The template
Copy this block into a doc or spreadsheet. One rubric per conversation you certify.
CERTIFICATION: [conversation, e.g. first discovery / pricing / renewal]
WHY NOW: [launch, new segment, or the stage where deals stall]
METHOD: [your methodology, e.g. MEDDPICC, Gap Selling, ValueSelling Framework®]
PASS RULE: all must-pass items met AND scored items average at or above [bar]
on the scenario AND on one live call within 14 days
| # | Item (observable in a transcript) | Must-pass? | Level 3: strong + reason | Level 2: adequate + reason | Level 1: miss + reason | Evidence: scenario | Evidence: live call |
|---|---|---|---|---|---|---|---|
| 1 | [what the rep does or the buyer confirms] | Y/N | [what it sounds like] / Reason: [why a marker gives 3] | [...] / Reason: [...] | [...] / Reason: [...] | [score + line] | [score + buyer's line] |
| 2 | | | | | | | |
| 3 | | | | | | | |
| 4 | | | | | | | |
| 5 | | | | | | | |
SIGN-OFF
Scenario passed on: [date] Attempts: [n] Marker: [name]
Live call signed off on: [date] Deal: [account] Reviewer: [name]
Named gap for next call: [item #, one line]
Hold check (3 weeks): [date] Still present? [Y/N, calls checked]
Five to eight items is the useful range. Each item must be something two reviewers could point to in a transcript.
How to use it
- Name the conversation and the reason. One rubric per conversation. Write the "why now" line so managers know which deals it applies to.
- Write the items from your playbook's exit evidence. If your sales playbook says a deal leaves discovery when the buyer has stated their success metric, item 1 is "buyer states the metric in their own words". Items describe what is on the call, so "confident" and "consultative" do not qualify.
- Mark the must-pass items. Accuracy and compliance items, such as pricing quoted correctly, are pass or fail. Skills are scored.
- Write the reason for every level. This is the column that matters most. In our call-scoring research with PeerLab, adding experienced markers' reasons lifted our scorer's average agreement with human scores from 0.62 to 0.72, the largest single gain; adding textbook examples added nothing.
- Calibrate before launch. Three managers score the same three recorded calls alone, then compare. Rewrite the reason wherever they split. In our call-scoring research, trained experts disagreed on 18% of scores even with a good rubric, so expect to do this twice.
- Score the scenario, then the live call. The rep passes the scenario, then runs the conversation with a real buyer within two weeks. The same rubric is applied to the transcript. A pass on both is the certification.
- Run the hold check at three weeks. Score two or three of the rep's later calls of the same kind. If an item has slipped, it goes to the next coaching call.
The full procedure, with the reasoning behind each step, is in how to certify sales reps. It is part of the sales readiness guide.
- Before launchCalibrate
- Three managers, three recorded calls
- ScenarioScore the scenario
- Within two weeksScore the live call
- Same rubric, a real buyer
- Three weeksHold check
- Two or three later calls
Filled example: MEDDPICC first discovery (illustrative)
The company, rep and lines are illustrative.
CERTIFICATION: First discovery
WHY NOW: deals stall after discovery without an agreed success metric
METHOD: MEDDPICC
PASS RULE: items 1 and 4 met; items 2, 3, 5 average 2.0+;
scenario AND one live call within 14 days
| # | Item | Must-pass? | Level 3 + reason | Level 2 + reason | Level 1 + reason |
|---|---|---|---|---|---|
| 1 | Buyer states the metric they will judge success by, in their words | Y | "On-time is 91% against 96%." Reason: measure and number from the buyer | Measure named, number to follow. Reason: metric named, number pending | "It's a big problem." Reason: no measure; deal cannot be sized |
| 2 | Rep quantifies the cost of the gap | N | Buyer puts a cost on it. Reason: cost in buyer's terms | Rep proposes a cost and buyer agrees to check. Reason: cost hypothesized | No cost discussed. Reason: no case for change |
| 3 | Rep identifies who else feels the problem | N | Two named roles beyond the contact. Reason: reach confirmed | One role named. Reason: partial reach | None. Reason: single-threaded |
| 4 | Rep agrees a dated next step with a buyer action | Y | "You'll send last quarter's report by Friday." Reason: buyer action, date | Meeting booked, no buyer action. Reason: date without commitment | "I'll follow up." Reason: no next step |
| 5 | Rep tests the decision process | N | Buyer describes who signs and how. Reason: process confirmed | Rep asks, buyer is unsure. Reason: process open, flagged | Not raised. Reason: process assumed |
Sign-off. Maya passed the scenario on her first attempt. Her live call with an illustrative account, Tavistow Freight, scored 3 on items 1 and 4, 2 on items 2 and 3, and 1 on item 5. Must-pass items met, scored average 1.67 against a bar of 2.0, so the live-call step was a retry. On her next discovery call the coach's prep covered the decision process, it scored 2, and the average reached 2.0. Result: certified, with item 5 noted as the item to watch. Hold check three weeks later: items 1 and 4 present on both calls checked.
Second filled example: the pricing conversation (illustrative)
Pricing is the other conversation teams most often certify, because a wrong quote costs money and a weak one costs margin. The rep and account below are invented.
CERTIFICATION: Pricing and proposal call
WHY NOW: discounts agreed on the call before value is restated
METHOD: your pricing playbook
PASS RULE: items 1 and 2 met; items 3, 4, 5 average 2.0+;
scenario AND one live pricing call within 14 days
| # | Item | Must-pass? | Level 3 + reason | Level 2 + reason | Level 1 + reason |
|---|---|---|---|---|---|
| 1 | Price and terms quoted match the current price list | Y | Correct and stated plainly. Reason: accurate | n/a | Any error. Reason: fails accuracy |
| 2 | No discount offered before the buyer asks | Y | Buyer asks first. Reason: rep held the line | n/a | Rep offers a discount unprompted. Reason: margin given away |
| 3 | Rep restates the buyer's problem and its cost before the number | N | Buyer's own words and cost, then price. Reason: value anchored | Problem restated, no cost. Reason: partial anchor | Price first. Reason: no anchor |
| 4 | Rep handles a price objection with a question | N | Asks what the buyer is comparing against. Reason: objection explored | Explains the price. Reason: defended without exploring | Concedes. Reason: no handling |
| 5 | Rep agrees who signs and by when | N | Named signer and date. Reason: path to close | Signer named, no date. Reason: partial | Not raised. Reason: path assumed |
Sign-off. Casey passed the scenario on a second attempt, after quoting an old price on the first. On a live call with an illustrative account, Pinebrook Clinics, items 1 and 2 were met and items 3, 4 and 5 scored 3, 2 and 2. Certified, with item 4 as the named gap for the next coaching call.
Turning vague items into observable ones
Most rubrics fail at the item column. If two reviewers cannot point to the same line in a transcript, they will score it differently. Rewrite every item until the evidence is something someone says, and keep the reasons column beside it: in our research, observable items scored without written reasons did worse.
| Vague item | Observable rewrite |
|---|---|
| Builds rapport | Buyer shares something about their role or team the rep did not ask about |
| Strong discovery | Buyer states the problem, its cost and who else feels it |
| Confident on pricing | Rep states the price without apologizing or offering a discount |
| Good close | Rep and buyer agree a dated next step with a buyer action |
| Knows the product | Rep answers the buyer's technical question correctly, or names who will by when |
A quick test: read the item aloud to a manager who was not on the call, then hand them the transcript. If they can find the evidence in a couple of minutes, keep it.
Three ways teams misuse it
Scoring on fractions the scale does not have. A 1-to-3 scale should produce 1, 2 or 3 per item. In one production engine we studied for our call-scoring research, 37% of scores were fractional, and rounding alone added 16 points of exact agreement. Decide the scale, then hold markers to it, and only average at the end.
Comparing two reps' calls side by side. It feels fair, and it is biased toward whichever call is read first. In our research, a model asked which of two calls was better picked the first one it read 62% of the time. Score each call against the rubric on its own.
Stopping at the scenario. The scenario columns get filled, the live-call columns stay blank, and the certificate goes in the LMS. That measures the afternoon of the attempt. Our longer argument is in certification is not the same as closing.
When a rep does not pass
A failed attempt is a coaching plan with the items already named. Handle it the same way every time, so reps see it as part of the process.
- Name the items. Tell the rep which items scored 1 and quote the line from the transcript. Leave the average out of the first conversation.
- Fix a must-pass miss before anything else. A wrong price or a compliance slip goes back to the source material, then a retry of the scenario.
- Coach one scored item at a time. Prep the rep's next real call around it, then score that call.
- Set the retry within two weeks. Long gaps turn a fixable gap into a habit.
- Keep the record. Attempts and dates go in the sign-off block, so a later hold check can see what slipped before.
Recertify when the conversation changes: a new price list, a new product, a new segment or a new method. Otherwise, the three-week hold check plus the rep's next certification cycle is enough.
Running the live-call sign-off for every rep
The live-call column is where most teams run out of manager time. Someone has to prep the rep on the deal, read the transcript, and check again three weeks later.
Our AI sales experts run that part: Caty for MEDDPICC, a pricing playbook or any house method, Keenan for Gap Selling, Val for the ValueSelling Framework®. The rep brings the live deal to the expert on a call, by voice or video, and it preps the conversation against your rubric and methodology, then debriefs the transcript afterward. Typing is the fallback. The expert scores the later calls the rep brings on the same rubric, so the hold check is already done, and Bryce gives managers the read across the team.
Scorecard setup is covered in the scorecard docs. For enablement teams: the enablers page and the build selling skills use case. To report on the program, see how to measure sales enablement, and for the term itself, what is sales readiness?.
Copy the template, fill in the reasons for each level, and calibrate on three recorded calls before the first rep sits the scenario. If you want the live-call column filled for every rep, talk to us.
FAQ
What should a sales certification rubric include? Five to eight items a reviewer can point to in a transcript, a must-pass flag on accuracy and compliance items, three levels per scored item, and the reason a marker would give for each level. Add two evidence columns, one for the scenario and one for a live call, and a sign-off block with a hold check three weeks later.
How do you score a sales certification? Score each item on the stated scale, with whole numbers, against the written reasons. Check must-pass items first, then average the scored items against the bar. Score each call on its own and avoid ranking two calls side by side, which invites order bias. Apply the same rubric to the scenario and to the rep's live call.
What is a good pass mark for a sales certification? Set it from your best reps. Have three of your strongest reps run the scenario, score them, and set the bar where their scored items cluster. All must-pass items should be required regardless of the average. A round number picked in advance tends to be too easy or arbitrary.
How do you make sure managers score certifications consistently? Calibrate before launch: three managers score the same three recorded calls alone, then compare and rewrite the reason for any level where they split. In our call-scoring research, trained experts disagreed on 18% of scores, so expect two rounds. Written reasons for each level were the biggest single fix in that research.
What happens if a rep fails a sales certification? They get a coaching plan built from the items they missed. Fix any must-pass miss first, coach one scored item at a time on the rep's next real calls, and set the retry within two weeks. Record the attempts in the sign-off block.
How often should reps recertify? Recertify when the conversation changes, such as a new price list, product, segment or method. Between changes, the three-week hold check shows whether the skill held.