HG Insights|Phoenix

QBR Outcome Tracker

Did the last QBR's commitments land? One verdict, with the signal deltas to prove it.

QBR Outcome Tracker
Sample preview

Sample output for this workflow will appear here once it is captured.

Run the workflow in Claude, ChatGPT, or Phoenix Playground using the buttons below to see real output.

Overview

Pairs with `qbr-pre-read` — that's forward-looking (next QBR agenda). This is backward-looking: given the previous QBR notes, did the commitments land? Reads firmographic, technographic, and intent deltas since the last QBR, contrasts against what was promised, and surfaces a one-line 'is this customer trending the right direction' verdict. Bake into the QBR-prep checklist.

Use cases

  • QBR rituals that actually compound

    Priya runs CSM ops at a security tooling vendor. Every QBR makes commitments; six months later nobody remembers them. She bakes this workflow into the QBR-prep checklist: run it 2 weeks before each QBR with the prior notes as input, surface what landed and what slipped, walk in with credibility instead of vibes. Pairs with `qbr-pre-read`.

  • Verdict you can defend in 30 seconds

    'On Track' isn't 'I feel like it went OK'. It's '3 of 4 commitments visibly landed in HG technographic + intent data, here are the dates and the citations'. Your VP CS gets a portfolio view that's grounded, not gut-feel.

View workflow prompt
# QBR Outcome Tracker

## Parameters

- `{{domain}}` *(required)* — Customer company domain. Example: `acme.com`
- `{{your_product}}` *(required)* — Your product or category. Example: `HG Insights`
- `{{prior_qbr_notes}}` *(required)* — Paste of the previous QBR's commitments / agenda / action items. Example: `Q3 QBR: committed to deploy HG Intent in Sales; named John Doe as champion; flagged renewal in Q1.`

## Purpose
Pair with `qbr-pre-read` (forward-looking next-QBR agenda). This workflow is backward-looking: given {{prior_qbr_notes}}, did the commitments at {{domain}} actually land for {{your_product}}? Output is a one-line verdict + the signal deltas that prove it.

## Process
1. **Parse commitments** — extract from {{prior_qbr_notes}}: what was promised, what was flagged as risk, what action items had owners. Surface 2-4 specific commitments.
2. **Signal deltas** — for each commitment, look up the relevant signal:
   - "Deploy product X by Q2" → `company_technographic` for {{domain}}: did product X show up after the QBR date with a recent `firstVerifiedDate`?
   - "Expand to function Y" → cross-check via FAI-style reasoning (workflow doesn't have direct FAI tool here — note as 'pending direct check').
   - "Maintain intent on category Z" → resolve Z via `list_intent_topics`, call `company_intent` with `limit: 200`, filter `topics[]` for Z, check `trend` (positive = held, negative = drifted).
   - "Manage CFO transition risk" → `company_firmographic` for leadership change. `sec_filing_section` for any 10-K Risk Factor flagging the same.
3. **Context check** — `web_search` for {{domain}} since the QBR date: any major event (M&A, RIF, earnings miss) that explains why a commitment didn't land?
4. **Verdict** — one of: On Track (≥75% commitments visibly landed), Mixed (1-2 commitments slipped with context), Off Track (≥50% commitments did NOT land).

## Output Format
- `# 📈 {{domain}} — QBR Outcome: <On Track|Mixed|Off Track>`
- `## Commitment review` — table: commitment | status (Landed / Slipped / Pending) | evidence + source
- `## Context` — 1-2 sentences on any major event since the last QBR
- `## Recommended next-QBR thread` — one specific topic to surface in the next QBR (e.g., "Acknowledge the genAI buildout — the commitment to maintain intent on our category slipped 30%; we need to defend our role in their AI stack")

## Quality Checklist
- Every commitment cites a specific signal source for its status
- Pending checks are flagged honestly (don't claim landed without evidence)
- Verdict matches the named thresholds
- No fabricated commitments — only what's in {{prior_qbr_notes}}
- Cap tool calls at ~10