Engineering · 4 min read
AI calling tools end calls on the wrong cues all the time. Here’s the discipline that fixes it.
The single most expensive AI calling failure mode we’ve seen is the agent ending a call on the wrong cue. “Aur kuch nahi” gets misclassified as “not interested”. The agent says goodbye. A high-intent lead — who simply ran out of questions — drops.
We codified this explicitly. Every strategy template ships with two lists. GREEN lights mean stay on the call: “sochenge”, “dekhte hain”, “family se baat karke bataenge”, “aur kuch nahi”, silence under 4 seconds. RED lights mean end politely: “mujhe interest nahi hai”, “wrong number”, confirmed appointment booked, three escalating “hello?” with no response.
The lists are reinforced at three layers: the system-prompt prelude, the end_call tool description, and a final compliance trailer. If one layer gets overridden by a long conversation context, the others catch it. End-call false-positives are under 1 % across our deployed customers.
We learned this the hard way after a customer’s top lead got dropped on “aur kuch nahi” in early April. The discipline is the fix.
Want to see how this works on your business?
Book a 20-min demo →