CloudresearchReviews of cloudresearch.com

here is my own formula i made to figure this all out. hope it helps! dont even need to care about their "email" BOOGERMANFORGENESIS Weighted Linear Combination (WLC) ,Weighted Additive Scoring Model (WASM) Microsoft Entra, Google Cloud Identity, AWS IAM classify accounts using anomaly detection, risk scoring, and weighted behavioral models. The Actual Math Behind the Score weighted risk/quality score, also known as a weighted linear model plus normalization. Raw Score Formula Q = w_s·S + w_f·F + w_a·A - w_r·R - w_t·T - w_c·C - w_ai·AI Where: Q = raw quality score S = successful submissions (rate) F = positive feedback score (avg rating) A = profile completeness / “About You” freshness R = rejected submissions (rate) T = time‑out submissions (rate) C = flagged conversations (per 100 submissions) AI = AI‑detected responses (rate) w_x = weight for each factor Normalization to 0–1000 raw score converted into visible score: Q_display = round( ((Q - Q_min) / (Q_max - Q_min)) · 1000 ) 449 is this normalized value, not raw internal score. Positive Factors (Scored as Rates) Successful submissions rate S = successful_submissions / total_submissions Positive feedback score F = average_feedback_rating Profile freshness A = 1 if “About You” updated recently A = 0 if stale or incomplete Positive score Q_positive = w_s·S + w_f·F + w_a·A Negative Factors (Penalties) Rejection rate R = rejected_submissions / total_submissions Time‑out rate T = timeouts / total_submissions Flagged conversations C = flags_per_100_submissions AI‑detected responses AI = ai_detected_responses / total_submissions Negative score Q_negative = w_r·R + w_t·T + w_c·C + w_ai·AI Full Raw Score Q = Q_positive - Q_negative Why Old Perfect Accounts Score Extremely High Example using stats: total = 6162 + 10 = 6172 success rate ≈ 6162 / 6172 ≈ 0.9984 rejection rate ≈ 10 / 6172 ≈ 0.0016 Example weights: w_s = 500 w_r = 800 Positive Q_positive ≈ 500 · 0.9984 ≈ 499.2 Negative Q_negative ≈ 800 · 0.0016 ≈ 1.28 Raw score Q ≈ 499.2 - 1.28 ≈ 497.92 Normalized → near the top end of the scale. why perfect accounts look “too good” to anomaly detectors. Access Level Bands Q_display < 300 → Restricted 300 ≤ Q_display < 600 → Limited 600 ≤ Q_display < 800 → Standard 800 ≤ Q_display ≤ 1000 → Priority 449 sits in the Restricted band. Why a Near‑Perfect Worker Can Still Land at 449 Example current pattern: S = 0.97 R = 0.03 T = 0.02 C = 0.01 AI = 0.02 F = 4.5 A = 1 Positive Q_positive = 500(0.97) + 150(4.5) + 50(1) Q_positive ≈ 1210 Negative Q_negative = 400(0.03) + 300(0.02) + 250(0.01) + 600(0.02) Q_negative ≈ 32.5 Raw Q ≈ 1177.5 If Q_min = 800 and Q_max = 2200: Q_display ≈ 270 Small changes in weights push into the 300–500 band → 449. How to Move the Score Up Negative weights are heavy. Small improvements matter. Examples: Drop AI from 0.02 → 0.00 Gain = 600 · 0.02 = 12 raw points Drop R from 0.03 → 0.01 Gain = 400 · 0.02 = 8 raw points Drop T from 0.02 → 0.01 Gain = 300 · 0.01 = 3 raw points Total gain = 23 raw points → easily +70 to +100 normalized. yours truly boogermanforgenesis!

I have been a participant for years. I had a high rating, only two rejections, and thousands of sucessful surveys completed. Now I am at 1 star and restricted. This is unacceptable. You should not treat your survey takers like this. I am not the only person dealing with this mess. Do better!

If I could rate it lower than one star I would. After having been a participant for years they arbitrarily instituted a rating system that has disenfranchised many, many long time participants. Effectively they banned us all by caveat. This site is no longer recommended, and should be avoided in favor of other similar sites.

Cloud Research is not transparent, mistreats participants, and can't really be relied upon or trusted. They recently mentioned punishing participants who return surveys (which violates consent and ethics regulations), then hid the infographic that displayed that information when challenged about the legalities of it. I would suggest other's look for work elsewhere. I have been using Amazon MTurk and Cloud Research for years now with no issues (2 rejections on MTruk due to inexperienced researchers errors, 0 on Cloud Research). I never use a VPN, Only use Dark Reader as a browser extension when doing surveys, never use Ai, always close other tabs and programs when instructed to, always enter full-screen when instructed to, have never had a disagreement with a researcher, have always been honest and accurate in my responses, read everything word-for-word, do longitudinal studies to the end, and provide long open-ended responses. Everything I can think of to be a good-quality participant. I used to have a perfect rating on the site 850, I have over 1500+ approvals (0 rejections), now my account is restricted at 384 overnight with no explanation other than "keep doing good work!". This is very untrustworthy for everyone involved. The participants can do everything correct, put in a lot of effort and time, and be told that they should keep giving good answers despite being told simultaneously that everything they do is bad for no specific reason. The researchers have no way of telling who is and isn't a trusted participant. Some people who get rejections regularly for poor quality work are now rated higher than people with a perfect record of following instructions. Any communication from them is either avoidant, manipulative, or blaming others for issues on the system they have complete control over. They will gaslight, lie, and hide whenever anything goes wrong. Even now while writing this review, I have to double-check there is no identifying information, because I fear the company will retaliate against me for providing an honest review here. You shouldn't have to be literally afraid of a company and hide from them as someone who does work for them. This company is creepy, like an abusive ex that you can't avoid because they are your manager, and they abuse their power and position over you.

I was a diligent user, even going as far as to contact researchers when there were issues or questions. Despite this, I have been given a 2 star rating and unable to do any surveys. I've been a user for 2+ years. So dissappointed. I reached out to their customer service multiple times via email. I also contacted them via Facebook and never got as much as a response or explanation.

The website was great for years until they implemented a broken rating system that has hindered my ability to get worthwhile paying projects. Look elsewhere for work. This is not a good platform.

My experience has been OKAY up until a few weeks ago. I was really appreciating your 17-25 cent studies because prolific was so slow. But I was wondering why I'm not getting ANY studies for a few days and noticed I suddenly have 1 star and a 435 credit score. This is complete bullshit and you do not appreciate us. You're treating us as disposable and degrading with the ridiculous amount you pay for the studies. And you add insult to injury with this shit. I'm glad everyone on Reddit is just as mad as I am and a lot of people are leaving as soon as they get paid. And I'm one of them. You would not exist without us!!! And you better believe it's going to be hard to get new people because they're going to read all these ratings and the Reddit stuff and say hell no!!!

Horrible experience. Cloud connect does not value loyalty and will just waitlist you for no reason