Lead generation · 2025
Bringing an abandon rate back under the line without losing contact volume
A predictive dialer tuned for raw throughput had pushed the abandon rate into territory the client’s compliance counsel would not sign off on. Fixed the pacing without surrendering connects.
- Abandon rate
- 2.6% → 0.9%
- Sustained over 90 days
- Connected calls
- +11%
- Despite less aggressive pacing
- AMD false positives
- −72%
- Live humans no longer dropped
- Compliance exceptions
- 0
- Since deployment
The situation
The operation had grown by adding lines rather than tuning pacing. Abandon rate sat above the threshold their counsel considered defensible under TCPA.
The obvious fix — dial less aggressively — would have cut connected calls by roughly a third, which the business would not accept.
Answering-machine detection was mistuned badly enough that live humans were being dropped as machines, which inflated the abandon figure further.
What I did
- 01
Instrumented the dial pipeline to separate genuine abandons from AMD misclassifications, which had been counted together.
- 02
Retuned answering-machine detection against several thousand recorded calls rather than defaults.
- 03
Rebuilt pacing per campaign against measured agent-availability curves instead of one global ratio.
- 04
Implemented per-state calling rules and DNC scrubbing as a hard gate ahead of the dial list.
- 05
Set up scheduled compliance reporting so the abandon rate is visible daily rather than discovered quarterly.