed? Both Camps Were Right.

You cannot manage what you cannot see. Here is how to build a team capacity dashboard that prevents burnout, eliminates bench time, and keeps projects staffed correctly.

A
Agency Script Editorial
March 21, 2026ยท12 min read
Operations

Optimizing Daily Standups for Distributed AI Agency Teams

Optimized standups keep distributed AI agency teams aligned without consuming the focused work time that engineers need to ship quality deliverables.

A
Agency Script Editorial
March 21, 2026ยท10 min read
Operations

Complete Utilization Rate Management Guide โ€” The Metric That Makes or Breaks Agency Profitability

A 5% shift in utilization can swing agency profit by 30% or more. Here is the definitive guide to measuring, managing, and optimizing the most important metric in your agency.

A
Agency Script Editorial
March 21, 2026ยท13 min read

Ready to certify your AI capability?

Join the professionals building governed, repeatable AI delivery systems.

Explore Certification