Workflow Audit Overview
Workflow audits are a specialized audit type focused exclusively on HubSpot workflows. While a standard portal audit evaluates workflows as one of many areas, a workflow audit goes deep — analyzing individual workflow structures, detecting conflicts between workflows, mapping dependencies, and scoring workflow health.
How Workflow Audits Differ from Portal Audits
Section titled “How Workflow Audits Differ from Portal Audits”| Aspect | Portal Audit | Workflow Audit |
|---|---|---|
| Scope | Entire portal across all hubs | Workflows only |
| Data points | 450+ across 6 sections | Workflow-specific metrics |
| Output | Portal score, section scores, block insights | Workflow health scores, conflict reports, dependency maps |
| Use case | General portal health assessment | Deep analysis of automation layer |
| AI insights | Block-level and executive summary | Workflow-specific findings and recommendations |
Workflow audits are not a replacement for portal audits. They are a complementary, focused analysis for when workflows are the primary concern.
When to Run a Workflow Audit
Section titled “When to Run a Workflow Audit”Run a workflow audit when:
- A client reports automation issues — Contacts receiving duplicate emails, wrong list placements, or conflicting actions
- Workflows have grown organically — The portal has dozens or hundreds of workflows built by different people over time
- Before a major automation overhaul — Understand the current state before making changes
- After importing workflows — Verify that imported workflows do not conflict with existing ones
- During regular maintenance — Periodic workflow audits catch issues before they affect contacts
Workflow Audit Components
Section titled “Workflow Audit Components”The workflow audit engine consists of six components that work together:
1. Asset Fetcher
Section titled “1. Asset Fetcher”Retrieves all workflows from the connected HubSpot portal, including their full configuration — triggers, actions, branches, filters, and enrollment criteria. The fetcher pulls both active and inactive workflows to provide a complete picture.
2. Graph Builder
Section titled “2. Graph Builder”Constructs a dependency graph of all workflows. The graph maps:
- Workflow-to-workflow connections — Where one workflow enrolls contacts into another
- Shared resources — Lists, properties, and objects that multiple workflows read from or write to
- Trigger overlaps — Multiple workflows that trigger on the same criteria
The graph is the foundation for conflict detection and dependency analysis.
3. Conflict Detector
Section titled “3. Conflict Detector”Analyzes the dependency graph to identify conflicts between workflows:
- Competing actions — Two workflows trying to set the same property to different values for the same contact
- Enrollment conflicts — A contact meeting criteria for conflicting workflows simultaneously
- Suppression gaps — Workflows that should suppress certain contacts but do not
- Circular dependencies — Workflow A triggers B, which triggers A, creating an infinite loop
- Race conditions — Multiple workflows acting on the same contact in an unpredictable order
Each detected conflict includes the affected workflows, the nature of the conflict, and its potential impact.
4. Health Scorer
Section titled “4. Health Scorer”Evaluates each workflow against best practices and assigns a health score:
- Naming conventions — Does the workflow follow consistent naming patterns?
- Enrollment criteria — Are triggers well-defined and specific?
- Branching logic — Are if/then branches structured cleanly?
- Suppression lists — Are appropriate contacts excluded?
- Re-enrollment settings — Are re-enrollment rules configured correctly?
- Goal criteria — Is there a defined exit condition?
The health score is a 0-100 value per workflow, similar to data point scoring in portal audits.
5. Insight Generator
Section titled “5. Insight Generator”Produces structured findings from the conflict detection and health scoring results. Insights explain what was found, why it matters, and what to do about it.
6. AI Insight Generator
Section titled “6. AI Insight Generator”Uses Claude Haiku to generate natural-language summaries of the workflow audit findings, following the same model and parameters as portal audit AI insights (temperature 0.3, structured JSON output).
What You Get
Section titled “What You Get”After a workflow audit completes, you receive:
- Workflow health scores — Individual scores for each workflow
- Conflict report — All detected conflicts with severity ratings
- Dependency map — Visual representation of workflow relationships
- AI insights — Natural-language findings and recommendations
- Prioritized recommendations — Actions to resolve conflicts and improve health
See Running a Workflow Audit to get started, and Understanding Results for a detailed guide to interpreting the output.