How to Fix Your Broken MCAE Automation
You inherited a Pardot setup from someone who left the company 18 months ago, and now you're sitting in Engagement Studio staring at a program that has been "active" for three weeks with zero prospects flowing through it. The automation rules exist. The lists are populated. Everything looks configured. And yet nothing is actually working. If this sounds familiar, you are not broken. The setup is.
This article is not a beginner's walkthrough of what Account Engagement does. It's a diagnostic guide for the person who already knows what Salesforce Marketing Cloud Account Engagement (MCAE), formerly known as Pardot, is supposed to do and cannot figure out why it isn't doing it. The reasons automations break in B2B marketing automation platforms like this are almost never where you think they are, and they almost never have a single cause. Here are the real failure points, starting with the ones most teams overlook.
When your Pardot setup looks fine but clearly isn't
Inherited MCAE configurations have a specific kind of dysfunction. Everything is technically present: automation rules, dynamic lists, Engagement Studio programs, completion actions on every form. The problem is that all of it was built by someone who made decisions you weren't there for, and those decisions are now quietly running (or not running) in the background without explanation.
The inherited configuration trap
A legacy Pardot setup typically lands in new hands looking like a full kitchen with every appliance plugged in. What you don't realize until you dig in is that two automation rules are running in parallel doing contradictory things, the field values triggering your branching logic were updated in Salesforce six months ago and no one touched MCAE, and there is zero documentation. This pattern shows up consistently in inherited instances across industries, it's just what years of accumulated workarounds look like. It is not a reflection of your competence.
Why the problem isn't always where you're looking
Most teams assume the issue lives inside Engagement Studio. In reality, the breakdown is often upstream. A sync error is blocking prospect records from updating. A dynamic list's criteria silently excludes the right prospects because an old field value no longer matches what Salesforce sends over. A completion action fires a list add before the prospect ever reaches the Engagement Studio entry criteria. By the time you're debugging the wait step, the real problem is three layers back.
Pardot automation failure points in MCAE
These are the patterns that appear in almost every broken Pardot setup, regardless of industry or company size. Check these before you touch anything else, and keep in mind that most teams encounter at least two or three of them stacked on top of each other.
Engagement Studio logic that breaks at the branch level
The most frequent Engagement Studio failure is a wait step that never advances because the rule condition it's waiting on is never actually met. Here's a concrete example: a program has a "Wait up to 5 days" step followed by a "Has the prospect clicked any email?" check. If the program was built using a plain "Wait" step instead of "Wait up to a maximum of," every single prospect holds for the full duration regardless of their behavior. The distinction is small, but the impact is that your entire nurture sequence stalls.
That's not the only way branch logic breaks. "If/then" branches built on outdated picklist values send prospects down paths that no longer exist. Entry criteria that are too narrow leave programs with zero prospects in queue despite showing as fully active, a silent failure that produces no error message and no dashboard alert. Each of these issues is fixable, but only once you know where to look.
Pardot sync errors silently stopping prospect updates
This one is the sneakiest because the automation itself looks perfectly healthy. If a prospect's Salesforce record can't sync due to a validation rule mismatch or a missing required field, scoring updates, list membership changes, and field updates may never apply. Salesforce evaluates validation rules against the entire record even when Pardot is only updating a single field like Score. The sync fails, the prospect stays stuck, and the automation keeps running as if nothing is wrong. The prospect just never moves through it.
The silent gaps teams miss until a campaign completely flops
These failure points don't throw error messages. They don't show up in any dashboard. They quietly destroy campaign performance while you're busy wondering why lead volume is down, and because they produce no visible signal, they're often the last thing teams think to check.
Completion actions firing in the wrong sequence
Completion actions and automation rules run independently in MCAE. That is not a bug, it's by design. But here's what happens when teams don't account for it: a form submission fires a completion action that adds a prospect to List A, while an automation rule running simultaneously removes them from List A based on a field value that was just updated by that same form. The result is unpredictable, and it's different for every prospect depending on which process runs first. If you're using completion actions and automation rules to manage the same list or field, audit those overlaps before anything else.
Lead scoring and list membership rules that have aged out
Scoring models built two or three years ago reflect how a completely different team defined a qualified lead. As a common risk, outdated thresholds can push prospects into nurture streams they don't belong in, or suppress them from campaigns entirely because their score exceeds a threshold that no longer means anything. Dynamic list criteria does the same when it still references retired lifecycle stages, old campaign names, or field values that haven't been active in Salesforce for a year. Prospects match or get excluded based on rules that no longer reflect reality, and nothing in the platform flags it.
How to diagnose broken Pardot automations instead of guessing
Auditing an inherited MCAE instance is not glamorous work, but it is the only way to understand what you're actually dealing with before you start making changes that create new problems. If you want expert help working through this, I (Molly Purcell-Weatherwalks) work with B2B marketing teams on exactly this: diagnosing legacy Pardot configurations, identifying accumulated misconfigurations, and rebuilding automation logic against current MCAE best practices. Start with the audit below regardless, it tells you where to focus first.
The four areas to audit before touching anything else
Run through this checklist before you change a single automation rule or rebuild any Engagement Studio program:
- Sync error log in the Salesforce connector: Check for validation rule failures and missing required fields that are blocking prospect updates.
- Engagement Studio prospect activity view: Confirm that actual prospects are moving through your programs, not just that the program shows as active.
- Dynamic list criteria: Verify that every filter still maps to valid, current Salesforce field values and that "match all" versus "match any" logic is intentional.
- Completion action and automation rule overlap on key forms: Find every place where both systems are acting on the same list, field, or assignment, and decide which one owns that action.
When patching the old config creates more problems than it solves
Some inherited Pardot setups are too layered with workarounds to fix incrementally. Every patch applied to a misconfigured foundation creates new edge cases. At a certain point, the faster path is a clean migration of the legacy automation logic into a properly structured Marketing Cloud Account Engagement configuration. I do this work regularly: untangling years of accumulated configuration debt, documenting what was actually intended versus what was built, and rebuilding it so nothing gets missed and nothing quietly breaks a month after you think it's fixed.
Your broken Pardot automations didn't happen overnight
Broken Pardot automations are almost never caused by one thing. They're the result of several small misconfigurations stacking on each other over months or years, until a campaign completely misfires and someone finally has to dig in. You now have a clearer map of where to look: sync errors upstream, Engagement Studio logic that was never validated, completion actions conflicting with automation rules, and list criteria that stopped reflecting reality a long time ago.
Many teams find that working with an external specialist significantly reduces the time it takes to untangle complex inherited setups, especially when the layers of workarounds make it hard to know what's safe to change. If your setup has reached that point, shoot me a message for Pardot troubleshooting or to start a legacy migration conversation. Your campaigns will keep misfiring until the foundation is correct. That's the only thing worth fixing.