I want to thank you Kashmiah for reaching out to the community.
It wouldn’t have been a task for which I volunteered were I a part of the Fiverr team.
I will assume you sincerely like the ARS or you wouldn’t have taken it up as you have. Especially since there isn’t a single concession in the entire post.
Let me state, I believe we’re in the middle of STARS WARS. There is no element of this system I like, no not one. However, I will reply to your post section by section in a how to vanquish the ARS Menace.
kashmiah said: We made an update to the rating system to ensure a better, more transparent community for buyers and sellers. We believe the advanced rating system allows sellers to get proper feedback for the entire Gig experience - the more accurate the ratings become, the more reliable the marketplace will be, separating the mediocre from the great and benefiting the entire community long-term.
There is not "greater transparency" where there are more things opaque.
Sellers only see a "star summary" and the review. The elements that went into that ranking is withheld. And of course, the worse part is...there's no transparency at all because we do not see the thinking of the buyer at the time of the rating.
We don't know if they were leery to give a high rating simply because of a poorly crafted question.
And of course, it's not "greater anything" to consider a "Satisfactory" rating to be "Negative Feedback".
The flaws in this logic are too numerous to list. However, the biggest presupposition is that you can KNOW good work before you PERSONALLY evaluate it after delivery.
We already have mutual cancellation and request modification to get to the cream and evaluate a provider. Why do we think people need to know "before hand"? It's a fools errand as even a MasterChef can forget an ingredient or serve a dish over or undercooked.
Again, we don't get "greater decisions making ability" by merely having more information...especially poorly generated information.
kashmiah saidIndividual ratings will not have an immediate affect on the overall stars- it's one rating factored into the overall average. As a result this won't change much for sellers who are already featured, in the top results and deliver high quality work.
This seems to be little more than a "Divide and conquer" admission that the system MOST HARMS new sellers.
Well, I for one remain in solidarity with all sellers (and buyers). I for one don't want a buyer who was just being a jerk to prevent a number of other buyers from seeing for themselves.
This is the STARS WARS -- not a civil war among sellers.
We don't want it no matter who it hurts most or first.
kashmiah said We've run tests prior to releasing the advanced rating system and there weren't any changes to the way ratings were left.
The tests are irrelevant because they were just that internal.
Now the incentives have completely changed. But beyond that, if there were tests...and if they were in fact good and compelling tests, you bungled the launch. You should have released the tests, got buy-in, established stakeholders, but NO...
...even TRSs had no early warning.
This is a real chance to learn from the launch error and if you have genuine improvements for buyers and sellers, apply these launch principles in the future. However, now, it's all fruit from the poison tree.
kashmiah said Along with the updates to the advanced rating system, we've also introduced buyer ratings as well. This will let sellers leave ratings for buyers as well, creating greater transparency between buyers and sellers on their overall experience. We monitor these ratings too, so we can weed out buyers who are causing sellers issues by trying to abuse the ratings system.
Again, even if this was a good thing. It's too little, too late on the launch. You lost us already. The STARS WARS are over. However, I'm not convinced it's a good thing either.
I as a SELLER don't know how to rate a buyer.
The Normal Distribution Curve tells us that 66% of all seller and buyer experiences should be right around 3 stars.
Another 13.6% should be 4 stars and 2 stars each. That leaves only 2.4% for 1 and 5 stars.
But real people are HORRIBLE at dealing with randomness, normal distributions, or ratings. So why would you have us try to do what we're horrible at.
I bet your tests do not have 66% of people with about a 3-star ratings.
And if they don't, why do you think it will identify the "mediocrity". Or even that there is such "Mediocrity" that needs to be found?
kashmiah said We want to reiterate that during our test, we didn't notice any change in how ratings were left for sellers, so it shouldn't change sellers who are already doing great work on Fiverr.
I'm not sure I even understand what you're saying.
However, maybe your test wasn't large enough?
Because we have plenty of folks getting very strange results ... "Love it, excellent seller" in comments and a 1-star review.
I don't understand how you can think the Fiverr system/community is immune to human nature?