Recipes / Post-call content follow-up with a Highspot Digital Room

Post-call content follow-up with a Highspot Digital Room

Gong + Highspot

Who it's for

AEs who want to answer each open question and objection from recent calls with the right approved content, packaged in one Digital Room for the buyer. Gong can be swapped for Fireflies or Granola. No other sales enablement server is listed yet, so Highspot has no listed alternative.

What you get

Question-to-content table and a Digital Room waiting for approval. Sample output: every company and person in it is fictional.

Deal: Larkspur Freightworks (fictional) platform rollout. Gong summary of the last 3 calls with Dana Okafor and Sam Whitlock.

Open questions and objections (Gong summary, not quotes):
1. How long does a typical rollout take for a 200-seat team?
2. Concern: pricing feels high next to the tool they use today.
3. Does the product support single sign-on with their identity provider?
4. Who else in freight and logistics uses it?

Matched Highspot content:
1. 'Implementation timeline overview' (PDF, updated Aug 2026)
2. 'Total cost of ownership one-pager' (updated Sep 2026)
3. 'Security and SSO FAQ' (updated Jul 2026)
4. No match found. Ask marketing for a logistics reference, or answer this one yourself.

Digital Room (not created yet):
Name: Larkspur Freightworks: answers from our calls
Intro: "Dana, Sam, here are the materials for the three questions we left open on Tuesday."
Items: 1, 2, 3

Reply 'approve room', 'linked pitch instead', or tell me what to change.

Time and cost

Estimate About 10–15 minutes, most of it checking the content matches.

Assumptions: One deal with about three recent calls, one Gong question set and one Digital Room.

• Gong: One or two ask_deal questions for the deal. Each uses Gong credits based on how many calls and emails it analyzes. Read only. Every call uses credits. All three tools (ask_account, ask_deal, generate_brief) consume Gong credits based on the volume of calls and emails analyzed; briefs cost more per open-ended section. Same consumption as the equivalent API requests. Source: help.gong.io, checked 2026-10-16
• Highspot: A content search per question plus one recommendations request, then one Digital Room or linked pitch after you approve it. Credit use not confirmed by Highspot.

Exact credit costs depend on your plan with each vendor. Check the tool pages before a large run.

MCP servers

Every server is confirmed in: ClaudeChatGPT

• Gong Up · auth
  ClaudeChatGPTRead onlyEvery call uses creditsNeeds admin or IT
  https://mcp.gong.io/mcp
• Highspot Not probed
  ClaudeChatGPTRead + writeCredits not confirmed

Set up in a chat app

Every server in this recipe is confirmed for Claude and ChatGPT. Connect each one, then paste the prompts below into one conversation.

• Connect Gong without code (Claude, ChatGPT)
• Connect Highspot without code (Claude, ChatGPT)

Developer setup

{
  "mcpServers": {
    "gong": {
      "url": "https://mcp.gong.io/mcp"
    },
    "highspot": {
      "url": "None"
    }
  }
}

Step-by-step prompts

Copy all prompts

1. Use Gong to ask about my deal with {company}: what open questions and objections came up in the recent calls, and which calls did you use? Give me a numbered list in your own summary words, not quotes.
2. For each item, search Highspot for content that answers it, and also ask Highspot for content recommendations for this deal. Show me a table: question, best match, last updated, and one line on why it fits. Write 'no match' where nothing fits. Don't create anything yet.
3. Draft a Digital Room for {contact names} with a two-line intro and only the items I keep from the table. Show me the room name, intro and item list.
4. After I approve it, create the Digital Room in Highspot. If I say 'linked pitch instead', create a linked pitch with the same items only after I approve it. Give me the link, and don't share it with anyone.

Run it on a schedule (optional)

• This works best run by hand the day after a call, once Gong has processed it.
• Don't schedule the Digital Room step. A room is meant for the buyer, so create it only after you approve each item.
• Gong needs a tech admin to create the MCP integration first, and Highspot needs an Agentic platform license. Check both before your first run.

Where this goes wrong

• A question that wasn't really asked. Gong returns AI summaries, not raw transcripts, so wording can drift and a passing remark can read like an objection. Check each item against your own memory of the call or open the call in Gong, and delete anything you don't recognize.
• Missing objections from private calls. Gong's MCP excludes private calls. If a key call was marked private, its questions won't show up. Ask Gong which calls it used and compare that list to your calendar.
• Content that doesn't answer the question. A search can match on keywords and return a file about something nearby. For each match, ask for a one-line reason it answers the question, and open the file in Highspot before you approve the room.
• Spending Gong credits on repeat questions. Every ask_deal call uses credits based on the calls and emails analyzed. Ask for all questions and objections in one request, and reuse that answer instead of asking again for each item.

Gotchas

• Highspot doesn't publish its MCP endpoint or auth method; the setup docs are in its customer-only help center. Connect it from the Claude or ChatGPT app directory instead of adding a URL by hand.
• Highspot's MCP needs a Highspot Agentic platform license.
• A Gong tech admin must create an MCP server integration under Admin center > Settings > Ecosystem > MCP before anyone can connect. Whether it uses a personal or shared authorization context is fixed when it's created.
• Gong's MCP is read only and doesn't return raw transcripts, so you get summaries, never word-for-word quotes.

Download

Get the prompts, guardrails and expected output as one Markdown file. Paste it into a Claude Project or ChatGPT project's instructions, or save it as CLAUDE.md for Claude Code.

Download post-call-content-room.md