DataAnnotation vs Outlier: Pay and Access
Facts checked 13 October 2026
DataAnnotation publishes higher pay ceilings but gives you one attempt at its entry assessment and faces a misclassification lawsuit. Outlier publishes no rates, onboards in a fixed sequence, and reviewers flag availability dropping to zero. Both can leave gaps between projects.
Side by side
| DataAnnotation | Scale AI / Outlier | |
|---|---|---|
| Operator | Run by Surge AI (Surge Labs, Inc.), per reporting; site says "powered by surge" | Operated by Scale AI, per Outlier's site |
| Published pay | Up to $150+/hr (software/ML), $125+/hr (finance, legal, medical, STEM), up to $50/hr generalist, up to $40/hr languages | No rates published; "competitive pay plus quality-based rewards" |
| Our record | $20-60/hr | $15-60/hr |
| Payment | PayPal; deposits within a few days of a withdrawal request | Weekly; method not stated on the page |
| Entry step | Starter assessment, about 1 hour (general) to 2 hours (specialist), one attempt only; ID check | 30-90 min general onboarding (account, expertise, skill screenings, ID check), then project-specific onboarding |
| What you need | Fluent English, reliable internet; track-specific skills | Valid ID and mobile number from your country, a resume and a LinkedIn profile |
| Countries | 50+ per the About page; much of the work is US-based on our record | Says "50 countries"; list not published |
| Our rating / reliability | 4.1 / Reliable | 3.6 / Very Reliable |
| Public reviews | Trustpilot about 3.9 from roughly 1,960 reviews (55% five-star, 20% one-star) | See Trustpilot and community forums; no consolidated figure used here |
| Main risks | Weeks with no projects, account suspensions, May 2025 class action (allegations unproven) | Task availability can drop to zero; support can be slow |
Platforms change rates and rules often. Where a figure comes from a platform's own pages it is that platform's claim, not a guarantee.
Pay: one publishes numbers, the other does not
DataAnnotation lists specific ceilings by field, and they are high for specialists: up to $150+ an hour for software and machine learning, $125+ for most professional fields. Its own FAQ quotes lower figures for general work, so the best reading is that the top rates apply to scarce skills. Outlier's site names no rates at all, which makes any claim about its pay a third-party estimate. Our record lists both in the same broad band, roughly $15 to $60 an hour.
The practical point is that neither number tells you what you will bill in a month. Check DataAnnotation's pay detail and our Outlier platform page for the assumptions behind our figures.
Getting in: one attempt versus a sequence
DataAnnotation puts everything on one assessment, and its FAQ says you cannot retake it. That makes preparation important; we cover it in the starter assessment guide. Outlier uses a longer but more forgiving sequence: general onboarding of about 30 to 90 minutes, then separate onboarding for each project. Expect to supply a resume and a LinkedIn profile for Outlier.
Reliability: both can leave gaps
Neither platform guarantees work. DataAnnotation reviewers describe weeks without projects and account suspensions, and the operator faces a proposed class action alleging misclassification and unpaid assessments (allegations only, no ruling found). Outlier's main complaint on our record is that availability can fall to zero. If steady income matters, keep a second platform active. Our platform rankings show alternatives.
Bottom line
Pick DataAnnotation if you are a specialist who can use its higher published ceilings and you are comfortable with a single-shot assessment. Pick Outlier if you prefer a longer, repeatable onboarding and are willing to work around variable project availability. If you are new to this kind of work, apply to both and see which actually sends you tasks.
Common questions
Keep reading
AI Work Expert is an independent publisher. We do not hire or pay anyone, and nothing here is financial or legal advice.