7-Stage Call Framework
How Aura's AI scores your sales calls — the seven things it grades, plus the bigger-picture scores it produces
7-Stage Call Framework
After every recorded call, Aura's AI watches the meeting and grades it. It doesn't just give you a single score out of 100 — it scores the call across seven specific stages of a sales conversation, then rolls those up into a few high-level numbers you can track over time.
This is the framework behind the "Comprehensive" analysis. (There's also a quicker "Outcome" analysis that just flags won/lost/no-show — that one doesn't do stage scoring.)
You'll occasionally see this called the "Apex Framework" in older copy — it's based on the Apex Sales methodology. The product just calls it the 7-stage framework now.
The seven stages
The AI grades each stage on a scale of 1 to 10. Higher is better.
| Stage | What it's looking at |
|---|---|
| Rapport | How well you set the tone at the start. Did you establish authority, agree on an agenda, make the prospect comfortable? |
| Discovery | Did you understand the prospect's situation? The AI looks for four layers: surface symptoms, underlying pain, dollar cost of the problem, and emotional weight. |
| Gap | Did you make the difference between "where they are now" and "where they want to be" crystal clear? A blurry gap leads to a weak pitch. |
| Financial Qualification | Did you confirm budget, decision-making authority, and that they can actually afford the offer — before the pitch? |
| Pitch | Did your solution presentation tie directly back to the gap you uncovered? Or did it feel generic? |
| Objections | Did you anticipate and resolve concerns, or get caught off guard? |
| Close | Did you make a clear ask, handle the price reveal cleanly, and lock in a next step? |
These names are the canonical ones the product uses. Older docs sometimes used words like "qualification," "presentation," or "next steps" — those aren't part of the framework anymore.
The big-picture scores
Beyond per-stage grades, you'll see five summary scores on every call:
| Score | What it means | Scale |
|---|---|---|
| Structure Score | The seven stage scores added up. Higher = the call followed the framework well. | 7 to 70 |
| Emotional Score | A separate read on how emotionally engaged the buyer was. Closing without emotion is rare. | 0 to 10 |
| Closing Score | A focused score on the pitch + objections + price reveal stretch — the part of the call where deals are made or lost. | 0 to 10 |
| Buyer Certainty Index | How ready the buyer was when you revealed the price. Low certainty + price reveal = lost deal. | 0 to 10 |
| Overall Score | The headline number on the call detail page. Combines everything into one read. | 0 to 100 |
The Structure Score is just a sum — every stage counts equally. There's no "Discovery is worth more than Rapport" weighting. The other scores capture things a raw sum would miss, like emotional engagement and timing.
Discovery in extra detail
Discovery is the one stage with structure inside it. The AI looks at four layers:
| Layer | What it captures |
|---|---|
| Symptoms | The surface problems your prospect mentioned ("our team is slammed") |
| Pain | The personal or business cost of those problems ("which means we're missing deadlines") |
| Cost | The actual dollar impact ("which costs us about $50k a quarter in lost contracts") |
| Emotional | The emotional weight ("I'm exhausted and frustrated and worried I'm going to lose my best people") |
Plus two extra dimensions:
- Emotional depth (1-10) — did discovery actually get into emotional territory?
- Logical clarity (1-10) — can the prospect articulate their situation clearly?
To score high on Discovery, you can't just collect symptoms. You need to land on all four layers.
Pass/fail checks
Three yes/no boxes the AI also ticks:
| Check | Question it answers |
|---|---|
| Need acknowledged | Did the prospect actually say "I need help with this," in their own words? |
| Gap is big enough | Was the problem you uncovered material enough to justify pitching your solution? |
| Temperature check before price | Before you said the number, did you check if they were ready to hear it? |
These show up as little check marks on the call detail page. They're not part of the numerical score — they're flags.
Coaching, not just grading
The comprehensive analysis isn't just numbers. The AI also produces a fixed coaching package for every call:
- 3 missed questions — questions a great closer would have asked but you didn't
- 5 line rewrites — specific things you said with a stronger alternative the AI suggests instead
- 3 action steps — what to practice for the next call
The counts are always exactly that — three, five, three. The AI is forced to stick to those numbers, so the coaching is consistent across calls and you don't get a wall of suggestions on one call and nothing on the next.
Where you'll see the scores
| In the app | What's shown |
|---|---|
| Call detail page | Every stage score, all five summary scores, the discovery sub-layers, the three checks, the full coaching package |
| Team performance dashboard | Per-rep averages over time — Structure Score, Overall Score, and per-stage trends |
| Analytics → Performance tab | Rollups across cohorts (e.g., "what's our average Objections score on the new product launch") |
| Partner API | Same data, available via API for anyone integrating their own dashboards |
How the AI runs
Two flavors run on each call:
| Type | What you get | When |
|---|---|---|
| Outcome | Quick won/lost/no-show classification, plus confidence | A few seconds after the call ends |
| Comprehensive | The full 7-stage analysis above | A few minutes after the call ends |
You can re-run either one anytime — useful if you tweak your business profile or want a fresh read.
Related
- AI Analysis — what happens behind the scenes
- Call Lifecycle — where in the lifecycle each analysis runs
- Business Profile — the context the AI uses to tailor its grading to your business