Your longest calls tend to start with your best leads. We know, because we scored the openings and then checked the clock.
CallCoach, Icana's call-analysis product, listens to the first minute of a sales or service call and rates the enquiry from 0 to 10. The score comes from two things: the caller's tone and emotions, and what they actually say. It reads the opening the way a sharp agent does, before your agent has finished their second question.
Then we lined those opening-minute scores up against how long each call ran. Across millions of minutes of calls, from multiple organisations, one pattern held.
| Call length | Average lead quality in the opening minute (0 to 10) |
|---|---|
| Under 2 minutes | ~3.4 |
| 2 to 5 minutes | ~4.6 |
| 5 to 10 minutes | ~5.1 |
| 10 to 20 minutes | ~5.7 |
| Over 20 minutes | ~6.2 |
The longer the call, the higher it started. Every bracket, no exceptions. Calls that ran past 20 minutes began with leads scoring close to twice as high as calls that died inside 2 minutes.
The score never sees the ending. It reads the emotions and content of the opening, before anyone knows how the call will go. The score comes first; the duration follows. Each bracket covers tens of thousands to hundreds of thousands of calls, and no single organisation dominates any of them.
Why the eager caller stays on
There's no mystery here. A caller with a real problem settles in. They have questions, they want to know whether you can help, and they are willing to spend the time finding out. A lukewarm caller finds a reason to get off the phone.
Your agents have always known this in their gut. It's why the argument between marketing and your floor never ends. Marketing says they sent you plenty of leads. Your agents say the leads were rubbish. Each side is holding a different impression of the same first minutes, and nobody has measured either one. Score those minutes at scale and the argument finally has an answer.
What it means for your handle time
Here's the edge. If the lead largely sets the length of the call, then ranking agents, teams or campaigns on average handle time measures your lead mix as much as anyone's productivity.
Picture your own dashboard. The agent working your eager enquiries runs long calls and looks slow. The campaign pushing weak leads produces short calls and looks efficient. A team leader who tells agents to tighten up their handle time may be cutting short the exact conversations that were going somewhere.
Keep in mind that there is some bias in this data set: the shortest calls (under 2 minutes) include callers who hang up, or who do not have time at that moment. Take the pattern as something we've noticed, not necessarily a target to hit.
None of this retires average handle time. It keeps its job in capacity planning, where you need to know how long calls run to staff the floor. As a way to compare people and campaigns, it needs a lead-quality column beside it. Without one, it quietly rewards whoever got the worst leads.
So go and look. Pull last quarter's calls and bucket them by duration, the way we did. Then ask what you actually know about the quality of the enquiries in each bucket.
If the honest answer is nothing measured, that's the gap. Every handle-time conversation you have ever had was missing its control variable. Measuring every call, rather than a sampled few, is the same coverage problem we wrote about in the QA coverage gap.