Operations

CRM Activity Classification with AI

Discovery, Demo, Negotiation, Follow-Up, Closed Won, Closed Lost — classify activity notes by sales stage so pipeline reports stop lying.

Open in Data Classification Prompt

Overview

Pipeline reports are only as good as the stage labels on activity notes — and reps label inconsistently. This setup classifies CRM notes into six stage labels with definitions that anchor each stage to observable activity ("active discussion of terms, pricing, or contract details" = Negotiation, regardless of what the rep titled the note). Conservative ambiguity fits reporting: choose a label only when the definition genuinely covers the note, and when torn between stages, the closer definition wins — not the more optimistic one.

How to use this resource

One note, one label

Single-label mode: a note describing a demo AND price talk gets the primary-purpose rule — what the meeting mainly was.
Anchor stages to activity

The definitions describe what happened, not how the rep felt about it — that's what makes the labels comparable.
Watch the Negotiation drift

Conservative mode plus "the closer definition wins" stops every promising call from being labeled Negotiation.

Why This Works

Activity-anchored definitions remove rep optimism from stage data
Primary-purpose edge rule handles the call that covered three stages
Conservative ambiguity biases toward accuracy over completeness — right for reporting

Best for

RevOps teams whose stage data is rep-dependent
CRMs full of notes titled "call" with no stage at all
Pipeline reviews that need Closed Lost honestly labeled

Not for

Extracting the next step or contact details from the note — that's extraction
Forecasting deal outcomes — labeling the past, not predicting the future

Use cases

Backfilling stage labels on free-text activity notes
Normalizing stage definitions across reps and regions
Feeding pipeline dashboards labels that mean the same thing everywhere

FAQ

How does it label a note that covered a demo and pricing in the same call?

The edge-case rule picks by primary purpose — "what the writer mainly wants" — so a note that was mostly a demo with some price talk lands on Demo, not both. If two labels genuinely fit equally, the ambiguity policy re-reads both definitions and takes the closer one. It returns exactly one label as JSON, and a human still owns pipeline decisions.

What stops every promising call from getting tagged Negotiation?

Conservative labeling plus definition-anchoring. Negotiation is defined as "active discussion of terms, pricing, or contract details," and the rules judge by content against that definition, not by the label name or an optimistic read. When torn, the closer definition wins rather than the more hopeful one — the guard against the Negotiation drift the workflow warns about.

Does the assigned stage decide whether the deal will close?

It doesn't forecast anything. This labels the past — matching a note against six activity-anchored definitions — not predicting the future; forecasting is explicitly out of scope. The label reflects the model's read of the note text, so accuracy depends on the note and your review. Treat it as normalized stage data for reporting, not a deal-outcome prediction.

Customize This Resource

Opens this setup in Data Classification Prompt. Generate to get the full classification prompt — then adjust the labels, ambiguity mode, and confidence output.

Open in Data Classification Prompt

Prompt Template

Copy it as-is, or use Open in Data Classification Prompt to load it pre-filled and customize it with your own context.

TASK
Classify CRM activity notes by sales stage for pipeline reporting.
Assign exactly one of the labels defined below to the text in the input.

LABELS
- Discovery Call: An initial call to understand the prospect's needs and context.
- Demo: A product demonstration session or a request for one.
- Negotiation: Active discussion of terms, pricing, or contract details.
- Follow-Up: A scheduled or promised continuation of an earlier conversation.
- Closed Won: A deal that has been signed and won.
- Closed Lost: A deal that ended without a sale.

CLASSIFICATION RULES
- Read the entire text before deciding — the deciding signal may come late.
- Judge by content and intent, not by tone, length, or formatting.
- Match the text against the definitions, not just the label names.
- Use only the labels defined above — never invent a new label.

EDGE CASE RULES
- If the text fits more than one label, choose the one matching its primary purpose — what the writer mainly wants.
- If two labels fit equally, prefer the more specific definition over the more general one.

AMBIGUITY POLICY
- Choose a label only when its definition genuinely covers the text.
- When torn between two labels, re-read both definitions and pick the closer one — do not guess from the names.

OUTPUT FORMAT
Return your classification as a single valid JSON object with a "label" key.

EXAMPLE OF A VALID RESPONSE
{
  "label": "Discovery Call"
}

OUTPUT RULES
- Return only the JSON object — no text before or after it.
- Do not wrap the output in markdown code fences.
- Return the label text exactly as defined above — no paraphrasing, no abbreviations, no new labels.

More resources from Data Classification Prompt

Resource

Classify Support Tickets with AI

Billing, Technical, Account, How-To, Feature Request — ticket triage with definitions that decide the borderline cases for the model.

Support

Resource

Text Classification Prompt — the Anatomy

The blocks a reliable classification prompt needs: defined labels, classification rules, edge-case rules, an ambiguity policy, and a confidence contract.

Prompt Engineering

Resource

Categorize Customer Feedback with AI

Praise, Complaint, Feature Request, Bug Report, Question — multi-label feedback categorization where one message can carry three signals.

Product

Resources that pair well

Resource

Extract Data From Text with AI

Free text in, named fields out. The extraction prompt pattern that turns any unstructured text into consistent, parseable records.

Prompt Engineering

Resource

Force JSON Output from AI

Stop getting 'Sure, here is the JSON…' — the output-contract pattern that forces models to return only parseable JSON: schema, example, and a strict rule block.

Prompt Engineering

Resource

Fix Invalid JSON from AI

The JSON won't parse and you can't see why. Deterministic cause-sniffing — trailing commas, single quotes, unclosed brackets — and the repair prompt that fixes it.

Engineering

Related tools

Tool

Data Classification Prompt

Build classification prompts that assign labels from a closed set — with label definitions and edge-case rules.

Structured Output

Tip: Save time by exploring related resources and tools that integrate with this resource.