Gengo takes pride in its QA system armed with GoCheck and LS's. And we, translators, are evaluated with the GoCheck score in this system. Gengo has many advantages including a good support system - they are quick responsive and kind as well. Also I understand reviewing isn't an easy job so that I believe Gengo named them as specialists and I believe many of them will work as best as named. However, I don't feel as such in my pair.
I sincerely hope that this won't be the case in your language pairs — indeed.
One re-review was requested on Feb. 1. Two days later, one error was corrected but the other two errors were not corrected, where one error was explained with a grammar that I have never heard or seen before and the other one was simply repeating the same explanation in the original review with disregarding my claim. On the same day, I wrote to the support team another long e-mail explaining the correctness of my translation and requesting any reliable source of the grammar that was based on the error; this time I didn't request re-grading. On Feb. 8 - aside from some absurd communicating-, what I had was another score, with one error changing from major to minor with ANOTHER standard unlike the original review's standard, and with the other error not explaining the bizarre grammar and not providing any source of the bizarre grammar but claiming ANOTHER new theory. I requested the explanation again through the support team, and have not been hearing for a week.
This is just one instance showing review gradings without any clear standard, re-review results with the same repeating explanation in the original review or with non-rational explanation, and cutting off the communication. I had a lot more cases so I often feel like that this QA system wields their power from way over the ivory tower.
We work for clients and Gengo as well. However, if the re-review is simply repeating the same claim in the original review without explaining and resolving our arguments, if the re-review is creating a new grammar or theory for them to justify the score of the original review, if original reviews are graded without any standard criteria so two different jobs are reviewed and scored on a different basis but they don't give a clear instruction upon our request and give only reasons or only alterative sentences, if we, genuine, "independent" translators, should work for the non-interpretable scores, not for the benefits of clients, if the review is based on some intuitions and the re-review is justified on some bizarre theories and the quest for them is being completely disregarded in the environment that the score is all we are judged upon, where would we find democracy in this QA system?