Connects get out of hand.

[Richard L]

Jure,

Unresponsive clients is also a huge problem because somehow freelancers get dinged for it through the mysterious "success score" algorithm. I've put in that complaint several times only to see it marked as "resolved." I end up having to hound clients for reviews -- which is something Upwork should enforce on their end. Instead, I appear like a nag.

Yeah, and I'm not sure how many times clients just disappear. It gets better the more I charge, but I wonder when I'm going to price myself out.

[Petra R]

---

Richard L wrote:

Unresponsive clients is also a huge problem because somehow freelancers get dinged for it through the mysterious "success score" algorithm.

---

No, they are not. Urban legend.

---

Richard L wrote:

 I end up having to hound clients for reviews --

---

You don't have to do that at all. As long as some money was at some point paid, just close the contrat if the client goes awol. As long as you don't have an insane percentage of "no feedback" clients, it has **NO** effect on your JSS in any way, shape or form. I just saw a profile of a freelancer with 85% of no feedback contracts and 100% JSS. I

[Richard L]

Petra,

Then go tell the technical support person who told me that I had too many clients with no feedback and it was hurting my score that you are right and she was not.

If you know the JSS algorithm, please make it public, because Upwork won't. You are making stuff up!

[Douglas Michael M]

---

Richard L wrote:

Petra,

Then go tell the technical support person who told me that I had too many clients with no feedback and it was hurting my score that you are right and she was not.

If you know the JSS algorithm, please make it public, because Upwork won't. You are making stuff up!

---

Petra does not "mak[e] stuff up." She is more knowledgeable about the workings of Upwork than many Upwork employees. It is repeatedly reported here that frontline CSS gets things wrong, misunderstands their canned answers, and even, yes, makes stuff up.

There's not that much to know about the JSS algorithm. It rewards good work as measured by customer satisfaction. It's all but a direct function of the My Stats "Clients Who Would Recommend" metric.

[Richard L]

"Who the clients recommend" is totally false.  Interesting that when I
followed the advice of the technical support person who you claim may
not know what they are talking about, my score rose back up to where I'd
expect it to be.

The algorithm is not simple. If it were it would be easy to predict. It
isn't.

[Douglas Michael M]

---

Richard L wrote:
"Who the clients recommend" is totally false. 

I point you to the one simple and useful metric Upwork provides as a measure of our success, as reported by clients. But it's "false." Well, OK, then. As you wish.

Interesting that when I
followed the advice of the technical support person who you claim may
not know what they are talking about, my score rose back up to where I'd
expect it to be.

Correlation is a necessary but not sufficient condition for causation.

The algorithm is not simple. If it were it would be easy to predict. It
isn't.

---

The JSS algorithm, as it's been variously described by Upwork, is not simple. "Simplicity" is neither a necessary nor a sufficient condition for its being fairly easy to predict—which it is.

[Richard L]

"I point you to the one simple and useful metric Upwork provides as a
measure of our success, as reported by clients. But it's "false." "

Yes, false. Because if it were true, the two would be equal, and they
are not.

[Richard L]

Douglas,

The claim made was that the success score matched one other parameter -- whether the client would recommend you or not. I said it wasn't true and that there were various things included in calculating the algorithm and Upwork was intentionally keeping the factors 'secret.' There is no place where I found that the algorithm is described, and live support refused to.

You arguing my point, not against me. And why everyone on this forum seems to feel flogging all other opinions is chichi is really odd if this is truly a progressive and friendly community.

[Douglas Michael M]

---

Richard L wrote:

Douglas,

The claim made was that the success score matched one other parameter -- whether the client would recommend you or not. I said it wasn't true and that there were various things included in calculating the algorithm and Upwork was intentionally keeping the factors 'secret.' There is no place where I found that the algorithm is described, and live support refused to.

You arguing my point, not against me. And why everyone on this forum seems to feel flogging all other opinions is chichi is really odd if this is truly a progressive and friendly community.

---

Richard,

The only person I have ever heard make the claim of identity between Clients Who Would Recommend and JSS is you. The complicated JSS is roughly predictable from and has a correspondence with Clients Who Would Recommend. That has been stated many times in many threads. That is not identity.

You didn't say the identity was false—in fact you subsequently asserted it out of the blue—you said the Clients Who Would Recommend—the one piece of hard, client-driven data you seem to be seeking—is false. When a person claims data are false, or a fact is false, that pretty much ends the conversation.

I'm not arguing anything. I, like others, am trying to point out facts, and discrete (and patterned) bits of information that Upwork has published over the years, and practices that it has modified or clarified over the years. That's not an argument. It's information. If you choose to reject it all, and take something an underpaid, undertrained, underinformed phone or chat worker told you as gospel, and take disagreement and correction as personal attack, so be it.

It would be nice, though, if you stopped putting words in other people's mouths, and misquoting even yourself.

[Richard L]

"The only person I have ever heard make the claim of identity between Clients Who Would Recommend and JSS is you. "

So your job success score correlates exactly with the percent who would recommend? THAT is what someone claimed, and I can tell you for a fact that isn't true.

"That is not identity."

What does that mean?

"You didn't say the identity was false"

I have seriously no idea what you are talking about at this point.

"you said the Clients Who Would Recommend—the one piece of hard, client-driven data you seem to be seeking—is false."

No. What I said was that JSS and clients who would recommend are clearly not a one-to-one correlation. Upwork says as much. Whomever stated that they were a one-to-one correlation was making a false statement.

"When a person claims data are false, or a fact is false, that pretty much ends the conversation."

Well...For argument's sake, if I say the moon has a chilling effect and moonlight cools water and you say "that's not right!" your sentence suggests the conversation is over. Exactly what was the outcome there? If you can just declare that I'm wrong without either having context or quoting correctly, it seems like having the ability to say "nuh-uh!" and walking away makes you correct.

"I'm not arguing anything."

Perhaps we need a definition of 'argument' like **Edited for Community Guidelines**

"If you choose to reject it all, and take something an underpaid, undertrained, underinformed phone or chat worker told you as gospel, and take disagreement and correction as personal attack, so be it."

I'm not sure what you are talking about at this point as you yourself typed words very similar to "It [the calculation] is not simple."

"It would be nice, though, if you stopped putting words in other people's mouths, and misquoting even yourself."

Um...I'm pretty sure you aren't reading your own posts before clicking 'send'.

0 Kudos

ReplyAccept As Solution

[Richard L]

"The only person I have ever heard make the claim of identity between Clients Who Would Recommend and JSS is you. "

So your job success score correlates exactly with the percent who would recommend? THAT is what someone claimed, and I can tell you for a fact that isn't true.

"That is not identity."

What does that mean?

"You didn't say the identity was false"

I have seriously no idea what you are talking about at this point.

"you said the Clients Who Would Recommend—the one piece of hard, client-driven data you seem to be seeking—is false."

No. What I said was that JSS and clients who would recommend are clearly not a one-to-one correlation. Upwork says as much. Whomever stated that they were a one-to-one correlation was making a false statement.

"When a person claims data are false, or a fact is false, that pretty much ends the conversation."

Well...For argument's sake, if I say the moon has a chilling effect and moonlight cools water and you say "that's not right!" your sentence suggests the conversation is over. Exactly what was the outcome there? If you can just declare that I'm wrong without either having context or quoting correctly, it seems like having the ability to say "nuh-uh!" and walking away makes you correct.

"I'm not arguing anything."

Perhaps we need a definition of 'argument' like **Edited for Community Guidelines**

"If you choose to reject it all, and take something an underpaid, undertrained, underinformed phone or chat worker told you as gospel, and take disagreement and correction as personal attack, so be it."

I'm not sure what you are talking about at this point as you yourself typed words very similar to "It [the calculation] is not simple."

"It would be nice, though, if you stopped putting words in other people's mouths, and misquoting even yourself."

Um...I'm pretty sure you aren't reading your own posts before clicking 'send'.