Getting Started

A plain guide to the app

This guide walks you through Score one piece at a time — what it is, when to use it, and what every button does. Read it start to finish, or jump to the part you need. Nothing here is urgent or complicated; if you ever feel stuck on the app, this page is meant to get you unstuck.

What it is

Score reads the transcript of a sales or customer call and gives it a score. The score answers one question: how much of the standard the manager set did this call actually meet?

It looks at eleven specific selling skills — things like how the rep opened the call, whether they uncovered the real problem, how they handled pushback, and how they closed. For each one it gives an actual score, compares it to the level you expect, and rolls everything up into a single number with a short written explanation and a list of next steps.

It does not record calls or listen live. You give it a transcript that already exists, and it gives you a read on that one call.

Why we use it

Reviewing calls by hand is slow, and two people often grade the same call differently. Score applies the same standard to every call, so a score means the same thing across reps, managers, and weeks. That makes coaching fairer and the patterns easier to see.

It also makes the good stuff repeatable. When you can point to exactly which skills a call hit or missed — with the lines from the transcript behind each one — you can teach the rest of the team to do the same thing on purpose.

When to use it

Reach for it whenever you have a transcript and want a quick, even-handed read. Typically:

• Right after an important call, while it's still fresh.
• Before a one-to-one, to decide what to praise and what to work on.
• When you want to compare how a skill is landing across several calls.
• When a rep wants honest feedback on their own call before their manager sees it.

Who it helps

The screen at a glance

The main working area is the Skill-Based Diagnostic — reached from the left menu, and where you'll spend almost all your time. From top to bottom you'll see:

• Top menu — Quick Help (a short in-app cheat sheet), Getting Started (this guide), FAQ, and Project Moneyball.
• Left menu — switches between pages: the Skill-Based Diagnostic (scoring, the page described here), the Cross-Skill Diagnostic (how the skills lift and drag each other), Anonymize Transcript (an optional privacy tool), and Sign out at the bottom. It collapses to an icon strip with the chevron at its top. (A few further entries are on the roadmap.)
• Toolbar — the main action buttons: Import transcript, Run, Stop, Download PDF, and Clear.
• Message Center & Runtime Monitor — the box on the left reports progress and warnings; the grid on the right lights up each skill as it's scored. The grid opens with a Sanity Check step (see below).
• Call Details — the rep, role, business, call type, and call duration, filled in for you after import.
• Lifecycle Stage — the customer journey shown as a row of stages; the one identified for this call glows gold.
• Final Score — the big number, its meaning, and how much of the call had evidence to score.
• Performance table — the actual result next to the expected minimum and weights you can adjust.
• Score Analysis & Recommendations — what worked, what didn't, and a single combined list of next steps.

Step by step

1. Import a transcript

   Click Import transcript and pick a file, or drag the file onto the notification box. Plain text works best (.txt, .vtt, .srt, .md, .log, .csv). The app reads the call, works out the call type, and loads a matching scorecard.

2. Check the Call Details

   The rep's name, role, business, and call type fill in automatically. These are read-only — they come from the transcript, so there's nothing to type. On calls with more than one rep on your side, the name and role shown are the call owner — the rep who actually drove the conversation and was scored, not whoever opened the call.

   Rep's Role is shown only when the call gives enough evidence to determine it reliably. If it can't be determined, the role is left out (rather than guessed) and a note appears in the Message Center.

3. Review the expected minimum and weights

   In the performance table, each skill has an expected minimum (the bar you set, 1 to 5) and a weight (how much it counts). The defaults are already sensible. Adjust them if this call should be held to a different standard. See Expected minimum & weights for what each setting does.

4. Run

   Click Run. The Runtime Monitor opens with a quick Sanity Check — a brief pre-flight pass that confirms the transcript looks like a real call before scoring begins — then lights up each skill as it's scored. The first run on a new transcript takes a couple of minutes; after that, re-running is quick and free — until you change a setting, which scores again.

5. Read the score

   The big circle shows the final number, with its meaning underneath (for example "Below expected") and how much of the call had evidence to score. Green means the call met or beat the standard, amber means it fell a little short, red means a serious gap. See Reading the score.

6. Read the analysis and next steps

   Score Analysis lists what worked and what needs improvement. Recommendations is a single, de-duplicated list of next steps. Use the Copy button on either box to paste it into your notes or an email.

7. Download the report

   Click Download PDF for a clean, one-page branded report you can save or share.

Every button explained

Top menu

Toolbar

Performance table

Expected minimum & weights

These are the two dials you control for each skill. Together they tell the app what "good" means for this kind of call.

Expected minimum (the bar)

Click the segments to set the level you expect for a skill, from 1 to 5. The score for that skill is measured against this bar: meet it and the skill is on target, fall below it and it counts against the call.

Set the bar to 0 to make a skill optional. An optional skill is never required — if the rep does it, it adds a small bonus; if they don't, there's no penalty. Use this for skills that are nice-to-have on a given call type but not expected.

Weight (how much it counts)

Critical does one extra thing: it acts as a safety gate. If a critical skill badly misses its bar, the whole call's score is capped — so a strong call can't paper over a critical failure. More on that below.

Reading the score

The final score is built around 100. 100 means the call met exactly the standard you set. Above 100 means it beat the standard; below means it fell short. The scale runs up to 120.

Evidence coverage & confidence

Under the score you'll see something like "72% – Medium Confidence". Coverage is how much of the scorecard the call actually gave evidence to score — not every skill comes up on every call. Confidence follows from coverage: High at 85%+, Medium from 60–84%, Low below 60%. A lower coverage doesn't mean a bad call; it means there was less to go on, so read the score with that in mind.

When a score is capped

The app holds a score back in a few cases, so a high average can't hide a real problem:

• A critical skill far below its bar — the score is capped at 74.
• Two or more critical skills below target — capped at 79.
• Low coverage (under 60%) — capped at 79 and confidence set to Low.

A capped score shows in red as a flag that something needs a closer look.

The eleven skills

Every call is scored on these eleven. Each gets its own actual score, compared to your expected minimum.

More on Next Steps — the closing wrap-up. Before agreeing next steps, a strong rep wraps up the call: they play back, in their own words, what they understood — "let me see if I got this right — you said…, you mentioned…, you clearly said you cannot accomplish…" — and invite the buyer to confirm or add anything. This repetition is deliberate: it catches misunderstanding and gives the buyer a last chance to surface more before the conversation moves on, so the rep advances with full ammunition and zero confusion — no money left on the table. The skill scores both this wrap-up and the quality of the next step itself, so a call that jumps straight to scheduling without the recap is capped even if the meeting gets booked. (Distinct from Call Prep, which recaps previous meetings at the start, and Active Listening, which is reflection throughout the call.)

The Cross-Skill Diagnostic

The eleven skills don't work in isolation — they lift and drag each other. Open the Cross-Skill Diagnostic from the left menu (after you've scored a call) to read the connected-skills story: why each skill matters, how one skill feeds the next, and what a call loses when a foundational skill is missing.

On a strong call it narrates how the foundation paid off — setting expectations and preparation feeding Problem Development, the gap, and a genuine A-Ha Moment. On a weaker call it does something more useful than a list of low scores: it traces the result back to a single root cause — the upstream skill whose weakness rippled downstream — and names the one place to start, so the whole chain improves with it. It's the difference between "these five skills were low" and "fix this one thing first, and here's why it dragged the rest."

It reads the scores from your most recent run, so score a call first and the page fills in automatically — there's nothing to configure.

MEDDPICC qualification scan

Alongside the score, Score runs a separate MEDDPICC qualification scan on the transcript. MEDDPICC is the industry-standard checklist for how well an opportunity is qualified. The scan reads the call and reports, for each of the eight elements, whether the call showed strong evidence (buyer-validated, quantified, named, or mapped), weak evidence (merely mentioned or assumed), or no evidence at all. Buyer-stated evidence always counts for more than anything the rep merely asserts.

Templates

A template is a saved set of expected minimums and weights. When you import a call, the app loads a template that fits the call type, so you rarely start from scratch.

• Save template — once you've set the bars and weights the way your team wants, save them under a name so everyone can reuse the same standard.
• Open template — load a built-in template or one you saved earlier.
• Reset — go back to the defaults for this call type if you've changed things and want to start over.

Finding what your top performers do differently

Score isn't only for grading one call — run it across a set of calls and it becomes a way to find out what your best reps actually do that others don't, so you can teach it on purpose instead of hoping it spreads.

Before you start, decide the lens you want to diagnose by: product line, segment, rep, or whatever comparison actually matters to your team. The steps below assume you're comparing top performers against reps who are struggling to hit their goals, but the same approach works for any two groups you want to compare.

1. Set the bar once

   Use the built-in preferences — they adapt automatically to the identified call type — or define your own desired state for each skill and save it as a template. Either way, score every call in the comparison against the same standard, so the results actually line up against each other.

2. Score your top performers

   Import their call transcripts one at a time and review the Actual Observed Performance for each. Download a PDF report for every call so you keep a record to compare later.

3. Score the comparison group

   Do the same for the reps you're comparing them against — for example, the reps who are struggling to meet their goals.

4. Compare and act

   Line up the two sets of results. The gap between them shows you exactly which skills separate your best reps from the rest — evidence you can turn straight into team coaching, one-to-one coaching, or an enablement playbook.

Anonymize your transcript

Score never stores your transcript — every call is processed in memory for a single run and then discarded, and only the derived results are kept briefly (full details in the Trust Center). For teams under strict privacy, legal, or compliance rules that would rather confidential details never leave their own machine at all, the Anonymize Transcript page (left menu) offers a self-serve prompt that does exactly that. It's an optional extra layer — playing it safe by choice, not patching a gap.

Copy the prompt, paste it into any AI assistant — ChatGPT, Claude, Gemini, or a local/offline model for maximum privacy — add your raw transcript, and it hands back a synthetic copy: real company, people, and product names, the buyer's quoted pains, and commercial figures are all swapped for consistent stand-ins, while the call's structure, timing, and every behavioral gap are preserved — so the synthetic version scores the same as the original. Review it, save it, and import that copy on the Skill-Based Diagnostic page.

If something looks off

A few things you might run into, and what they mean:

Still stuck? The notification box at the top of the app explains what's happening at each step — read the latest message there first. For anything else, reach out using the contact below.