Glossary

Bank transfer payout

Payment sent directly to a bank account — via ACH, SEPA, or a wire transfer — without an intermediary wallet. It is the standard route for higher-paying AI-training and transcription work, often paid weekly or monthly.

Cryptocurrency payout

Payment in Bitcoin, Ethereum, or a stablecoin instead of fiat currency. A handful of microtask and get-paid-to platforms offer crypto as a primary or optional payout route.

Defunct platform

A platform that has stopped operating or paying. This directory does not delete dead platforms — it marks them, because the warning is the value: balances on a defunct platform are usually unrecoverable, which is exactly why low minimum payouts matter.

E-wallet payout

Payment delivered to a digital wallet service — PayPal, Payoneer, AirTM, Skrill, and similar — rather than directly to a bank. E-wallets matter for contributors in countries where direct bank transfer or PayPal is unavailable, and many AI-training platforms pay exclusively through regional wallets.

Listed on every platform page

Effective hourly rate

What you actually earn per hour of work after accounting for unpaid screening, rejections, search time, and speed — as opposed to the advertised rate. A "$60/audio-hour" transcription rate is really $10–15 per working hour at beginner speed.

Identity verification

The checks a platform runs before letting you work or pay you: email and phone confirmation, government ID upload, selfie matching, or tax forms. Expert-tier platforms almost always require ID verification (often via third-party services); casual survey apps may need only email.

Listed on every platform page

Independent contractor status

The legal classification of nearly all paid-task work: you are a contractor, not an employee. That means no minimum wage guarantee, no benefits, no tax withholding — you track and pay your own taxes — and platforms issue tax forms only above reporting thresholds.

Instant payout

A payout cadence in which money is released immediately or within hours of completing a task, rather than on a billing cycle. It is most common in mobile survey apps and per-test usability platforms.

Minimum payout

The amount you must accumulate before a platform releases any money. Thresholds range from $1 to $100 depending on the platform and payout method.

Shown on every platform page

Net-30 (and other net terms)

An invoicing convention meaning payment arrives 30 days after the billing period closes — "net-15" and "net-60" work the same way. Net terms are standard in higher-end freelance-adjacent work, including some AI-training and transcription contracts.

Pay model

The unit a platform bills your work in: per task, per audio minute, per message, per study, per hour, or in points. The model predicts your earnings shape better than any advertised rate — per-task pay fluctuates with task availability, per-audio-minute pay depends on your speed, and per-study pay is high-variance.

Listed on every platform page

Payment schedule

How often a platform releases earned money: instantly after each task, daily, weekly, monthly, or once you clear a threshold. Schedule plus minimum payout together decide how fast work becomes cash — a platform that pays after every approved test is fundamentally different from one that holds funds for 30 days.

Listed on every platform page

Qualification test

An entry exam a platform uses to decide who can work: grammar quizzes, sample transcriptions, guideline comprehension checks, or domain assessments. Performance gates both access and pay tier.

Listed in How-to-join on platform pages

Referral program

A scheme that pays you for recruiting other users, usually a flat bonus or a percentage of their activity. Referral income is legitimate when it is a side feature of a real task platform, and a red flag when it is the primary earnings path — platforms where recruiting matters more than working are MLM-shaped and excluded from this directory.

Verification sweep

The recurring process of re-checking every tracked platform against its official signup, payout, and policy pages, with the check date recorded per source. Sweeps run on a rolling monthly schedule; the last full sweep date is printed in the site footer and each platform page carries its own last-verified date.

W-8 BEN form (international)

The IRS form non-US contributors submit to claim treaty benefits and certify foreign status when working for US-based platforms. It is the international counterpart of the W-9 and is commonly required before payout on American AI-training and transcription platforms.

W-9 form (US)

The IRS form US contributors submit to platforms so payments can be reported correctly. Most US-facing platforms request it before your first payout; it collects your name, address, and taxpayer identification number.

Conversion rate (points to cash)

The fixed exchange rate a platform applies when converting your points balance into cash or gift cards — for example, 100 points = $1. It is the single number that decides whether a points offer pays fairly.

Diary study

A longitudinal research format in which participants record their behavior, purchases, or opinions over days or weeks — via app entries, photos, or short surveys. Diary studies pay cumulatively meaningful amounts ($50–$200+) but demand consistent daily participation; incomplete entries are usually pro-rated or unpaid.

Disqualification (DQ)

Being ejected from a survey partway through because your answers no longer match the target profile — the most frustrating and least understood part of survey work. DQ rates of 50–80% are normal on some panels.

Focus group

A moderated group discussion — in person or online — where 4–10 participants give opinions on a product, service, or concept for 60–120 minutes. Focus groups pay far more than surveys, commonly $50–$150 per session, but are infrequent, screen hard, and often require webcam participation or facility attendance.

Used by: survey and market-research platforms

Get-paid-to (GPT) site

An aggregator platform paying small rewards for miscellaneous activities — surveys, videos, offers, app installs, games. GPT sites are legitimate but low-paying intermediaries: they take a cut of advertiser budgets and pass on a fraction.

Gift card payout

Rewards redeemed as store credit instead of cash. Gift cards often unlock at lower thresholds than cash payouts — some panels cash out at $3 in gift cards versus $25 via PayPal — and retailers sometimes run bonus-card promotions.

Honorarium

The fixed thank-you payment for participating in a research session — focus groups, interviews, usability studies — disclosed before you commit. Honoraria are usually cash, gift cards, or prepaid cards paid within days to weeks of the session.

In-depth interview (IDI)

A one-on-one qualitative research session, usually 30–60 minutes by video or phone, in which a researcher explores your experiences and opinions in detail. IDIs pay more than surveys — often $40–$100+ — and recruit for specific professional or consumer profiles, so matching depends heavily on your profile completeness.

Points system

A pay model in which earnings accumulate as points rather than cash, converted at a fixed rate when you cash out. Points hide the real wage: a 500-point survey can be worth $5 or fifty cents depending on the panel.

Used by: most survey panels

Product testing

Receiving a physical product — or early app access — free in exchange for trying it and completing feedback surveys or reviews. Real product testing never asks you to buy the product first.

Screener (screening questionnaire)

The qualifying questions at the start of a survey or study that decide whether you fit the target demographic. Screeners are unpaid time — typically one to three minutes — and dishonest answers to force a match get accounts banned and incentive claims clawed back.

Survey panel

A research community you join by building a demographic profile, then receive survey invitations matched to that profile. Panels are the backbone of consumer market research and the largest category in this directory.

Used by: survey platforms

Survey router

Infrastructure that routes panelists from one survey into others across multiple research clients, often through repeated screeners. Routers are why finishing one survey sometimes chains into several more — and why disqualification feels constant.

AI training data work

The umbrella category for paid tasks that improve AI models: labeling images and text, transcribing audio, evaluating and ranking model responses, writing prompts, and reviewing code. Work ranges from simple microtasks to expert-domain projects.

Annotation

Marking up data for machine learning: drawing boxes around objects in images, tagging entities in text, segmenting audio, or labeling video events. Annotation is the entry level of AI-training work — no degree required, but precision matters, since rejected batches are typically unpaid.

Bounding box annotation

Drawing rectangles around objects in images or video frames so computer-vision models can learn to detect them — the classic image-annotation task. It requires precision and patience: loose boxes are rejected, and quality benchmarks decide continued access.

Data collection tasks

Paid tasks that create new training data rather than label existing data: recording speech samples, photographing objects in specific settings, writing questions and reference answers, or capturing receipts. Collection work often pays better than annotation because it is harder to fake and requires specific hardware or languages.

Expert tier

The pay band on AI-training platforms reserved for domain-qualified contributors — coders, lawyers, doctors, linguists, STEM postgraduates. Expert work (evaluating code, checking legal reasoning, writing reference answers) pays $15–40+/hr versus a few dollars for generalist microtasks.

Microtask

A small, self-contained unit of work — labeling one image, transcribing one sentence, rating one search result — taking seconds to a few minutes. Microtasks pay cents each and aggregate into hourly earnings only at volume.

Used by: AI training and microtask platforms

Named entity recognition (NER) tagging

Text-annotation work that labels entities — names, organizations, locations, dates, products — inside sentences so language models can learn to identify them. It demands reading precision and consistent guideline application across thousands of examples.

Per-task pay

A pay model in which each accepted unit of work — one annotation batch, one test, one transcription file — pays a fixed amount. It is the dominant model in microtask and AI-training work.

Used by: microtask and AI-training platforms

Prompt evaluation

AI-training work in which contributors rate and compare AI model responses to prompts — scoring helpfulness, accuracy, and safety, and explaining their judgments. It is the most widely available expert-tier AI task and typically pays per task or per hour.

Used by: AI training platforms

RLHF (reinforcement learning from human feedback)

The technique behind most AI-training gig work: humans rank or score model outputs, and those preferences train the model to behave better. RLHF is why platforms hire thousands of contributors to evaluate responses — the ranking data is the product.

Search evaluation

Rating search-engine or recommendation results for relevance and quality against defined guidelines, usually for large platforms via long-running contracts. Search evaluation is one of the oldest paid AI-adjacent tasks — stable hours, hourly-equivalent pay, but strict quality monitoring and country-specific demand.

Text classification

Labeling text into categories — sentiment, topic, intent, spam/not-spam, or custom taxonomies defined by project guidelines. It is the most common text microtask and the easiest AI-training entry point: no coding, no degree, just careful reading and consistent judgment.

Unpaid rejection

Work a platform declines to pay because it failed review — mislabeled data, a silent usability test, a transcription that missed accuracy bars. Rejection risk is the hidden tax on per-task models: your true rate is the task price multiplied by your acceptance rate.

Beta testing

Using a pre-release app or game over days or weeks and reporting issues through structured feedback channels. Unlike one-off usability tests, betas are longitudinal: participation can span the whole test period with milestone surveys.

Bug bounty

Paying security researchers per accepted vulnerability report, with amounts scaling by severity — from $50 for low-impact issues to five figures for critical exploits on major programs. Bug bounty is skilled testing work: findings require proof-of-concept reproduction, and duplicates pay nothing.

Moderated interview

A live, scheduled session in which a researcher leads the tester through tasks or questions in real time, usually over video call. Moderated sessions pay substantially more than unmoderated tests — often $30–$90 for an hour — because they require scheduling, reliability, and strong communication.

Playtest

A scheduled game-testing session in which you play a build while recording structured feedback on difficulty, bugs, and enjoyment. Playtests pay per session — commonly $10–$30 for an hour — and differ from casual "play and earn" games: they require selected participants, specific hardware, and written or spoken feedback after the session.

Sample test (entry test)

The practice usability test every testing platform requires at signup — your graded audition. Reviewers score think-aloud quality, task completion, and audio clarity; a failed sample usually means a retry after a waiting period, and repeated failures end the application.

Think-aloud protocol

The core method of usability testing: the tester narrates their thoughts — what they see, expect, like, and find confusing — while completing tasks on a website or app. Platforms train testers in the protocol during onboarding, and clear think-aloud commentary is the single biggest factor in passing the sample test and receiving more test invitations..

Used by: app and website testing platforms

Unmoderated test

A usability test you complete alone: the platform presents tasks, records your screen and voice, and pays after review — no live researcher. Unmoderated tests are the volume business of testing platforms, typically $5–$10 for 15–30 minutes.

Used by: app and website testing platforms

Usability testing

Paid tasks in which real users explore a website, app, or prototype while verbalizing their experience, so product teams can find friction. Tests typically run 5–60 minutes, are completed from home with a microphone (often a webcam), and pay per completed test.

Audio hour

The standard billing unit for transcription work: one hour of recorded audio, regardless of how long the file takes to transcribe. Transcription platforms advertise pay per audio hour (or per audio minute), so a $60/audio-hour rate is not an hourly wage — beginners typically need four to six hours of work time per audio hour, making effective pay a fraction of the headline figure..

Used by: transcription platforms

Audio minute

One minute of recorded audio — the finer-grained cousin of the audio hour, used by platforms that price work per file rather than per hour of footage. A rate of $0.50 per audio minute equals $30 per audio hour.

Used by: transcription platforms

Captioning

Creating time-aligned text for video: dialogue plus speaker labels and meaningful sound descriptions, synced to the frame. Captioning pays more per minute than plain transcription because timing precision adds real work — captions must hit frame-accurate cues.

Used by: transcription and captioning platforms

Clean-read transcription

Transcription that lightly edits speech for readability — removing filler words, false starts, and repetitions while preserving meaning — as opposed to strict verbatim. Most business, podcast, and research clients actually want clean read.

Foot pedal

A USB hardware pedal that controls audio playback (play, pause, rewind) with your feet, keeping both hands on the keyboard. Professional transcriptionists consider it the single biggest speed upgrade — reaching for the mouse hundreds of times per audio hour adds up.

Subtitling

Translating or adapting dialogue into on-screen text for viewers — either same-language subtitles for accessibility or translated subtitles for localization. Translation subtitling pays the best rates in the video-text family because it combines two skills, and platforms gate it with source- and target-language tests.

Timestamping

Inserting time codes into a transcript at regular intervals or at speaker and topic changes, so readers can navigate the audio. Most platforms require some timestamping standard, and formats differ — the guideline document is authoritative.

Transcription gear

The equipment baseline for paid transcription work: good closed-back headphones, quiet listening environment, reliable internet, text-expander software, and optionally a foot pedal. Gear requirements are verified and listed per platform — most general work needs nothing beyond a computer and headphones, while legal and medical transcription assume professional setups and reference materials..

Verbatim transcription

Transcription that captures every spoken word exactly as said, including filler words ("um", "uh"), false starts, and sometimes non-verbal sounds. Verbatim work pays more than clean-read transcription because it is slower and more demanding.

Community moderation

Managing the health of an online community: answering questions, enforcing rules, running events, and escalating policy violations — usually for a specific brand, game, or app. It blends customer support with light content moderation.

Companion chat

Conversational messaging work on social platforms where you role-play friendly or flirtatious personas for users paying per message. It is adult-adjacent work: 18+, emotionally demanding, and paid per message — around $0.35 per message is commonly cited.

Used by: companion chat platforms

Content moderation

Reviewing user-generated content — posts, images, videos, ads — against platform policy: removing violations, escalating edge cases, and documenting decisions. Professional moderation is employment-style work with training and scheduled shifts, and it carries documented psychological load; reputable contractors provide wellness support.

Per-message pay

The billing unit of companion and some support chat work: a fixed amount per message sent, often with minimum response-time requirements that effectively cap concurrent conversations. Per-message pay rewards responsiveness and persona consistency, not typing speed alone.

Trust and safety (T&S)

The professional discipline behind platform integrity: policy design, abuse investigation, fraud detection, and moderation operations. As paid work it spans frontline moderation queues, specialist escalation roles (fraud, self-harm, minor safety), and policy writing.