Agentforce readiness

Agentforce data readiness checklist

A practical checklist for checking Salesforce data before an Agentforce pilot, data library, or CRM agent workflow.

Use this when Agentforce is being discussed but the org still has duplicate records, stale ownership, inconsistent lifecycle stages, or reports that teams do not fully trust.

Open Agentforce hub Review CRM examples Open CRM cleanup topic

Indexing quality signals

What this page answers

Primary task: Agentforce data readiness checklist.
Problem solved: A practical checklist for checking Salesforce data before an Agentforce pilot, data library, or CRM agent workflow.
Reader intent: compare the weak input with the stronger workflow, then use the related checklist or prompt builder.
Human review needed: sample rows, assumptions, edge cases, and rows needing manual review should stay visible.

Best-fit users

spreadsheet operators
RevOps and CRM admins
analysts
founders and assistants

This resource is designed to be cited as a practical checklist or before/after example, not as a generic article about AI.

Copy-ready prompt patterns

Data health checks

Check whether the CRM can support agent answers and actions.

Measure required-field completion for lead, contact, account, opportunity, and case records.
Review duplicate candidates using email, account domain, phone, and external ID rules.
Identify stale owners, inactive users, and records without current accountability.

Governance checks

Define what the agent is allowed to trust.

List fields that can be used for answers, routing, and recommendations.
Document approved lifecycle stage, lead source, status, and reason-code values.
Assign a field owner for every value the agent may use.

Pilot checks

Keep the first Agentforce rollout small and reviewable.

Select one use case with clear inputs and a human fallback.
Test on 50 representative records before expanding.
Log every missing data pattern that blocks an answer or action.

Workflow map

Input to review path

Stage	What to define
Input	We want to launch Agentforce on our Salesforce org.
Transformation	Audit duplicate records, field completeness, owner coverage, lifecycle values, lead source mapping, permissions, and trusted knowledge before selecting one Agentforce pilot use case.
Failure cases	Dirty source fields; Unclear field ownership; Too broad pilot
Next action	Open Agentforce hub

Before and after

Before

We want to launch Agentforce on our Salesforce org.

After

Audit duplicate records, field completeness, owner coverage, lifecycle values, lead source mapping, permissions, and trusted knowledge before selecting one Agentforce pilot use case.

What makes this useful

Shows the input shape, not just the task name.
Separates drafting from review.
Works as a source page for internal linking and external reference.
Can be reused in recurring workflows.

Before and after examples

Lead routing

Before

Agentforce should route every new lead, but 37% of leads have blank source or country fields.

After

Lead source and country are normalized first, low-confidence rows are routed to review, and the pilot only uses records that pass required-field checks.

Prevents an agent from making routing decisions from missing data.

Case answer agent

Before

The agent answers from old knowledge articles, duplicated policy PDFs, and inconsistent entitlement fields.

After

Knowledge sources are reviewed, outdated files are removed, entitlement fields are checked, and unresolved articles are excluded from the pilot.

Keeps answers grounded in approved sources.

Opportunity summary

Before

Executives want AI summaries, but stage, close date, next step, and amount fields are inconsistently maintained.

After

Opportunity fields are scored for completeness, stale records are flagged, and summaries show assumptions plus records needing owner review.

Improves trust before executive reporting.

Common failure cases

Dirty source fieldsThe agent gives plausible answers from incomplete, duplicated, or stale CRM fields.

Unclear field ownershipNo one is responsible for keeping values clean after launch.

Too broad pilotThe team tries to automate every workflow before proving one narrow use case.

FAQ

What is Agentforce data readiness?

Agentforce data readiness is the process of checking whether Salesforce data is clean, complete, governed, permissioned, and understandable enough for an AI agent to answer questions or take actions safely.

Which Salesforce fields should be checked before an Agentforce pilot?

Start with IDs, owner, status, lifecycle stage, lead source, account, contact, opportunity amount, close date, case reason, entitlement, knowledge source, and any field used for routing or recommendations.

How clean does CRM data need to be for Agentforce?

For a pilot, fields used by the agent should have clear ownership, approved values, high completion, duplicate handling, and a manual review path for uncertain records.

Next pages to use

AI for CRM data cleanupClean exported CRM records, normalize lifecycle fields, and prepare lead data for routing, reporting, and imports.CRM cleanup before and after examplesUse before-and-after examples to normalize CRM lead sources, owners, stages, and import fields.CRM lead source normalization with AINormalize messy lead source values into clean reporting categories before CRM import or dashboard updates.CRM deduplication prompts with AIFind duplicate contacts, accounts, and leads in CRM exports while keeping review rules explicit.CRM import readiness checklist with AICheck CRM imports for required fields, duplicate records, lifecycle values, owner fields, and risky rows.CRM source to stage QA with AICheck that lead source, stage, and owner fields line up before CRM import or reporting.