The Challenge
Aspen Creek Wealth Strategies uses service tiers to determine the level of attention each client project receives. The firm had an automation that was supposed to assign these tiers automatically whenever a new project was created.
But the automation had a logic error. It was using the wrong criteria to classify projects, which meant hundreds of projects had incorrect or missing tier assignments. A team member spent 2.5 hours trying to fix it and could not resolve the issue.
The broken automation was not just an inconvenience. Wrong tier assignments meant some clients were getting less attention than they were paying for, while others were consuming resources above their service level. The firm needed it fixed immediately.
What We Built
We rebuilt the automation from scratch. The new system checks the right criteria, handles edge cases that the old one missed, and includes safeguards to prevent the same kind of logic errors from happening again.
We also built in protections so the automation does not override manual assignments. If someone on the team manually sets a tier, the system respects that and does not overwrite it.
For the 1,917 existing projects, we ran a one-time cleanup that audited every single record, identified the 83 that had wrong data, and corrected them. Zero errors in the cleanup process.
The Result
Every new client project now gets the correct service tier automatically the moment it is created. No manual assignment needed.
The 83 misclassified projects were corrected with zero errors. Clients are now receiving the service level they are paying for.
The team member who spent 2.5 hours trying to fix the old automation no longer needs to think about it. The system runs reliably in the background without oversight.
The Takeaway
1. Broken automations cost more than no automation.
A system that assigns wrong tiers is worse than no system at all. When automation handles something important, it needs to be right, not just fast.
2. Fix the data, not just the process.
Rebuilding the automation was only half the job. The 83 existing records with wrong data needed to be cleaned up too, or the same problems would persist.
3. Safeguards prevent repeat failures.
The new system includes checks that the old one lacked. It cannot override manual assignments, it handles edge cases, and it validates its own output.
Ready to be our next case study?
Book a call to discuss your workflows. We'll show you exactly what we can automate.
Get in touch