Skip to content

Module 04 · Lesson 13

Set up the client workspace and source of truth

Make ownership, lifecycle, and authoritative records clear before connecting an agent.

Updated August 2026 · AI Agency Academy

What you will learn

Make a clear, safer operating decision.

You will be able to prepare a client workspace so an agent receives reliable context and people can see who owns the next step.

Why this matters

Good agent work is useful before it is impressive.

An agent cannot repair a system where the same customer exists in three places, stages mean different things to different people, and nobody owns corrections. A source-of-truth map is basic operating hygiene.

Field note 13

Make the relationship visible.

Core concepts

The language that keeps the work clear.

Authoritative recordThe system or field that wins when two copies of a fact disagree.
Saved on this device.
Lifecycle stageA shared description of where a customer or request is in the operating process.
OwnerThe individual or queue responsible for moving the record forward.
Data correctionThe controlled way an operator fixes a wrong or stale fact and prevents it from spreading.
The practical method

Work through the decision in order.

  1. 01

    Inventory customer records

    List where leads, customers, appointments, notes, preferences, and status live today.

  2. 02

    Declare the source

    Choose the authoritative system and field for each important fact; do not make the agent decide.

  3. 03

    Normalize lifecycle and ownership

    Define the few stages and owners that guide a real next action.

  4. 04

    Test one record journey

    Create or inspect one sample request from entry to handoff and confirm the right context follows it.

Worked example

A realistic, bounded implementation.

A recruiting firm keeps candidate notes in email, a spreadsheet, and the CRM. The proposed agent would otherwise see inconsistent role preferences and owner names.

The firm makes the CRM contact and opportunity records authoritative. It defines the stages ‘new enquiry,’ ‘needs qualification,’ ‘scheduled,’ and ‘closed,’ each with a responsible queue.

The agent can now summarize an inbound request into a consistent record and create a task for the right recruiter. It never treats an old spreadsheet as current policy or truth.

Build it in practice

Use this copyable working template.

Adapt it to the client’s evidence, policy, people, and tools. Do not treat placeholders as approved instructions.

Fact: [field]. Authoritative system: [system]. Record owner: [role]. Lifecycle stage: [stage]. Who may correct it: [role]. Agent may: [read / write / propose].
Spacebrain implementation

Put the operating system around the agent.

Configure contacts, pipelines, tags, stages, ownership, notes, tasks, and permissions before building automation around them. Use the CRM timeline to preserve the decision trail.

Practice

Before you move on

  • Map the source of truth for five facts the proposed agent needs.
  • Find one conflict between systems and decide how it is resolved.
  • Confirm every lifecycle stage has a human or queue owner.
  • Important facts have an authoritative source.
  • Stages mean the same thing to operators.
  • Ownership is not implied.
  • The agent’s write permission is limited and reviewable.

Put the learning to work

Build an AI service people can trust.

Create a free Spacebrain account and use the operating layer around your AI service.

Start free trial