Back to use cases
Sales

Recurring objections across mid-market deals

The same pushback, said three different ways, by ten different prospects this quarter.

Best for
Sales enablement, messaging
Primitives
DecisionsRationaleSentiment
Token savings
75%

The prompt

You're identifying patterns in why mid-market deals stall. Using Sentra: 1. Pull entities around mid-market deals and stage in {evaluation, procurement, legal}. 2. Filter to Interactions where Sentra extracted a Decision or Rationale with polarity = "concern" or facet = "objection". 3. Cluster the extracted objections by theme using whatever clustering the Semantic Agent exposes — pricing structure, security review, integration depth, time-to-value, vendor consolidation, etc. 4. For each cluster, give the count, the three most-quoted phrasings, and the three accounts where it was loudest. Output: - Top 5 objection themes, ranked by frequency. - For each: one sentence on what it actually means, three quoted phrasings with sources, and a suggested response we've used successfully in another account (cite the Interaction).

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: sales enablement, messaging

Sentraover MCPMid-market deals · last 90 days

One objection, three costumes

"We already have this in our BI tool" · 6 prospects

calls · 6 sources

Always raised by the data lead, never by the buyer. "Our dashboards tell us what happened; I do not see the delta" is the canonical form.

"What happens to our data if we churn" · 4 prospects

security reviews

Surfaces only in security review, never earlier. Two deals stalled a week each because the answer was ad hoc.

"Rollout will eat a quarter" · 3 prospects

calls

Appears after pricing, never before. It is a value question wearing an ops costume.

What kills it

The BI objection died in every deal where we showed a meeting-to-decision trace live.

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 mid-market deal in evaluation, procurement, or legal stage from the CRM. Iterate over each.
  2. 02For each deal, pull all call transcripts, emails, and Slack threads. Read each for objection-shaped moments.
  3. 03Tag each objection with a theme (pricing, security, integration, etc.) by re-reading and judging.
  4. 04Aggregate counts across deals. Identify recurring themes manually.
  5. 05For each theme, pick the three most-quoted phrasings — re-skim source material to extract verbatim.
  6. 06Look for prior accounts where the same objection was successfully countered — search for our follow-up language.
  7. 07Compose the top-5 ranking with citations.

~240,000 tokens

With Sentra

  1. 01Filter `Opportunities` where segment = mid-market and stage ∈ {evaluation, procurement, legal}, last quarter.
  2. 02Pull all `Interactions` for those opportunities where Sentra extracted a `Decision` or `Rationale` with polarity = concern or facet = objection.
  3. 03Cluster extracted objections by `Theme` — already learned by Sentra's Semantic Agent.
  4. 04For each theme: count, top-three phrasings, three accounts where loudest.
  5. 05Pull successful counter-objection patterns from prior `Decisions` in won deals.
  6. 06Compose the ranking.

~60,000 tokens

Book a DemoStart Free