By: Dr. Tamara Patzer
Most chief executives have never put a price on the binder, wiki or shared folder that explains how their company runs. On October 9, San Francisco AI data company micro1 announced that it will spend $1 billion over 12 months on exactly that kind of material through Company Data Partnerships, a program for established companies. If your company runs on procedures, playbooks, saved message threads and shared documents, you may hold more of that material than you realize, and it may be worth more than you expect.
Key Takeaways
- micro1 says it will spend $1 billion over 12 months licensing company operating data.
- The program lists standard operating procedures, process documents, knowledge bases, workflows and templates.
- It sets a floor of 30 full-time employees, so it reaches about 2 percent of U.S. businesses.
- Payment is not ensured, and the company calls its estimates “indicative only.”
- Rights, exclusivity and liability are the open questions that an executive’s counsel should settle first.
What Could Count: Your Messages, Procedures and Documents
Many executives assume the valuable material lives somewhere else, in a vault or a data warehouse. More often it is already in daily use. It is the Slack thread where a manager walked a new hire through a hard call. It is the email chain that settled a disputed contract. It is the standard operating procedure that someone wrote after the third time a shipment went wrong. A company can own a great deal of this without ever thinking of it as an asset, and for many leaders it turns out to be among the most valuable things the business holds.
Executives often have their own words for it. You may call it “the way we do things here.” In this world, it is called a standard operating procedure. The binder in the operations office is a playbook. How your best manager handles an unhappy customer is a workflow. The email everyone copies is a template. The shared drive nobody can find anything in is a knowledge base. The filing cabinet of paper manuals is a document archive, and it can be scanned.
That last point matters. Paper binders, printed manuals and signed forms can be scanned, and files that already exist in digital form can usually be exported. micro1’s page lists procedures, process documents, knowledge bases, workflows and templates. Message and email archives are a separate question, so ask the program which record types it accepts before assuming either way.
A short test: ask three department heads what a new hire would need to learn in the first year. Whatever they point to may be what a buyer wants to see. If you are near the 30 full-time employee line, run the online estimator.
What Is the Company Buying?
The page describes records of how work gets done: standard operating procedures, process documentation, internal knowledge bases, workflows, templates, CRM records, support tickets and project histories. micro1 says it uses the material to build reinforcement learning environments, which are practice workplaces where AI agents attempt business tasks and get feedback.
The value sits in the logic of the work. A clear procedure for a disputed invoice teaches an AI agent the same lesson whether or not it names a customer. micro1 says personal and identifying information is removed or transformed before data is shared with third parties, and a cautious executive can begin with templates that carry no customer names.
Who Qualifies?
The eligibility requirements describe companies with at least 30 full-time employees. Several years of records across multiple business systems make a company a stronger fit, the page says, and the data should be primarily in English. Demand is strongest in the United States, the United Kingdom and Canada.
How Many Companies Fit?
Fewer than the headline suggests. The U.S. Small Business Administration’s Office of Advocacy counts 36.2 million small businesses, using 2022 Census Bureau data. Of those, 29,811,495 have no employees, and about 6.4 million have a payroll. Advocacy’s 2025 profile lists 654,501 firms with 20 to 499 employees, and its earlier tally counted 21,041 with 500 or more.
That is roughly 675,000 companies with 20 or more workers, and the count that meets micro1’s threshold of 30 is lower. Census brackets do not break at 30, so no public source gives the exact number. These companies are about 2 percent of American businesses.
What Does the Company Say It Pays?
As of November 7, the page lists three tiers: $100,000 to $500,000 for qualified enterprise partnerships, $500,000 to $3 million for large operational datasets or work across several teams, and $3 million and up for highly unique proprietary data.
These are advertised ranges. The page calls its estimate “an indicative estimate only” and says the final offer depends on data sources, size, quality, access terms and due diligence. The technology site Magica noted in October that such figures are not observed average payouts, and earlier reports quoted different top-tier numbers. Executives should read the current page before relying on any figure.
What Should an Executive Settle First?
Rights come before price. Rajesh Beri, writing at Beri.net on October 10, said owning files does not mean owning the right to license what is inside them. Customer contracts and nondisclosure agreements may bar use for AI training, privacy rules can limit new uses of personal data, and knowledge bases often include third-party material.
He also noted that micro1’s public terms leave open who the downstream buyer is, whether a license is exclusive and where liability falls. Those answers belong in a contract reviewed by counsel.
De-identification has limits too. A 2019 Nature Communications study by Luc Rocher, Julien Hendrickx and Yves-Alexandre de Montjoye estimated that 99.98 percent of Americans could be re-identified from 15 demographic attributes. That is a model estimate, not a test of any program, but it is why reviewing a sample matters. micro1 says companies keep ownership, can review samples, and that raw data is deleted on an agreed schedule.
How Does the Process Work?
The application process has six stages: an online estimator covering size, years in business, entities, share of English data, location and data sensitivity; an indicative estimate; a short application; discovery and pricing; contracting; and delivery. micro1 has 30 days to review delivered data, and payment follows within 30 days of invoicing after approval. Payment depends on the data being accepted, and no offer is guaranteed.
Frequently Asked Questions
Does the company have to anonymize its own data?
micro1’s page says no. Confirm that in the contract.
Does the company keep ownership?
micro1 says companies retain ownership. The license terms, which are not published, decide what micro1 may do with the data.
Who inside the company should own this?
Many advisers suggest one accountable executive, such as general counsel, with a written inventory of what exists, who created it and which contract governs it.
What if the company has fewer than 30 employees?
This program lists 30 as its threshold. Other AI data buyers set different ones.
Where can a qualifying company start?
Companies that meet the size requirement can start with micro1’s company partnership page.
Is there a minimum amount of data?
micro1’s page says no fixed minimum applies. It looks at historical depth, breadth, quality and relevance, and says several years of records across multiple systems are a stronger fit.
Who can I ask at micro1?
Start with micro1’s application page, linked in this article.
Dr. Tamara Patzer is a publisher and former newspaper editor who writes about AI and business data.
Sources (all read November 7, 2026)
- micro1, “micro1 commits $1B to enterprise data acquisition,” October 9, 2026: micro1.ai/blog/micro1-commits-1b-to-enterprise-data-acquisition
- micro1, Company Data Partnerships estimator and FAQ: micro1.ai/data-estimate
- Rajesh Beri, “micro1 commits $1B to buy enterprise data for AI training,” Beri.net, October 10, 2026
- U.S. Small Business Administration Office of Advocacy, “Frequently Asked Questions About Small Business,” February 2026, and 2025 Small Business Profile for the United States (Census Bureau 2022 data)
- Magica editorial team, October 2026
- Luc Rocher, Julien M. Hendrickx and Yves-Alexandre de Montjoye, Nature Communications 10:3069, July 23, 2019
- micro1, Data Partnerships page, micro1.ai/data-partnerships, read November 7, 2026
Disclaimer: The author is a referral partner of micro1. This article is information, not a promise of money, and not legal advice.
