The Agency Handbook

A secure, branded workspace where clients access project updates, approve work, share files, and communicate, without needing access to your internal tools.

A structured process for submitting content for review, collecting feedback, and getting explicit sign-off before publishing or delivering, so approvals are tracked, not assumed.

Net Promoter Score is a client loyalty metric that shows how likely clients are to recommend your agency, using a single 0 to 10 survey question.

The structured process of bringing a new client into an agency engagement: gathering information, running the kick-off, setting communication norms, and managing the first deliverable.

The ongoing practice of maintaining trust, communication, and structured feedback with agency clients to retain relationships and prevent churn.

An onboarding questionnaire is the structured intake an agency uses to collect goals, stakeholders, assets, approvals, access, and working preferences from a new client before delivery starts.

Client feedback is the structured review input an agency collects, consolidates, and acts on during delivery so revision rounds stay clear, scoped, and traceable.

An AI-driven process where an AI agent autonomously plans and executes a series of steps to complete a complex task, without a human directing each action.

An AI technique where the model searches your own documents or data before generating a response, so answers are grounded in your specific information, not just the model's training.

An AI system that can perceive its environment, make decisions, and take actions autonomously to achieve a goal. Unlike a chatbot that just responds, an agent acts.

An AI system design where a human reviews, validates, or approves AI outputs at key decision points, rather than letting the AI act fully autonomously.

Model Context Protocol, or MCP, is a standard way for AI tools to connect to external systems, data, and actions, so one model can work across your real stack without custom one-off integrations.

Prompt engineering is the practice of structuring instructions, context, constraints, and examples so an AI system produces a useful output, not just a plausible one.

An MCP server is the software layer that connects an AI model to external tools and data sources using the Model Context Protocol standard, acting as the bridge between what an AI can reason about and what it can actually do.

Agentic AI refers to AI systems that can plan and execute multi-step tasks autonomously: given a goal, they figure out the steps, use tools, check their own work, and keep going until the job is done.

An autonomous AI agent is an AI system that can receive a goal, break it into steps, use tools to execute those steps, and evaluate its own progress, all without step-by-step human direction.

An LLM agent is an AI system that uses a large language model as its reasoning engine: not just to answer questions, but to plan what steps to take, decide which tools to use, and work toward a goal with minimal human direction.

AI workflow automation means using AI to run multi-step business workflows automatically. The AI does not just execute predefined steps. It reads content, makes routing decisions, drafts outputs, and handles the variation that rule-based automation cannot.

AI project management tools use machine learning and automation to handle scheduling, status tracking, task creation, and reporting, freeing agencies from the coordination overhead that scales with client volume.

AI in project management means using artificial intelligence to handle the operational and administrative tasks in a client project lifecycle, freeing project managers to focus on client relationships, creative judgment, and the decisions that require context.

The structured process of running a client walkthrough, collecting formal sign-off, and handing off a completed website — covering pre-launch checks, the delivery call, and post-launch documentation.

A written agreement that defines exactly what an agency will deliver, what is excluded, and the conditions for sign-off.

A formal document that defines the work an agency will deliver, the timeline, the cost, and the acceptance criteria, signed before the project begins.

A written amendment to the original project scope that documents new work, additional cost, and timeline impact, issued and signed before the new work begins.

A discovery call is the first structured sales conversation where an agency diagnoses the prospect's problem, qualifies fit, and decides whether a proposal or scope should follow.

A document that captures the objective, audience, tone, deliverables, and constraints for a creative project, agreed by agency and client before production starts.

A contract for ongoing monthly services between an agency and a client, specifying the scope, hours or deliverables, payment terms, and exit conditions.

The four main structures agencies use to charge for their services: hourly billing, fixed-fee project pricing, monthly retainers, and value-based pricing.

Process automation means using software to execute repeating, predictable tasks automatically, so your team can focus on work that requires judgment, relationships, and creativity.

AI-native describes a business, agency, or product designed from the ground up to use AI as a core operating component. It is not a layer added on top of existing processes; AI is built into the workflows themselves.

A service level agreement sets the response times, revision windows, approval deadlines, and support coverage an agency commits to during an active client engagement.

A brand style guide is the operating document that defines how a client brand should look, sound, and stay consistent across creative work, reviews, handoff, and future revisions.

The process of assigning team members and their available hours to client projects based on capacity, priority, and skill, a critical practice for agencies running multiple concurrent clients.

A project charter is a short kickoff document that aligns the agency and client on goals, stakeholders, success metrics, constraints, and decision-making before execution begins.

Capacity planning is the process of forecasting how much client work your agency can realistically deliver over the next few weeks or months before assigning it.

A kickoff meeting is the structured start of a client engagement where the agency and client align on scope, stakeholders, approvals, assets, timelines, and post-meeting next steps.

Automated project management uses software triggers and rules to handle routine client project tasks automatically, so account managers spend time on relationship work instead of admin.