Your Marketo instance is probably messier than you think. If you inherited it from a previous team, have not reviewed it in 12 months, or noticed sync errors piling up in Salesforce, it is time for an audit.
A Marketo audit is not just clicking through the admin panel. It is a structured review of every layer: data quality, program architecture, scoring logic, CRM integration, and reporting. Here are the 10 steps we use when we audit Marketo instances for B2B clients. marketing ops and automation services
Step 1: Database Health Check
Start with the foundation. Pull your total record count and check:
- Duplicate rate: Export your database and run a match on email, company name plus domain, or first/last name plus company. Anything above 5% duplicate rate needs immediate action.
- Marketable records: How many records are actually marketable (valid email, not unsubscribed, not bounced)? If less than 60% of your database is marketable, you have a data rot problem.
- Field usage: Count your custom fields. Most inherited instances have 200 or more fields, and half are unused. Flag any field with less than 10% population rate.
Red flag: If your bounce rate exceeds 3% on any email send, your database health is critically low.
Step 2: Naming Convention Review
Open Marketing Activities and check whether programs, campaigns, emails, and landing pages follow a consistent naming pattern.
A clean naming convention looks like: `YYYY-MM_Type_Campaign-Name_Asset`. For example: `2026-04_WBN_Q2-Pipeline-Summit_Invite-Email`.
If you see names like “Test email copy 2 FINAL,” “Campaign – new version,” or folders named “Old stuff,” the instance has no governance. This slows down every person who touches it and makes reporting unreliable.
Step 3: Program and Folder Structure
Check how Marketing Activities is organized:
- Are programs grouped by quarter, type, or business unit?
- Are there orphaned programs sitting outside any folder?
- Are there active smart campaigns inside programs marked as “Archive” or “Old”?
Red flag: Programs with no members or zero activity in the last 6 months that are still running trigger campaigns. These consume processing power and create confusion.
Step 4: Smart Campaign Logic Review
Pull a list of all active trigger campaigns and batch campaigns. For each:
- Is the trigger filter specific enough? A trigger on “Fills Out Form” with no form filter fires on every form submission across the entire instance.
- Are there campaigns with overlapping logic? Two campaigns triggered by the same action that do different things create unpredictable results.
- Are there batch campaigns running daily that should be triggers, or vice versa?
Audit the request campaign queue for bottlenecks. If campaigns have high latency (processing delay above 30 minutes), the instance is overloaded.
Step 5: Lead Scoring Model Assessment
Open your scoring model and check: analytics and RevOps reporting
- Fit score: Are you scoring on firmographic data (industry, company size, job title)? Does it match your actual ICP?
- Behavior score: Are you scoring meaningful actions (pricing page visit, webinar attendance, demo request) or just email opens?
- Score decay: Is there a decay mechanism? Without decay, a lead who engaged 18 months ago and went silent still shows as “hot.”
- Threshold calibration: What score triggers MQL status? Pull your last 50 closed-won deals and check what their score was at MQL. If the threshold does not match reality, it needs adjustment.
Red flag: If more than 40% of your MQLs are rejected by sales, the scoring model is broken.
Step 6: CRM Sync Health
Check the Marketo-Salesforce sync:
- Sync errors: Go to Admin > Integration > Salesforce and check the sync error log. Any persistent errors (field validation failures, ownership conflicts, duplicate rules blocking inserts) need immediate fixes.
- Sync frequency: Marketo syncs with Salesforce every 5 minutes by default. If records are not updating for hours, something is blocking the queue.
- Field mapping: Review which fields sync bidirectionally vs. one-way. Misaligned field mapping is the most common source of data discrepancies between MAP and CRM.
Red flag: More than 100 unresolved sync errors. This means records are falling out of sync silently, and your CRM data does not match your MAP data.
Step 7: Email Deliverability Check
Review your email sending health:
- Bounce rate: Hard bounces above 1% per send indicate a list hygiene problem. Soft bounces above 5% indicate a reputation or infrastructure issue.
- Unsubscribe rate: Anything above 0.5% per send means your content or frequency is off.
- Spam complaints: If complaints appear in your metrics, your sender reputation is at risk.
- SPF, DKIM, DMARC: Check that all three authentication records are properly configured. Missing any one of these tanks deliverability.
Step 8: Landing Page and Form Audit
Review all active landing pages and forms:
- Are forms using progressive profiling or asking 10 fields on the first touch?
- Are landing page URLs following a clean convention?
- Are there published landing pages with no form or a broken form?
- Are UTM parameters being captured on forms for campaign attribution?
Red flag: Forms that do not capture UTM source, medium, and campaign. Without these, your attribution reporting is blind.
Step 9: Reporting and Attribution
Check what reporting exists:
- Are program success steps defined correctly? (A webinar program where “Registered” counts as success instead of “Attended” inflates your numbers.)
- Is multi-touch attribution configured?
- Are revenue cycle analytics (RCA) models set up if you have the module?
- Can you pull a report that shows marketing-sourced pipeline by channel?
If the answer to the last question is no, reporting infrastructure is the most urgent fix.
Step 10: Documentation and Governance
The final step is the one most teams skip:
- Is there a documented process for creating new programs?
- Is there an onboarding guide for new team members?
- Are there written rules for when to use triggers vs. batches, how to name assets, and how to handle data imports?
If the answer to all three is no, every fix you make in steps 1 through 9 will degrade within 6 months as team members build without guardrails.
Prioritizing Your Fixes
After completing all 10 steps, rank every finding by two factors:
- Impact on revenue: Does this issue directly affect lead quality, pipeline reporting, or sales follow-up?
- Effort to fix: Can it be done in an afternoon, or does it require a multi-week project?
Start with high-impact, low-effort fixes. Sync errors, scoring threshold adjustments, and naming convention documentation fall here. Save the large projects (full database deduplication, scoring model rebuild, program restructure) for a phased roadmap.
If your Marketo instance needs more than a tune-up, or if you do not have the internal ops resources to execute the fixes, Kanvl runs full marketing ops audits and cleanups for B2B teams running Marketo and Salesforce. The engagement includes the audit, a prioritized backlog, implemented fixes, playbooks, and training so your team can maintain the system. book a strategy call
