River
Y CombinatorBacked by Y Combinator
FREE TEMPLATE

Sales Call Coaching Scorecard Template

Every criterion typed by who controls it, the conditions recorded before anything is scored, and follow-through counted in calls rather than weeks.

Free download  ·  No account needed

Call Scorecard · Norbeck procurement software · two weeks, four reps

Three reps score in the bottom band, and only on the calls somebody else booked.

CallRepBooked byBookedHeldMost senior attendeeRep-controlledVerdict
C-3041A. DuthiePartner3014Purchasing assistant11 / 20Conditions. Route it
C-3052A. DuthieRep4547Operations director18 / 20Skill confirmed
C-3085E. FaradayPartner3013Purchasing assistant10 / 20Conditions. Route it
C-3089E. FaradayRep6061Plant manager15 / 20Skill confirmed
C-3074C. MowbrayRep6022Purchasing assistantnot scoredNot scorable
C-3070L. RanjanRep4544Finance director11 / 20Skill gap. Coach it

Every partner-booked meeting ran under half its length with nobody who can spend money in the room, and every rep who took one scored in the bottom band on it and in the top band on their own calls. That is a booking problem with a partner manager's name on it. The last row is the opposite case and the reason this cannot be used as an excuse: full length, finance director present, and still eleven out of twenty. That one is coaching.

Page one for this query settled recently and settled completely. Five to twelve observable criteria per call type. Each one written as a trigger and an action so the grade rests on an event rather than an adjective. Anchors at one, three and five. Weighting by leverage against the win and loss split. Not applicable specified as a designed outcome. Manager calibration to a stated agreement threshold. Three to five calls per rep per week, one skill per session, scores compared week over week.

What none of it separates is a criterion the rep controls from one decided before they dialled. Scoring discovery depth on a fourteen-minute meeting a partner booked, with a purchasing assistant in the room, scores the calendar. Design of experiments has a name for this: a nuisance factor moves the measured result and is not the thing under study. Every criterion here is typed rep, jointly determined, or set before the call.

The second absence is follow-through. Acted on, yes, written the following Tuesday is not evidence, because everybody does the thing once while it is fresh. Each piece of feedback becomes one behaviour counted in calls: present, absent, or the trigger never fired. Two live behaviours per rep, and the third silently ends the oldest. Aviation requires a qualification programme to prove its training and evaluation are accomplishing their objectives. The review that inspects the deal is a separate job.

What each criterion measures, and whether last month's feedback survived

One fictional team throughout: Norbeck sells procurement software, four reps, twelve scored calls in a fortnight. Every figure ties to the others.

Criterion Control Map

Every criterion typed by who controls the outcome, with the owner a repeated failure routes to.

CriterionTriggerControl classWhere a failure routesIn the coaching score
Named the costThey describe a problemrepCoachingyes
Asked who else decidesAny call past five minutesrepCoachingyes
Booked the next step liveThe last five minutesrepCoachingyes
Clarified before answeringAn objection is raisedrepCoachingyes
Recap sent same dayThe call endsrepCoachingyes
Right people in the roomAny scheduled calljointly determinedRep and whoever booked itpartly
Enough time heldAny scheduled calljointly determinedRep and whoever booked itpartly
Meeting length bookedAny scheduled callset before the callWhoever books itno
Seniority of the attendeeAny scheduled callset before the callOwner of that lead sourceno
Stage fitAny scheduled callset before the callSales operationsno
Both sides audibleAny recorded callset before the callNot scorable when it failsno

Five criteria carry the coaching score. Four of the eleven were decided before anybody dialled, and on a conventional scorecard all eleven feed one number that gets read out to the rep. The test for typing is one question: could a rep at the top of their game fail this criterion on a badly booked call?

Behaviour Runs

Every piece of feedback as one named behaviour, counted in scored calls where its trigger fired.

RepBehaviourGivenCalls sincePresentAbsentRunLongestBroke onState
A. DuthieAsk what the problem costs before proposing anything12 May98155C-3041, a 14-minute partner callholding
A. DuthieBook the next step before hanging up2 Jun44044noneholding
L. RanjanAsk one clarifying question before answering an objection21 Apr144912C-3061lost, reopened 9 Jun
L. RanjanAsk who else has to agree5 May113801C-3067lost, never re-coached
L. RanjanSend the recap the same day9 Jun33033noneholding, and the third live on this rep
C. MowbrayName the cost as a range when they will not give a figure14 Apr161511111C-3059closed 12 Jun
E. FaradayAsk who else has to agree19 May76133C-3085, a 13-minute partner callholding
E. FaradayBook the next step before hanging up4 Jun52301C-3089at risk

Eight consecutive calls where the trigger fired closes a behaviour, which is why one is marked closed at a run of eleven. L. Ranjan is carrying three: the two oldest both lost their runs and one has not been mentioned in six weeks while the log still reads live. Two of the broken runs ended on thirteen and fourteen-minute partner calls, so those pause rather than reset.

Conditions Ledger

Patterns in the conditions, each with the person who owns the fix. None of these is a rep.

PatternCriterion affectedRepsCallsEvidenceOwnerOutcome
Partner-booked meetings run half the length they were booked forEnough time held3414, 16, 13 and 22 minutesPartner manager, OnwardRe-briefed, next four averaged 27
Partner-booked meetings put a purchasing assistant in the roomRight people in the room34Four of fourPartner manager, Onwardopen
Content-sourced inbound arrives at plant manager level and no higherSeniority of the attendee12Two of twoDemand generationopen
Demos booked before any discovery call on the castings segmentStage fit23No discovery call loggedSales operationsStage gate added 14 Jun
Audio drops on calls dialled from the mobile appBoth sides audible22Two calls unscorableITclosed 11 Jun
Objections answered before they are understoodClarified before answering13Full length, senior attendee, all threeNobody. This one is coachingBehaviour reopened

The last row is deliberate and it is what stops the ledger becoming an excuse machine. Every rep-controlled criterion is scored again across clean-condition calls alone, and anything still failing there is a skill gap owned by nobody but the rep. Three full-length calls with a plant manager, a head of supply chain and a finance director on them is not a conditions problem.

What is in the pack

01

Criterion Control Map

Every criterion typed rep, jointly determined or set before the call, with the trigger that fires it and the owner a repeated failure routes to.

02

Call Scorecard

One row per scored call carrying the conditions it happened under, the score split three ways by control class, and a verdict of coach, route or not scorable.

03

Behaviour Runs

Each behaviour counted in calls with its current run, its longest run ever, and the specific call that broke it, so a pause is not mistaken for a regression.

04

Conditions Ledger

The patterns nobody in a development plan can fix, each with the evidence in figures and a single named owner outside the sales floor.

05

Coaching Framework

The three control classes, the eight consecutive calls that close a behaviour, and the two-behaviour capacity limit that keeps the third from killing the first.

06

Scoring Conditions Standard

What must be recorded before any criterion is marked, the four cases where a call is not scorable at all, and why absent and not applicable are different.

07

Feedback Format

Behaviour, because, marked, closes at. Written so a stranger listening to the next recording can decide present or absent without asking anyone.

08

Coaching Session Note

A worked twenty-five minutes on one rep: three calls with clean conditions, one behaviour, two timestamped moments, and one dead behaviour closed out loud.

How it works

  1. 1

    Type the criteria

    Send your existing scorecard. Each criterion is tested against one question: could a rep at the top of their game fail this on a badly booked call?

  2. 2

    Score with conditions

    Recordings get the booking, the length held, the lead source and the attendee's seniority recorded first, then the score comes back split three ways.

  3. 3

    Route the rest

    A condition failing three times in twenty calls opens a ledger row with the figures and a named owner outside the sales team.

  4. 4

    Count the runs

    Every live behaviour is marked on each new call, and the report is the run length and the call that broke it, not a weekly average.

Frequently asked questions

Is this template free?

Yes. Four Word documents and four CSV sheets, downloaded as a zip, no signup. River is the optional half: it scores the recordings, records the conditions and counts the runs. Single-shot jobs live in River's tool index and the rest in the template library.

Does this replace our scorecard?

No, it types the one you already have. Most criteria survive unchanged. What changes is that four or five of them turn out to have been decided before the rep dialled, and pulling those out of the coaching score usually moves a rep out of the bottom band. The question set they ask from is separate work.

Why count runs in calls instead of weeks?

Because four calls a week and one call a week are not the same fortnight. Eight consecutive calls where the trigger actually fired means the same thing for both reps. A weekly trend line means the busier rep is judged on more evidence and nobody notices. Whether the volume itself predicts anything is a separate test on the activity log.

Is two behaviours per rep not too few?

It is what a person holds. A third does not add capacity, it silently ends the oldest, and no coaching log has ever recorded that happening. In the worked example one rep carries three, and the two oldest both lost their runs while the log still reads them as live.

Could a rep hide behind the conditions ledger?

No, and the pack is built to stop exactly that. Every rep-controlled criterion is scored again across clean-condition calls alone. The worked example carries a row for a rep failing on three full-length calls with senior attendees present, owned by nobody, marked as coaching.

What makes a call not scorable?

The buyer left before the criteria could trigger, the recording is incomplete, the call type does not match its own rubric, or the rep was not leading it. Mark the whole call, do not part-score it. Absent means the trigger fired; not scorable means it never did.

How is this different from a deal review?

Different unit. A deal review inspects one opportunity and asks who has the authority to unblock it. This inspects one call and asks what the rep did with the thirty minutes they were given, and whether the thirty minutes were actually thirty.

Find out whether your scores are about the rep

Send five recordings and say who booked each one. The first pass usually finds somebody marked down for a calendar.

Edit with AI