Find patterns across 5–10 call transcripts: objections, drop-offs, momentum
Analyzes a batch of transcripts as a dataset instead of one-off reviews: objections recurring across calls, the exact points where prospect energy dies, the moments that reliably create engagement, and gaps between what you pitch and what prospects care about. Ends with three evidence-backed changes your team can start Monday.
You are a revenue analyst who finds the signal individual call reviews miss. One call tells you about a rep and a deal; ten calls tell you about your market, your message, and your process. Your specialty is separating real patterns from coincidence — you never call two occurrences a trend. I'm going to paste 5–10 call transcripts, each labeled. Analyze them AS A SET: 1. RECURRING OBJECTIONS — objections appearing in 3+ calls, with one verbatim example each and a count. Note whether the handling was consistent across reps or all over the place. 2. DROP-OFF POINTS — where in calls does energy die? Look for: the topic after which prospect responses get shorter, the recurring question that stalls things, the segment where interjections stop. Cite calls by label. 3. MOMENTUM MOMENTS — what reliably generates engagement across calls: a phrase, a story, a demo moment, a question. These are your best assets; name them precisely. 4. MESSAGE-MARKET GAPS — things prospects consistently care about that we consistently talk past, and things we consistently pitch that consistently land flat. 5. PROCESS FINDINGS — patterns in call mechanics: next-step quality, stakeholder presence, call length versus outcome, discovery depth before demo. End with THE THREE CHANGES: the three highest-leverage adjustments this set of calls justifies, each tied to its evidence, each phrased as something a team could start doing Monday. Rules: every pattern needs 3+ occurrences and cited examples — flag anything weaker as 'worth watching, not yet a pattern'. If the calls are too different to compare (mixed stages, mixed products), say so and segment them. Before you analyze anything, interview me. Ask me these questions ONE AT A TIME, waiting for my answer each time: 1. Paste the transcripts, labeled (Call 1, Call 2... with outcome if known: advanced, stalled, lost, won). 2. What do these calls have in common — same rep, same stage, same segment, same campaign? 3. What prompted this analysis — what are you hoping to find or afraid is true? Once you have my answers, produce the analysis.
How to use it
- 1
Copy the prompt into Claude or Gemini — 5–10 transcripts demands a large context window.
- 2
Pick a coherent set: same stage, same rep, or same campaign. Mixed sets produce mush.
- 3
Label each transcript with its outcome — pattern-versus-outcome is where the best findings live.
- 4
Answer question 3 honestly; naming your fear lets the model confirm or kill it with evidence.
- 5
Take the three changes to your next team meeting with the cited examples attached.
Best practices
Run monthly on the same slice (e.g., all first discovery calls) — the month-over-month delta shows whether coaching is working.
The drop-off section is the sleeper hit: teams instrument their emails obsessively and have no idea where their calls lose people.
Insist on the 3+ occurrence rule when reading output; if the model calls two mentions a pattern, push back.
Compare won-deal calls against lost-deal calls as two separate runs, then diff the momentum moments.
Example: what this looks like in practice
A head of sales at a 40-person compliance platform batches eight stalled-deal discovery transcripts, afraid the problem is pricing. The analysis says otherwise: pricing objections appear in only two calls, but in six of eight, prospect responses shorten right after the security-questionnaire discussion — reps consistently answer 'we'll handle that later in the process,' and energy never recovers. Momentum moment, found in five calls: a specific customer story about a failed audit. The three changes: address security proactively in call one, move the audit story earlier, and stop leading with the integrations slide that landed flat in seven of eight calls. Next month's stall rate on new discovery calls drops visibly.
Best fit
This prompt is one gear in a bigger machine. We orchestrate 20+ tools into outbound systems our clients own — and guarantee the results.
Apply for a Pilot Spot → →Frequently asked questions
Five is the useful minimum, eight to ten is better. Below five, everything looks like a pattern and nothing is provable; this prompt enforces a 3+ occurrence rule precisely so two coincidences don't become a strategy change. Keep the set coherent — same call stage or same rep — or segment before analyzing.
More call transcript analysis & coaching prompts
Extract every objection from a call transcript and grade how it was handled
Pulls every objection out of a transcript — including the soft, disguised ones reps miss — with verbatim quotes, a diagnosis of the real concern, a letter grade on the rep's response, and a better line for next time. It's a full objection review in two minutes instead of re-listening to the call.
Score a sales call against MEDDIC or SPICED with a coaching scorecard
Turns any transcript into a framework scorecard with a 0–3 score per element, verbatim evidence behind every score, and the exact question that would close each gap. Stage-aware, so a first discovery call isn't punished for missing decision criteria. It's the deal-review prep that used to take a manager 45 minutes per call.
Analyze talk ratio and question quality from a call transcript
Measures what conversation intelligence tools measure — talk ratio, question count — but adds the layer they skip: classifying every question by type and quality, and pinpointing the exact moments a rep should have dug deeper instead of moving on. Works from any transcript, no Gong seat required.