Mercor vs micro1: connect our tools, or license our written processes?

Mercor connects to your tools, publishes no size rule and typically pays 2–4 weeks after signing. micro1 agrees redaction rules with you first, then pays firms with 30+ employees for written processes, on unpublished timing. Your records decide fit. Both pay the company itself, as of October 2026.

How does each one get our data?

Mercor connects to your accounts; micro1 does not say whether it connects to systems or receives files (checked October 5, 2026). Mercor’s program page lists four steps: Scope, Connect (“Choose from 100+ supported tools and link your accounts”), Process (“Our platform extracts and de-identifies data into an asset”) and Payout. It names Slack, Microsoft Teams, Gmail, Outlook, Confluence, Notion, Drive, Salesforce and GitHub, and says “No data leaves your systems until you approve it.”

micro1’s program page describes the order of decisions instead: “Before any participation, together we define” the approved data scope, security and confidentiality standards, “anonymization and redaction requirements,” and internal approvals. An assessment, a review and a discovery call come before the agreement.

Which one asks less of our team?

Mercor, on the two companies’ own descriptions. With Mercor, your team picks the tools and links the accounts; Mercor says extraction is “fully automated” and that it “covers all costs associated with data extraction, anonymization, and transfer” (program page). With micro1, you help write the redaction rules and, its FAQ says, have “the opportunity to review representative samples before use”; your representatives must also “communicate with the micro1 team in English throughout the partnership and onboarding process.”

If you want the redaction rules settled before data leaves, micro1 builds that step in. How much work is it to license company data? lists the tasks that fall on your side.

Will either consider a company our size and industry?

micro1 publishes a floor of 30 employees, and a ceiling of 200 on one page; Mercor publishes no size rule, and the two name different industries.

	Mercor	micro1
Headcount	No rule published	30+ on its program and referral pages; 30 to 200 on its application page
Industries named	Legal, healthcare, finance and insurance, professional services, manufacturing, retail and e-commerce, real estate, logistics and supply chain	Technology, BPOs and recruiting, financial services, consulting, professional services, legal and compliance, healthcare administration, logistics and operations
Language and location	No rule published	Companies that “operate in English”; “We currently prioritize US companies, followed by other Western markets”

The table reflects Mercor’s program page and micro1’s pages as of October 5, 2026. Mercor’s lack of a rule means its criteria surface during scoping, not that it takes everyone. Above 200, ask micro1 which of its two rules applies. Manufacturing, retail and real estate appear only on Mercor’s list, though micro1’s program page says “operational data from every industry can contribute.” Does my company’s data qualify? compares thresholds across buyers.

Which of our records does each want?

Where their industries overlap, they ask for different layers of the same business. Mercor’s examples are the records the work produces; micro1’s are mostly documents that say how the work is done:

Area	Mercor’s examples	micro1’s examples
Legal	Contracts, case files, discovery docs	Policies, contract workflows, review processes, compliance procedures
Finance	Statements, underwriting files, claims	Financial processes, SOPs, reconciliations, approvals, reporting workflows
Logistics	Shipping manifests, dispatch logs, invoices	Operational SOPs, logistics processes, fulfillment workflows, inventory management

The examples come from Mercor’s program page and micro1’s (checked October 5, 2026). Mercor’s connectors also reach email and chat. micro1’s page does not mention email, Slack or chat; it asks for SOPs, knowledge bases, templates and “operational frameworks,” plus CRM data and project histories. A company whose history sits in inboxes and line-of-business software matches Mercor’s list; one with a maintained wiki and written procedures matches micro1’s. A data inventory shows which, and whether you could offer messages to one and manuals to the other.

Which one pays more?

Public figures can’t settle it, because Mercor publishes none. It says compensation “is based on data volume, the tools you connect, and the depth of your data,” and “connecting more tools generally results in a higher payout” (program page). micro1 lists tiers of $100k+ for “Qualified Enterprise Data Partnerships,” $500k+ for “large-scale operational datasets or ongoing participation across multiple teams and business functions,” and $1M+ for “highly unique, proprietary operational data” (program page). Its referral page says $100K–$2M+ “for approved data packages,” and its estimator shows an example of $50,284–$92,776, “not a quote or binding offer.”

The only public prices from both concern one archive. In August 2026, Mercor’s $7.5 million offer for Spirit Airlines’ de-identified business data became the backup bid after Google was named the winning bidder at $10 million (Bloomberg Law). micro1 then proposed $12.5 million in a court filing (Inc.; Fortune). That archive held about 100 million emails and 500 million Teams chats from a defunct airline, so neither figure says what an operating company would get. Court approval of the sale was still pending as of late September 2026 (Moneywise).

To compare the two, get a written figure from each, after scoping, for the same records. They offered us $75K. Is that a good offer? shows how to read one.

When does each pay, and what triggers it?

Mercor pays one wire when the de-identified dataset reaches its buyer, and micro1 does not say when it pays. Mercor’s page says, “Payment is issued via wire transfer upon delivery of the anonymized dataset to the buyer. This typically occurs within 2-4 weeks of the signed agreement, depending on data extraction and processing time.” It says it takes no fees (program page). The trigger is delivery, not signing, so ask what happens if extraction runs long or the buyer is not ready.

micro1’s program, application, referral and estimator pages give no payment timing or structure (checked October 5, 2026). Its $500k+ tier mentions “ongoing participation,” and its referral page pays referrers on different triggers for “Workflow” partnerships (after onboarding and “the minimum revenue threshold”) and “Corpus” partnerships (once data “has been accepted by micro1”). It does not define either type for companies. Ask which you are offered, what triggers each payment to you, and how long any ongoing commitment lasts. What sets the date is covered in How long until I get paid?.

Who else gets paid if we sign?

Possibly whoever introduced you. Mercor’s program page offers a referral bonus, paid “when they go live and start earning,” and gives no amount (checked October 5, 2026). micro1 pays “up to $50K per referral, depending on the size and value of the partnership,” if the company applies through the referrer’s link (referral page). Both also approach companies directly. Eight days after HubSpot agreed to acquire Warmly, Mercor offered to buy or license Warmly’s codebase and task records, according to The Information as summarized by AI Weekly in August 2026. A micro1 job listing calls for “outbound prospecting” to business owners. If a contact recommends either, find out whether that person is paid when you sign.

How does each de-identify the data, and can we see it first?

micro1 says you can review samples; Mercor’s page doesn’t address it. micro1’s FAQ says PII “is removed when relevant,” some datasets “may undergo synthetic rewrites that preserve workflow structure while reducing direct ties to original records,” data is “retained only for agreed periods,” and deletion follows the agreed terms; its application page says “Original datasets are deleted after processing.” In September 2026 it published a model that swaps real identities for consistent synthetic ones. Its program page names no security certification.

Mercor says “No AI lab buyer or third-party vendor ever sees your raw data,” that it logs every access, and that it holds SOC 2 Type II with encryption keys in AWS KMS (program page). The page is silent on the de-identification method, on any review of output before delivery and on how long the extracted dataset is kept (checked October 5, 2026). Mercor lists clinical notes and claims among its healthcare examples, and micro1’s estimator asks whether your data is regulated or sensitive; if yours is, Is it legal to sell company emails and records? lists what to check.

Has Mercor or micro1 had a security incident?

Mercor has. It told TechCrunch on March 31, 2026 that a compromise of LiteLLM, an open-source project, had hit it along with thousands of other companies, and that it “moved promptly to contain and remediate the security incident.” According to Cybernews (April 1, 2026), the Lapsus$ group claimed to hold about 4 terabytes of Mercor data, source code included. Citing Wired, TechCrunch reported on April 9 that Meta had paused its Mercor contracts; citing Business Insider, it added that five contractors had sued over alleged exposure of their personal data. No report said whether the incident touched records that companies had licensed through Mercor’s data program.

Data Licensing Report found no reported security incident involving micro1 as of October 5, 2026.

Because Mercor’s connectors would hold access to your accounts, ask what it changed after the incident, how connector credentials are revoked, and whether its SOC 2 Type II report covers the months since. Ask micro1 which certification, if any, covers the data program.

Can we license the same data to Mercor and micro1?

Neither publishes its agreement or says whether a license is exclusive (checked October 5, 2026). Mercor’s page says “We license it, package it, and pay you for it.” micro1 says “Companies retain ownership of their underlying data,” which does not settle exclusivity: you can own data and still grant an exclusive license. Splitting different records between them may be possible, but only if neither agreement bars it. For how owning data and licensing it differ, see Am I selling my data or licensing it?.

What has each done with company data so far?

Neither names a price for any completed company-data deal. micro1 told the Spirit court it had completed more than 50 data transactions in the preceding 45 days, without naming sellers, prices or AI-lab customers, and a court-appointed privacy watchdog said it needed more time to investigate micro1 if the court considered it as buyer (Fortune, September 14, 2026). Its application page shows “$200M+ Generated for partners” without saying what that counts. Mercor’s reported approaches, to Spirit and to Warmly, both involved companies that were closing or being acquired.

Mercor was founded in 2023 and micro1 in 2022. TechCrunch put Mercor’s annualized recurring revenue at about $500 million in October 2025, and Fortune reported micro1’s gross annualized run rate at $500 million by August 2026; both figures cover the whole business, not payouts to data partners.

What should we ask Mercor and micro1?

Start with how data would move and what each would pay for, then deletion and the license:

1. How does data leave our systems: connectors you run, an export we prepare, or files we upload? How do we revoke access?
2. Which systems and record types are in scope, and how does adding one change the price?
3. Will you put a figure in writing after scoping, and what could change it before payment?
4. Can we review a de-identified sample before delivery, and when are our originals deleted?
5. Which AI labs would receive the data, for how long, and on exclusive terms or not?

When does Mercor fit better than micro1?

When your history sits in messages and business systems, your headcount is outside the 30-to-200 range on micro1’s application page, or you want a known payment date.

• Your history sits in email, chat and business systems such as Slack, Teams, Gmail, Salesforce or GitHub.
• You have fewer than 30 employees, or more than 200, the top of the range on micro1’s application page (its other pages set no ceiling).
• You want to know before signing when the money comes. Mercor says it wires payment once the buyer receives the dataset, typically 2–4 weeks after signing, and takes no fees.
• You accept connector-based extraction, provided Mercor’s account of what it changed after March 2026 satisfies you.

When does micro1 fit better than Mercor?

When your know-how is written down, you have 30 to 200 employees, and you want redaction rules agreed before any data moves.

• Your know-how is written down in SOPs, a maintained knowledge base, templates and playbooks.
• You have 30 to 200 employees, work in English and are based in the US, which micro1 says it prioritizes.
• You want published tiers, starting at $100k+, to compare a quote with.
• You want redaction rules agreed before data moves and a chance to review representative samples.