Back to use cases
Marketing

Draft a case study from the actual interactions

A customer story composed from what they said: not what we wished they'd said.

Best for
Case study production
Primitives
AccountsInteractionsOutcomes
Token savings
60%

The prompt

You are drafting a case study for {{account_name}}. Using Sentra: 1. Pull every customer-facing Interaction with {{account_name}} from before they were a customer through the present. 2. Extract: the original problem in their words, the alternatives they considered, the moment they decided to adopt, the rollout arc, and the measurable outcomes they've shared. 3. Pull the four most quotable moments, with speaker, title, and date. 4. Flag every quote with a consent status — Sentra should surface whether the speaker's permission is on file. Output: - A two-page case study draft with the standard structure (problem, decision, deployment, outcome). - Inline citations for every claim and every quote. - A separate sign-off checklist: which quotes still need explicit customer consent before publication.

See it work

Run it from

ClaudeCursorChatGPTAny agent · MCP

How it runs

  1. 01Copy the prompt into any agent connected to Sentra over MCP.
  2. 02The agent reads the company memory, scoped to what you are allowed to see.
  3. 03One grounded answer comes back, with every line traceable to its source.

Or automate it with Actions

The same prompt runs as a Sentra Action: on a schedule, or on a semantic trigger. Output lands in Slack or email, and every send waits for your approval.

Trigger idea: case study production

Sentraover MCPComposed from what they actually said

Brightline story · drafted from the record

Before

kickoff · May 12

"Every launch, someone was working from last month's plan." Their words from the kickoff, not ours.

The moment

thread · Jun 4

The launch-date conflict caught across four rooms in week three, resolved before it reached the customer.

Result

QBR · Aug 29

Their own QBR numbers: prep time for the weekly review down from a day to under an hour.

Pull quotes

2 sources

Two dated candidates flagged, both pending customer approval before publication.

The prompt above becomes this. Fictional data, real mechanics: every line cites the meeting, thread, or record it came from.

Without Sentra

  1. 01Pull every customer-facing interaction with the account from pre-customer through present.
  2. 02Read each to identify the original problem in their words, the alternatives they considered, the moment of decision, the rollout arc, and measurable outcomes.
  3. 03Pull the four most quotable moments — re-skim transcripts to confirm verbatim, speaker, date.
  4. 04Check consent status for each quote by searching prior conversations for explicit permission.
  5. 05Compose the two-page case study with inline citations.

~110,000 tokens

With Sentra

  1. 01Pull every customer-facing `Interaction` with the `Account` from pre-customer through present.
  2. 02Extract: original problem, alternatives considered, the decision moment, rollout `Decisions`, outcome `Commitments`.
  3. 03Pull four most quotable moments with `Actor` speaker, title, and date.
  4. 04Each quote has typed `consent` status — Sentra surfaces what needs sign-off.
  5. 05Compose the case study with citations.

~44,000 tokens

Book a DemoStart Free