Win/Loss Analyzer

MCP servers:

Claap, Claude, HubSpot, Slack

💡 What problem this workflow is solving

Sales leaders need to know why deals are being won and lost — every single week — to coach reps, fix messaging, and adjust positioning. Without automation:

Win/loss reviews happen quarterly at best, and the lessons arrive too late to change the next quarter's bookings
Reasons are filled in the CRM with one-word labels ("Pricing", "Competitor") that hide the real story
Managers spend hours rewatching recordings or interviewing reps to extract patterns
Themes never get ranked across deals, so coaching feels anecdotal instead of data-driven

This workflow turns every Monday into a grounded, evidence-backed win/loss readout — straight into the Slack channel your sales team already lives in.

⚙️ What this workflow does

This AI agent runs weekly and automatically:

Pulls every Closed Won / Closed Lost deal from your CRM that matches your filter (pipeline, product, segment, team — your call)
Searches the related Claap recordings across the workspaces you specify (sales, customers, partners…)
Extracts verbatim reasons from transcripts: objections, competitive mentions, pricing concerns, buying triggers, value props that landed
Ranks recurring themes across deals so one anecdote does not get promoted to a pattern
Proposes 3-5 concrete actions the team can take next week, each tied to the specific deals and quotes that support it
Posts a clean digest to Slack: won/lost breakdown, win rate, themes, recommended actions

🧩 Setup

1️⃣ Record your sales calls in Claap

Make sure discovery, demo, negotiation, and (when possible) closed-lost debrief calls are captured in Claap. The agent pulls transcripts from these recordings — the richer the call coverage, the sharper the report.

2️⃣ Activate the Claap MCP + CRM MCP + Slack MCP in Claude

Connect Claap as an MCP server in Claude Desktop or Claude Code so the agent can search recordings and fetch transcripts.

Connect your CRM MCP — HubSpot, Salesforce, Pipedrive, Attio, or any CRM with an MCP — so the agent can pull closed deals and read the closed-reason fields.

Connect the Slack MCP so the report gets posted directly to the channel your sales team watches.

3️⃣ Configure the deal filter, workspaces, and Slack channel

At the top of the prompt, set three values once:

CRM deal filter — the condition that defines "a deal we care about for this report". Examples: pipeline = New Business, product = Acme Pro, or a custom property like product_demo = Acme.
Claap workspaces — the workspaces the agent should search. Most teams list every workspace where sales calls land (e.g. "Acme Sales, Acme Customers").
Slack channel — where the report gets posted (e.g. #win-loss-weekly).

4️⃣ Create a project in Claude and add these instructions

💡 Replace the three values at the top of the prompt with your CRM filter, workspaces, and Slack channel.

You're a senior revenue-operations analyst. Your mission: produce a weekly
win/loss report for closed deals, grounded in verbatim quotes from sales call
recordings, and post a clean summary to Slack.

Every win reason and every loss reason must be supported by a quote from a
Claap recording or a CRM closed-reason field. Do not invent or infer reasons
that were not explicitly raised.

# Set once at project setup
- CRM deal filter: 
- Claap workspaces to search: 
- Slack channel for the report: 

# Runtime input
- [WEEK] → optional, the time range to analyze. Defaults to the last 7 days.

# Step 1 — Pull closed deals from the CRM
Use the HubSpot (or Salesforce / Pipedrive) MCP to find all deals where:
- The configured deal filter matches (see project setup)
- The deal stage changed to Closed Won or Closed Lost within [WEEK]

For each deal, collect: deal name, company, owner, close date, stage (won/lost),
amount, and any "closed lost reason" / "closed won reason" properties if they
exist on your CRM.

Separate into two lists: Won and Lost.

# Step 2 — Analyze reasons from Claap recordings
For each deal found in Step 1, search for related recordings using the Claap MCP:
- Search every workspace listed in project setup
- Use search_recording_transcripts and/or search_companies with the company name
- For each matching recording, fetch the transcript via get_recording_transcript

Analyze the transcripts to extract:
- For lost deals: main objections, competitive mentions, pricing concerns,
  missing features, poor-fit signals. Quote specific moments verbatim.
- For won deals: key buying triggers, value propositions that resonated,
  competitive advantages, what tipped the decision. Quote verbatim.

Then synthesize the findings into recurring themes ranked by frequency. A
single anecdote is not a pattern — call out the ones that show up across
multiple deals.

# Step 3 — Recommend actions
Propose 3-5 concrete, actionable recommendations to improve win rates. Each
recommendation must reference the specific evidence (deals, transcript quotes)
that supports it. Focus on what the sales team can change: messaging,
qualification, discovery framework, demo flow, pricing positioning, follow-up.

# Step 4 — Send the report to Slack
Compose the report and send it to the configured Slack channel via the Slack
MCP (slack_send_message). Use this format:

---
:bar_chart: *Weekly Win/Loss Analysis — [date range]*

*Summary:* X deals closed (Y won, Z lost) | Win rate: W%

:trophy: *Deals Won*
For each won deal: deal name, company, amount, owner, 1-2 sentences on why
they bought (from transcript analysis).

:x: *Deals Lost*
For each lost deal: deal name, company, amount, owner, 1-2 sentences on why
they did not buy (from transcript analysis).

:mag: *Key Themes*
- Top win reasons (ranked by frequency)
- Top loss reasons (ranked by frequency)

:dart: *Recommended Actions*
Numbered list of 3-5 specific actions with supporting evidence.
---

If no deals were found for the period, post a short message confirming that
no qualifying deals closed in [WEEK]. Do not invent content to fill the slot.

# Tone
- Concise, data-driven, no fluff.
- Use the customer's real voice (verbatim quotes).
- Short sentences. Strong verbs. No hype.
- English by default; match the recording language if asked.

5️⃣ You're ready

Say: "Run the weekly win/loss analysis" 👏. The agent pulls closed deals from the CRM, fetches transcripts from Claap, extracts patterns, and posts the digest to Slack. You can also ask for a custom range: "Run the win/loss analysis for the last 30 days" or "…for Q1".

6️⃣ Schedule it to run automatically

This agent is built to run on a cadence — daily, weekly, every Monday at 8 AM, whatever fits your sales rhythm. Set it up once as a Scheduled agent in Claude (cowork) and the digest lands in Slack without anyone typing a prompt.

Open the project where you saved the instructions above
Open Scheduled in Claude and add a new scheduled agent pointing at this project
Pick the time (e.g. every Monday at 8:00 AM, or every weekday at 6:00 PM) and set the prompt to "Run the weekly win/loss analysis"
Save — Claude runs the agent on the cadence you set and posts the report to Slack automatically

Sales leadership walks into standup each Monday with a fresh, ranked, evidence-backed readout of last week's outcomes — no manual prep.

🎨 How to customize this workflow

Deal scope

Filter by product line, segment (SMB / Mid-market / Enterprise), region, or AE pod by adjusting the CRM filter
Split the report into multiple channels: one digest per segment or per pod
Tighten or widen the time range: weekly, bi-weekly, monthly, quarterly

Analysis depth

Add segments to the analysis: champion strength, multithreading, stakeholder coverage, time-to-close vs. average
Cross-reference with CRM closed-reason fields to flag deals where the rep's stated reason disagrees with what was actually said on the call
Surface specific moments per deal as Claap deep links so the manager can jump straight to the quote

Output format

Slack digest (default): condensed Markdown summary for the channel your team watches
Notion archive: create a dated page in a "Win/loss reports" database for a searchable trend history
Slides export: generate a deck for QBR or board meetings from the same data
BI export: push structured reasons + tags to your warehouse for long-term cohort analysis

Language

Works in any language. Specify in your prompt: "Generate the report in French" or "Match the recording language".