Skills · 21 June 2026 · 2 min read

How to Score Yourself on Discovery Using Gong Data.

After a discovery call - you want to know honestly whether you ran it well, not just whether it felt good
Will Koning
Will Koning
Founder, meritt
meritt illustration: sales tech & ai fluency

After a discovery call - you want to know honestly whether you ran it well, not just whether it felt good

Calls that feel smooth are not always calls that move deals forward. A rep can have a warm, friendly conversation and still leave without understanding the buyer's real problem, who controls the budget, or what happens next. Gong gives you data points - talk ratio, questions asked, topics covered, next steps flagged - that cut through the feeling and show you what actually happened.

Where it goes wrong

Without a structured check, reps mark deals as healthy based on vibe. They miss that they never asked about impact, never confirmed a decision process, and never locked in a next step with a date. Those deals stall and the rep is surprised.

What you'll be able to do

You can run a five-point scorecard on any discovery call using Gong data, spot the gaps, and know exactly what to fix or follow up on.

How to do it

After each discovery call, score yourself 1-5 on five

After each discovery call, score yourself 1-5 on five dimensions: agenda and control, problem depth, stakeholder clarity, next steps, and customer talk time.

Use Gong's talk ratio and question count to check

Use Gong's talk ratio and question count to check your score on control and customer talk time - no guessing needed.

Use topic and keyword filters to check whether pain,

Use topic and keyword filters to check whether pain, impact, timeline, and decision process actually came up - or whether you assumed they did.

Use Gong's action item detection to check whether next

Use Gong's action item detection to check whether next steps were captured with a specific date, owner, and outcome.

Once a month, filter your last 20 calls and

Once a month, filter your last 20 calls and look at won vs lost - see which scorecard dimensions you covered consistently in won deals and where you dropped off in lost ones.

See the difference

Weak

Rep finishes a 30-minute discovery, logs the call as 'good conversation, interested', and moves the deal to next stage. No next meeting booked. No notes on who else is involved.

Strong

Rep opens Gong after the call. Talk ratio: 58% rep, 42% buyer - a bit high. Questions asked: six. Checks topics - pricing came up but impact and decision process did not. Action items: none flagged. Scores herself: control 4, problem depth 2, stakeholder clarity 1, next steps 1, customer talk time 3. Sends a follow-up email that afternoon: 'Two things I wanted to come back to - what does it cost you to leave this unsolved for another quarter, and who else would need to be part of a decision like this?'

You can run a five-point scorecard on any discovery call using Gong data, spot the gaps, and know exactly what to fix or follow up on.

How you'll know it's working

You have got it when you can fill in all five scorecard dimensions from Gong data within ten minutes of a call ending, and your self-scores match what the data shows.

Questions people ask

How do you score yourself on discovery using gong data?

Calls that feel smooth are not always calls that move deals forward. A rep can have a warm, friendly conversation and still leave without understanding the buyer's real problem, who controls the budget, or what happens next. You can run a five-point scorecard on any discovery call using Gong data, spot the gaps, and know exactly what to fix or follow up on.

What is the most common mistake to avoid?

Without a structured check, reps mark deals as healthy based on vibe. They miss that they never asked about impact, never confirmed a decision process, and never locked in a next step with a date.

Ready to hire

Hire with Assessment.

£7-10k flat fee. The methodology, delivered.

See Hire with Assessment
More reading

The methodology.

Four behaviours, role skills. Published in full.

Read the methodology