Account Intelligence Synthesis Engine
Transform raw account data into a structured, evidence-aware signal summary for strategic account planning.
Customise & copy
Add the context you have. The prompt updates as you work.
Enter the name and domain of the company being analyzed.
Paste recent press releases, announcements, or news articles.
Paste active job postings or department growth metrics.
Paste LinkedIn updates, executive appointments, or departures.
Paste tech stack detections, infrastructure changes, or tool usage.
Paste internal notes, meeting logs, and inbound email history.
Live prompt
6 fieldsYou are a precise intelligence analyst. Your job is to consolidate and organize disparate raw account intelligence for Acme Corp (acme.com) into a structured, evidence-aware signal summary. Raw input data provided: 1. Company News & Press Releases: None provided 2. Hiring & Job Postings: None provided 3. Role Changes & Leadership Updates: None provided 4. Technology Stack Lookups: None provided 5. CRM Notes & Inbound Emails: None provided Instructions: - Organize the summary strictly by data source and chronological date. - Maintain absolute epistemological separation: explicitly distinguish between verified facts, external signals, and internal subjective notes (CRM/emails). Do not flatten unverified signals or scraping data into hard facts. - Highlight recent momentum, hiring trends, and technical footprint changes objectively without hype, sales pitches, recommendations, or unsolicited commentary. - Include source attribution and dates for every individual data point.
Proof before you use it
A real example
Tested on gemini-3.1-flash-lite on 2026-09-04
{
"crm_notes": "Account manager noted in Salesforce on Jan 20, 2025 that the client is evaluating vendor consolidation for Q3.",
"news_data": "Apex health network acquired St. Jude Community Hospital on Oct 05, 2024.",
"tech_data": "Wappalyzer detection on Jan 03, 2025 indicates usage of Epic EHR and Salesforce Health Cloud.",
"hiring_data": "Posted 5 compliance officer and HIPAA security specialist roles on Nov 12, 2024.",
"role_changes": "Dr. Mark R. appointed Chief Medical Information Officer (CMIO) on Dec 01, 2024.",
"target_company": "Apex Healthcare Systems (apexhealth.org)"
}A small ritual that works
How to use it
- 01Paste your raw data from news, hiring, tech stacks, and CRM notes into the corresponding fields.
- 02Run the prompt to generate a chronologically organized, source-attributed intelligence brief.
- 03Review the output to ensure clear epistemological separation between verified facts and internal subjective notes.
Why it works
The prompt enforces strict epistemological separation between verified facts, external signals, and internal subjective notes. It structures unstructured multi-source inputs into categorized, chronological groupings while requiring source attribution and dates for every data point.
Where it fails
Both test outputs violated the instruction to organize strictly by data source and chronological date by misordering items and mixing source categories. They also introduced unsolicited analytical momentum summaries and overstated external scraping observations as active migrations.
Customisation tips
To enforce stricter adherence, add negative constraints explicitly banning 'analytical summaries', 'momentum interpretations', and inferences beyond explicit text. Require a literal one-to-one mapping section for each input variable.
Watch-outs
- LLMs naturally default to adding analytical 'so what?' summaries or trend sections that violate anti-hype instructions.
- Scraping and technical lookup data (e.g., DNS/BuiltWith records) are frequently over-interpreted as active corporate projects or migrations.
- Conflating internal CRM subjective notes with hard corporate facts compromises epistemological separation.
Works in
Was this useful?
Submit your variation
If your version helps, it may be published with your name and a link back to you.
Want to make AI useful across your team? Explore Atul's training.