Showing results for 
Search instead for 
Did you mean: 

What if most of us make more money with JSS?

Ace Contributor
Karolis K Member Since: Sep 5, 2015
11 of 25

But that's the thing: why was this person upset with the score of "4" (which sounds like an excelent score really)? Because everyone else is getting 5s so if you get anything lower than 5 you stand out as being worse.

I am new here. I did 4 jobs in total. I got straight 5s in all of those. And I gave straight 5s to every client. Not all of these were deserved on both sides.

In my opinion the stars just don't work. Ratings should differenciate between freelancers. Not make everyone feel good about themselves.

Community Guru
Ronald T Member Since: Sep 14, 2009
12 of 25

Great thread Preston! You can usually spot an interesting topic by the length of the replies, and there are some long ones here already..


Preston wrote in part: "Let's say 90% of contractors end up earning more money...." That could happen; however, I doubt it will.


Actually, I doubt the change regarding the JS score and the average star rating will make much difference because there are far more significant problems with Upwork e.g., persistent core site performance issues, buggy core platform, inconsistent and conflicting practices, not exactly the best support, lack of effective follow-up on the part of management overseeing support and engineering (open issues not closed appropriately), low Community turn-out, Community features promised yet not implemented, apparent Community purpose not exactly in the best interest of members (except as an alternative for support), poor qualify jobs and questionable clients, overcrowding by low quality contractors....


Note: Did I mention "performance issues"? While writing this post I attempted to access the core site and was met with "We'll be right back. Promise.." The page meta title read: "An Error Occurred: Internal Server Error." Fortunately for me, I save anything from Upwork that I might need to my local drive using the Zotero add-on for Firefox. OK, the page is back, but the header with navigation links is Maybe this is another "Hot Update"; that is, "hot" as in online, untested, and without warning -- another bad practice. Oops, errored out again!


I must say in general that recent changes (adds, deletes, and modifications) to the core site layout, navigation, features, etc., show improved usability, data flow, and integration of information -- both on the client side and more recently the contractor side.


I still believe that they would have done much better developing Upwork from scratch with oDesk running separately and in parallel with a new Upwork beta platform. Yes, I know how much of an undertaking that would have been -- been there, done that many times with sites in enterprise level wide area networks.


Why do I I doubt the change regarding the JS score and the average star rating will make much difference? As a seasoned individual with decades of experience within high levels of information technology and quality assurance, I realize that systems for quality awareness, performance measurement, and similar initiatives must be well planned, implemented, and maintained using established and documented baselines. In other words, don't make it a practice to continually change the rules of the game -- integrity becomes questionable, and people loose interest in the gimmicks. Yes, the JS score has become just another gimmick.


Beginning 2012, as a result of numerous changes in practices, most of my work as an independent contractor was off-oDesk (Upwork). I would like to move clients to Upwork; however, at the moment that might be too risky. In July 2016, I will make my final decision regarding whether to return to Upwork full-time or not.


I have already successfully transitioned from working in the diverse fields of "Web, Mobile & Software Dev (Web Development)" and "Sales & Marketing (SEO - Search Engine Optimization)" to specializing in "Data Science & Analytics (Data Mining & Management, Data Extraction / ETL, Data Visualizatio, and more!)."


I really enjoy using KNIME, Tableau software and other tools used in the data sciences field and hope to share with the Community some of the results of my studies at a later date.


"It takes a lot of work just to work via Upwork."

Community Guru
Jennifer M Member Since: May 17, 2015
13 of 25

I'm with Ron. I couldn't give a rat's patootey about JSS. The slowness and daily site crashes and major bugs are what's costing me money and likely lots of freelancers and Upwork as well.


I don't want to jinx it but notifications have been working for a full week now. That's a record.

Community Guru
Ronald T Member Since: Sep 14, 2009
14 of 25

Hey Jennifer, do you use a thesaurus, or do those things just come naturally? "Patootey' !?!? I always look forward to your posts. They keep me laughing....

Community Guru
Jennifer M Member Since: May 17, 2015
15 of 25

lol Ron, I have a potty mouth that would get me banned in a second if I let it out, so I have to think of other words to keep myself out of trouble. "Patootey" sounds PG enough. 

Community Guru
Darrin O Member Since: Jan 20, 2015
16 of 25

The quoting system is a bear for posts this long, but here . . . we . . . go!  Preston H wrote:

And right off the bat I'm seeing yet another "bug" in how the system operates.  Smiley Sad

What if we all (or most of us, anyway) make more money with the change to Job Success Scores being shown at the top of profiles and star feedback only shown alongside individual jobs? Would you still be against it?

In the immortal words of philosopher Jayne Cobb: Yeah, and if wishes were horses, we'd all be eatin' steak.


I'm not going to engage in magical thinking.  I'm a scientist, so I will absolutely change my mind based on evidence.  The problem is that Upwork seems to never provide any actual evidence either before or after they set out to make these changes.  If you want to "What if" a scenario for their success, the burden is on you to specify a believable mechanism by which I'm going to make more money.

Mathematically speaking, there are only two possible future events with regards to the total amount of money earned by contractors:

a) it will increase

b) it will decrease

It's a zero-sum game, and earnings may bear very little correlation to JSS, let alone causation.  The total amount of money available is dependent on the Upwork platform as a whole.  You need to show the mechanism that will cause that to change before you even begin to talk about how individual freelancers will be affected.  As others have already noted, the biggest issue on this front is the negative affect that site errors are having.  The only way I can see JSS being a factor in that is as a misallocation of developers to work on adding/removing useless things rather than fixing core site issues.

Let's say 90% of contractors end up earning more money...

Let's not just say that.  Let's back it up with a mechanism for change.  Otherwise, all I'm expecting from a JSS-centric mechanism (I'm now down to 77% as of the latest update) is much less future work that I would have gotten from my still-perfect-5-star rating.

And let's say you're one of those people. You end up making 25% more money per month...

How?  Am I going to be hired 25% more?  Are clients going to paying 25% more?  What is the mechanism that fuels your fantasy?

The converse proposition here is that if moving to Job Success Score ONLY at the top of profile pages causes Upwork profits to DECREASE measurably... don't you think they'll go back to the way things are now?

No.  And it won't be their choice.


Community Guru
Preston H Member Since: Nov 24, 2014
17 of 25

Excellent posts all around on this subject... I appreciate the discussion and hope others find it interesting and maybe even helpful.


If I could clarify one thing:

In my original post I threw out some numbers (90% of contractors seeing a 25% increase in earnings). That was not meant in any way to be a prediction. I'm not predicting that will happen.


I just wanted to have numbers out there to be used to address the question of whether objection to the change to JSS-only in profile headings was primarily about concern about earnings, or concern about the score on moral grounds, a mixture of both, etc.


I probably did not communicate clearly enough that those are not meant to be realistic projections, or projections of any kind, but just a framework for thinking about why people object to the change, and what could get them to change their minds.


Also, a few people here have made an excellent point, in that if you're a contractor already working at peak capacity, there is no clear way that this change would increase (or impact) your earnings.


This makes a lot of sense. Realistically speaking, the change in JSS display is most likely to benefit contractors with high job success scores who are working less than they want to be working and have room to get hired. Logically speaking, the people this change would seem to help the *most* would be contractors with high job success scores and relatively low average star feedback ratings. This is precisely the objection that many commenters in other threads have mentioned.

Community Guru
Darrin O Member Since: Jan 20, 2015
18 of 25

@Preston H wrote:

In my original post I threw out some numbers (90% of contractors seeing a 25% increase in earnings). That was not meant in any way to be a prediction. I'm not predicting that will happen.


I just wanted to have numbers out there to be used to address the question of whether objection to the change to JSS-only in profile headings was primarily about concern about earnings, or concern about the score on moral grounds, a mixture of both, etc.

My objection is on a purely mathematical basis (in some circles, that counts as moral grounds Smiley Wink).  JSS, as implemented, exhibits all the same problems (skew, clustering, etc.) that the star system does, with the added insult of having secret information being a large part of the calculation.  A proper fix, such as normalization as I've mentioned before, could have been applied just as well to the star system.


And the fundamental math impacts any projection of success you'd like to make, whether it's 25% or 0.25%.  If you try to imagine a future where the rising tide raises all ships, you have to tell the story of how that is actually going to happen based on the change made.  I'm going from a 5 star rating to a 77% (and falling) JSS.  How do you suggest that is going to benefit me?  How do you think most people who are being clustered in the 4.8+ range are going to benefit when the result is that they're going to be (regardless of the specific calculation used) spread out over the lower side of a larger range of values?


Look at it from the client's point of view.  What Upwork used to offer is a bunch of nearly 5 star candidates.  Now they'll be getting a list (of the same people) who now arbitrarily have lower scores.  Is that going to inspire greater confidence?  Lead to more hires?  Lead to higher pay?  It just doesn't add up to what you suggest it might (most of us making more money).


Community Leader
Mark C Member Since: Aug 8, 2015
19 of 25

The very name Job Success Score is a misnomer, how can the fact that a client can't be bothered to leave you feedback reflect how successfully the job was carried out?


I also don't think that the many adverse comments about the JSS are against the principle of the JSS but against the secretive methods that are used in the calculation. 


Furthermore, there are freelancers on here who have seen big decreases in their JSS for not having feedback. on one of my closed contracts nobody has received feedback - 48 out of 48 contracts have no feedback. When I started this, no feedback didn't matter, but to then introduce a system that retrospectively punishes no feedback is not right.


Then there are freelancers that have a 100% JSS but have contracts that are over 6 months old with less than 1 hour worked. To me, this seems that the system is being selective, or not working correctly.


Either way, it is wrong that a metric that is secret and still flawed should be the primary way of rating freelancers.



Community Guru
Cairenn R Member Since: Aug 19, 2015
20 of 25

One of the problems with the rating system on Elance was that there were numbers, ranks, stats and levels all over the place - chaotic. I understand the love of getting stars and likes and sparkly things on profiles, but this is supposed to be a professional business platform, not Facebook. 


Keeping both systems would be a mistake. We're already experiencing enough chaos with the bugs and tech issues.


The JSS is here. We won't know for awhile if it works, or doesn't, or makes any difference at all, but sitting around crying about it definitely isn't going to make the score go up.