Creative agency briefs, client email and project chat, 17 agencies
Summary
Client email, public Slack channels, Asana projects and shared-drive briefs from 17 US brand, digital and B2B content agencies. Work runs from new-business pitches and creative briefs through concept rounds, client feedback, legal review of claims, production schedules and campaign wrap reports. Client brands, people and products are coded, and casting files, talent photos, audio and video, consumer lists supplied for campaigns and HR threads are left out.
Collected to order. We collect it from US companies for your order. Scope, volume and timing are confirmed in writing before you sign.
Pricing on request
Request scope and pricingAt a glance
- Planned scope. Volume, sources and timing are confirmed in writing before you sign.
- Records
- 26.4Memails, channel messages, tasks and documents
- Size
- 52.3 GB
- Tokens, estimated
- 5.2B
- Years
- 2015 to 2026
- Firms
- 17
- Owner profile
- 17 US marketing and creative agencies, 8 to 120 staff each, Northeast, West Coast and Texas
- Industry
- Professional services
- Data type
- Company operations archive
- Headcount
- 8 to 120 staff per agency
- Region
- United States (Northeast, West Coast, Texas)
- Languages
- English
- Formats
- JSONL
- Updated
Sources
| System | Kind | Records |
|---|---|---|
| Google Workspace mail | 14.6M messages | |
| Slack public channels | Team chat | 9.8M messages |
| Asana | Projects and wiki | 1.9M tasks and comments |
| Shared drives | Shared files | 112K briefs, decks and scripts |
Good for
- training agents to write and critique creative briefs
- modeling client feedback rounds and predicting revision scope
- evaluating project management agents on production schedules and handoffs
- tone and register modeling for client-facing agency email
Fields
| Field | Type | Description |
|---|---|---|
| record_id | string | Stable record id |
| agency_code | string | Code of the contributing agency |
| client_ref | string | Client brand code, stable across projects |
| source | enum | email, slack, asana or document |
| project_ref | string | Project or campaign code |
| phase | enum | pitch, brief, concept, revision, legal_review, production, launch or wrap |
| created_at | timestamp | Creation time in UTC |
| author_role | enum | account_lead, strategist, creative_director, designer, copywriter, producer or client |
| text | string | Message, comment or document text |
| thread_refs | array<string> | Linked records in the same conversation across tools |
Example record
{
"record_id": "AG11-E-3381094",
"agency_code": "AG-11",
"client_ref": "[COMPANY_4]",
"source": "email",
"project_ref": "[PROJECT_19]",
"phase": "revision",
"created_at": "2025-04-08T21:14:52Z",
"author_role": "client",
"text": "Thanks for round 2. Leadership liked Concept B, Built for the Long Haul, but legal flagged guaranteed uptime in the headline because we cannot back it up across every fleet size. Can we see B with the depot photography instead of the illustrated route map, and a version of the 15-second cutdown without the offer end card? We need the final key visual by the 14th to make the trade show print deadline in [CITY_2].",
"thread_refs": [
"AG11-E-3380712",
"AG11-S-9917345"
]
}Licenses available
Exclusive
One buyer. Nobody else can license it during the exclusive term, usually 2 years.
Non-exclusive
Licensed to more than one buyer.
First look
One buyer gets 2 to 4 weeks to make an offer before anyone else sees the dataset.
What ships with it
- Datasheet: what it is, what it is good for, verified size, sources, years, format and limits.
- Provenance file: the owner shown by code, staff notice and consent status, exclusions, mapped to the EU AI Act Article 53 training-data summary and California AB 2013.
- Manifest with a hash for every file.
- QC report and de-identification report.
- Your own copy, with unique canary records.
How it is de-identified
Default exclusions removed, files flattened to text, identifiers found in three layers and replaced with consistent placeholders, and the result measured against a 99% bar for each identifier type in each source.