Programme profile
micro1 licenses de-identified operational data, such as SOPs, knowledge bases, CRM data and project histories, to build RL environments. It publishes three tiers ($100K to $500K, $500K to $2M and $2M+), pays in full within 30 days of acceptance, asks for 30 or more full-time employees with data mainly in English, and pays referrers $50K, or 7.5% on larger deals.
Page updated 19 Oct 2026. Figures checked 18 Oct 2026 at their sources.
Data partnerships for RL environments
Tiers from $100K to $2M+; paid within 30 days of acceptance; 30+ employees.
5
sourced facts on record
As published
Operational material: SOPs, knowledge bases, internal documentation, CRM data, project histories and QA processes, plus decision patterns and human feedback on AI outputs. It is used, de-identified, to build reinforcement-learning environments.
Three tiers: $100K to $500K for qualified partnerships, $500K to $2M for large datasets or ongoing work across several teams, $2M+ for highly unique proprietary data. It says payments typically run from $150K to over $2M, and it licenses qualifying agent trajectories for $1M to $2M a year. Its estimate page shows an indicative $50,284 to $92,776, described as not a quote or binding offer. Pay depends on quality, uniqueness and value.
A lump sum for the initial data. After transfer, micro1 has 30 days to review and approve it; the company then invoices and is paid within 30 days. Agent trajectories are licensed per year. Companies keep ownership, can review samples before use, and data is licensed only to labs and enterprises in the US and allied countries.
(micro1, checked 18 Oct 2026)
30 or more full-time employees, mature operations and several years of records across business systems. Data mainly in English, and the team must work in English. US companies come first, then other Western markets.
(micro1, checked 18 Oct 2026)
$50,000 for enterprise deals up to $700,000, or 7.5% of the deal value above that. For companies that no longer operate: $75,000 for deals up to $1 million, or 7.5% above. Paid by wire once the referred company reaches set milestones.
(micro1, checked 18 Oct 2026)
On 9 October 2026 micro1 committed $1B over 12 months to enterprise data acquisition, with capital from Citi and Hercules Capital. Forbes reported a raise of over US$100M at a US$4B valuation, and a late US$12.5M bid for part of Spirit Airlines' data, which Google won at US$10M.
(micro1, checked 18 Oct 2026)(TNW, checked 18 Oct 2026)(Forbes Australia, checked 18 Oct 2026)
These points were not on micro1's public pages when we checked on 18 Oct 2026. They are the questions to put in writing before signing anything.
The licence clause comparison shows how other programmes answer the same points.
Every statement above links to one of these pages. All were read on 18 Oct 2026.
Listing a programme is not an endorsement. How we choose and check is on the methodology page.
Keep reading
FAQ
It publishes tiers of $100K to $500K, $500K to $2M and $2M+, depending on scale and how unique the data is, and says payments typically run from $150K to over $2M. An indicative estimate on its site shows $50,284 to $92,776, labelled not a quote or binding offer.
30 or more full-time employees, mature operations and several years of records, mainly in English. It prioritises US companies, then other Western markets.
$50,000 for enterprise deals up to $700,000, or 7.5% of the deal value above that. Referrals of companies that no longer operate pay $75,000 up to $1 million, or 7.5% above. It is paid by wire after set milestones.
It de-identifies operational data such as SOPs, knowledge bases and project histories and uses it to build reinforcement-learning environments.
It announced on 9 October 2026 a commitment of $1B over 12 months to enterprise data acquisition, backed by Citi and Hercules Capital.
After you transfer the data, micro1 has 30 days to review and approve it. You then send an invoice and are paid within 30 days, so up to about 60 days after transfer. It describes this as payment in full, not instalments.
Stay current
Once a month: deals that disclosed a value, programmes that changed their published figures, and corrections to this site. Free, and one click to leave.
Holding an offer already? Compare it with published ranges.