Back to use cases
Marketing

Language our best customers actually use

Phrases your customers used to describe your product: ranked, dated, sourced.

Best for
Positioning, web copy, ads
Primitives
InteractionsThemesQuotes

The prompt

You are building a language library from customer Interactions. Using Sentra: 1. Pull entities around customer-facing conversations and sentiment about us (sales calls, CS calls, support, community, etc). 2. Extract any sentence where the customer described our product, our category, or the problem we solve — in their own words. 3. Cluster phrasings by theme (the pain we solve, what we replaced, the moment of "click", the metric they care about). 4. Rank phrasings within each cluster by recency × frequency × seniority of the speaker. Output: - A theme-organized library: per theme, the top 5 phrasings, each with the quoted sentence, the account, the speaker's title, and the date. - A short closing note: which phrasings have grown vs. faded vs. only appeared in the last 60 days (a possible new emerging story).

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: positioning, web copy, ads

Sentraover MCPTheir words, dated and sourced, ranked by reuse potential

How customers describe us

"The only place the why survives"

call · Aug 29 · 00:12:18

Brightline, onboarding call. Their ops lead, unprompted, describing decision traceability.

"Project plan versus project reality"

QBR · Jul 17

Meridian QBR, describing the delta view. Already tested well in one deck.

"Our new hires stopped asking archaeology questions"

call · Jun 30

Harbor, casual aside in a check-in. Strongest onboarding line in the library.

Clearance

All three flagged for approval before public use; none are cleared yet.

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

Without Sentra

Not reachable without persistent memory: the context this recipe needs was never written down anywhere an agent could retrieve it.

With Sentra

  1. 01Pull `Interactions` tagged surface = customer-facing across the last 12 months.
  2. 02Extract sentences where the customer described our product, category, or problem — already extracted as `Theme`-tagged quotes.
  3. 03Cluster by theme: pain solved, what we replaced, the click moment, the metric they care about.
  4. 04Rank phrasings within each cluster by recency × frequency × speaker seniority.
  5. 05Identify phrasings that have grown vs. faded vs. emerged in the last 60 days.
  6. 06Compose the theme-organized library.

~65,000 tokens

Book a DemoStart Free