Reply
cancel
Showing results for 
Search instead for 
Did you mean: 
Reply

Re: Open Jobs Trending Down - Upwork Statistics

makeyourmark
Ace Contributor
Mark G Member Since: Jan 25, 2015
41 of 75

Great work Setu! Very interesting stats.

 

I've recently become self-employed full-time and I'm already starting to worry whether it's completely the wrong time. Finding quality jobs on here has become a struggle in recent weeks. That may just be a lull during the summer but I can't be sure.

 

What's frustrating is the controversial credits system. I feel like I'm wasting credits because the majority of jobs I apply for are just remaining open - they're not hiring. Don't get me started on the rest of the jobs with the clients who seem to think it's okay to pay $5 for what is essentially 10 hours work. Unfortunately, they're only encouraged to continue in this fashion because some people continue to apply for them.

setumonroe
Community Guru
Setu M Member Since: Jan 26, 2014
42 of 75

@ Scott,

 

Yes. March, April and June would have been nice but the data is not available, remember the data cannot be plucked from thin air. This is the nature of baseline data, it has to start somewhere and someone has to decide to do the dirty work. I decided to do it, to provide it for you guys, and I am not being paid. Long term data is also useful, but once again I am not being paid. I am doing this purely because I am a scientist by default.

 

Correlation does not equal causation. Therefore traffic does not equal number of posted jobs. So many other factors could be the reason. If I decide to collect monthly data on "open jobs," I will continue to share it. But there is no doubt that there has been a drop in jobs posted, between last year and this year.

 

@ Mark,

 

I have something in the pipeline for assessing "actual number of hires" but it may be a bit skewed because I was defeated by Upwork's system initially. So for my interest, I may run another test concurrently.

 

On a different note, at least more people are taking interest in the data and what it means.

---- easy like Sunday morning ----
stencil_media
Community Guru
Scott E Member Since: Jul 26, 2015
43 of 75

"Correlation does not equal causation. Therefore traffic does not equal number of posted jobs."

 

Absolutely. I was a Six Sigma wizard in a previous life, but traffic is going to be a factor at some point though. 

 

"March, April and June would have been nice but the data is not available"

 

I didn't realise that's all the data you had, I thought you were just picking the highlights!

 

"But there is no doubt that there has been a drop in jobs posted, between last year and this year."

 

I can't buy in to that. Not yet anyway. I know stock footage sales are slightly different, but I've seen year on year increases, and the sales for June/July and August can be half of what February usually is. If we had a February 2015 data point that also showed a drop then I'd be happy to come onboard. 

 

The data is appreciated though and it would be nice to keep an eye on the stats. If I was a gambling man, which I am, then I;d say September and October will see quite an increase. 

"Welcome, humans. I'm ready for you!"
- Box, Logan's Run (1976)
setumonroe
Community Guru
Setu M Member Since: Jan 26, 2014
44 of 75

Absolutely. I was a Six Sigma wizard in a previous life, ...

 

Sorry for the late reply, I was busy 'googling' Six Sigma

 

I didn't realise that's all the data you had, I thought you were just picking the highlights!

 

I have been following the data for quite a while, but only recently started logging it, since the community was in such an uproar about "no hires." 

 

I can't buy in to that. Not yet anyway..... If we had a February 2015 data point that also showed a drop then I'd be happy to come onboard. 

 

I can appreciate that.

 

I;d say September and October will see quite an increase. 

 

Undoubtedly it will, because there will be two new variables introduced into the equation.

  1. Elance Clients
  2. Elance Freelancers

Therefore you couldn't prove an upswing just due to seasons, any more than I could prove a decrease because of uneven timeframes.

---- easy like Sunday morning ----
setumonroe
Community Guru
Setu M Member Since: Jan 26, 2014
45 of 75

So to be organized and have the information clearly disseminated, the assessment is in 4 areas:

  1. long-term stats on number of open jobs
  2. long-term stats on number of active freelancers
  3. short term stats on actual number of posted jobs (1 month)
  4. short term stats on "actual posted jobs" (percentage hires from 500 posts after 1 month)

I have been very busy but I will catch up on two of these things now (freelancer numbers and open jobs over time).

The data on "no hires" jobs have been collected, but no time to put it together as yet. Later this week hopefully.

 

Picture1.jpg

Picture3.jpgPicture2.jpg

 

Number of freelancers is clearly increasing, while job posts are on the rebound. I assume both these changes are due to influx of clients and freelancers alike. Next months data will make it clear.

---- easy like Sunday morning ----
jsutherland
Community Guru
Jean S Member Since: Oct 22, 2007
46 of 75

Thanks Setu.

 

I really appreciate the time and effort you put in to get these numbers for everyone to see.

setumonroe
Community Guru
Setu M Member Since: Jan 26, 2014
47 of 75

This is the Terminal Data on the "Open Projects" "No Hires" problem.

 

The aim was to shed some light on if and why clients are not hiring. All has been explained before so I will just provide the data and any thoughts.

 

After 30 Days

  • 192 or 38.4% of 500 posts remained open
  • 54 of the 192 had hires, so really 138 of the 500 remained open
  • 257 or 51.4% of 500 posts hired a contractor, therefore 48.6% hired no one
  • Verified clients accounted for 136 or 70.8% of posts left open
  • Verified clients accounted for 257 or 100% of 257 hires
  • 'Unverified' clients accounted for 87 or 35.8% of 243 'No Hires'
  • Verified clients conducted interviews on 62 or 40% of 155 'No Hire' posts
  • Only 1 post had no applicants
  • Posts for verified clients had and average 25.7 applicants vs unverified at 16 applicants

The most striking result is that verified clients account for 100% hires. Not all clients are verified at the beginning, but all that hire are verified.

 

The second thing that stands out is that only half of projects posted result in anyone being hired. While unverified clients make up 17.4% of project posts, they account for a third of the 'no hires.'

 

The last consideration in my mind is the quality of freelancers. Using verified clients, 40% of their projects posted conducted interviews but did not hire. It cannot be said that these are 'scam' clients, since most have a history of hiring and actually conducted interviews. It could be poor freelancer suggestion, or unqualified applicants, or even test runs. However this is food for thought on this matter.

---- easy like Sunday morning ----
pandoraharper
Community Guru
Pandora H Member Since: May 11, 2010
48 of 75

@Setu M wrote:

This is the Terminal Data on the "Open Projects" "No Hires" problem.

 

<snip>

 Mucho thanks Setu...

 

<gives Setu milk and cookies>

setumonroe
Community Guru
Setu M Member Since: Jan 26, 2014
49 of 75

@Pandora H wrote:

<gives Setu milk and cookies>


My favourite. Thanks Pandora 🙂

---- easy like Sunday morning ----
petra_r
Community Guru
Petra R Member Since: Aug 3, 2011
50 of 75

@Setu M wrote:

The most striking result is that verified clients account for 100% hires.


 That is because they physically can not hire without a verified payment method on file 😉

 

Setu, I don't understand all of it (yet, but I will) but I am just in awe of you taking the time and care to not only crunch the data for us, but to present it and explain it for us.

 

Thanks Setu!

 

 

TOP SOLUTION AUTHORS
TOP KUDOED MEMBERS