Call quality scorecard

What the total is, and what it is not

The arithmetic is trivial on purpose: each category's share of the weight, multiplied by the score you gave it, added up. There is no model inside it and no hidden adjustment. What the total is worth depends entirely on whether two people scoring the same call would land in the same place, which is a property of your anchors rather than of this page.

So treat the number as a way of writing down a judgement you already made, not as a measurement that arrived from outside. It is useful for three things: telling one caller's pattern from another's, telling this month's pattern from last month's, and settling an argument about which of two weaknesses is actually costing more. It is not useful as a stick, and a floor that works out you are ranking them on it will start scoring the scorecard.

Watch the biggest-gain figure rather than the headline. A caller sitting at seventy with all the loss in one category has one conversation ahead of them; a caller at seventy with the loss spread evenly across six has a training problem, and coaching them on the category that happens to be listed first will not touch it.

Calibrate before you coach

Before the first score counts for anything, have two managers score the same three calls without talking, then compare category by category rather than total by total. Two scorers often agree on the total and disagree about every line in it, which averages out to a number that looks stable and means nothing.

Where they disagree, the fix is almost always the anchor rather than the scorer. The descriptions in the sheet are written as observable events -- did the caller name the company in the opening, did they establish who signs, did they ask for a specific date -- because a scorer can be wrong about whether a call was good and cannot be wrong about whether a sentence was said. Rewrite any anchor you find yourself arguing about until it names something you could point at in the recording.

Score a handful of calls per caller per week rather than one, and pick them at random from the whole shift. A sample of one is a story, and a sample drawn from the calls a supervisor already remembers is a sample of the memorable ones.

Why a zero in the disclosure category ends the argument

One rule in this sheet is not arithmetic. If you score zero for compliance and disclosure, the verdict comes back as coach whatever the weighted total says, and the sheet tells you it overrode the total rather than quietly adjusting the number. A scoring model that lets an excellent pitch average away a missed disclosure is teaching the floor that the disclosure is optional, and it will be learned.

Which rules bind your dialing is a question about your business rather than about ours: the states you call into, whether the number was consented, what your own script requires the caller to say. Write the answer into the anchor for that category in your own words, review it with whoever advises you on it, and score against that. This page is a scoring sheet, not legal advice, and it deliberately names no regulation for you to score against on our say-so.

The same reasoning explains why the category carries real weight rather than sitting outside the total as a checkbox. A category worth nothing is a category nobody prepares for.

Worked example, using the scores the page loads with

The sheet opens scored against its own starting anchors: opening 4/5, discovery 3/5, objection handling 3/5, compliance 5/5, close 3/5, notes 4/5, against the starting weights of 10/25/20/20/15/10. Each category's earned points are its share of the hundred times its score out of five, and the six add up to seventy-two -- five points under a pass mark of eighty, so the default call comes back coach.

Discovery is where the total goes to die: at 25 percent of the weight and a score of three out of five, it is leaving ten of its own points on the table, more than any other category. Objection handling is next at eight lost. Compliance, scored a clean five, loses nothing -- which is also why it cannot be the reason this call misses eighty. Coach the call on discovery first; the arithmetic says so before anyone has to argue about it.

Where this fits

• Our quality control system -- the longer argument about calibration, live monitoring and coaching that this sheet is the one-page version of.
• Call center outsourcing -- what to ask a vendor about how they score calls, before you sign one.
• Roofing call center services -- the floor these categories were written for.
• Speed-to-lead decay curve -- the other half of a bad month, and the half no scorecard can see: a perfect call placed forty minutes late.