Cap applicants!

[ri3dviz]

It's rediculous, I get an email from Upwork with on average 15 jobs that might interest me but when I get the email there are 50 applicants already. Upwork you should cap applicants on job proposals at a reasonable number like 20. Then if the client hasn't found what they need in those 20 they can tick a box to allow more. I'm not going to apply to a job with 50 applicants.

[versailles]

---

Robert I wrote:

Upwork you should cap applicants (...)

I'm not going to apply to a job with 50 applicants.

---

Et voilà. Done.

Anything else?

[ri3dviz]

Am I suposed to respond to that, or you're here to be snarky? I don't want to read into your reply so maybe clarify what you want.

[lysis10]

---

Robert I wrote:

It's rediculous, I get an email from Upwork with on average 15 jobs that might interest me but when I get the email there are 50 applicants already. Upwork you should cap applicants on job proposals at a reasonable number like 20. Then if the client hasn't found what they need in those 20 they can tick a box to allow more. I'm not going to apply to a job with 50 applicants.

---

Yeah, so don't apply. I rarely apply to those jobs. That letter seems to be a list of jobs that maybe they think can be sold but the client isn't finding the right person. You need to see it from Upwork's side, not yours. The don't care *who* gets the job. They just want someone to get it so that they can get their fees. They obviously want someone with higher fees to get it, but this list seems to be a last resort to get it sold to anyone.

[ri3dviz]

I need to see it from Upworks side? You can't be serious, we're paying 20% of our fees, buying connects to apply and we need to see it from their side? 

Cap the applicants, have a button the client can use to accept more if they don't find a freelancer. That's not too much to ask.

Every time I post here I get someone defending Upworks policies, if it's not a two way street it's not a good thing. Now someone will say "so don't use Upwork." Which is my point.

[prestonhunter]

re: "I need to see it from Upwork's side? You can't be serious, we're paying 20% of our fees, buying connects to apply and we need to see it from their side?"

I have a better understanding of how Upwork works and why it does things it does because I am able to see things from Upwork's perspective.

Jennifer is not saying that you required to "accept" or "embrace" someone else's perspective. Nor is she saying that anyone can not advocate for change or put forth their ideas for how to improve things.

It's simply a fact that seeing things from Upwork's perspective lets us understand many aspects of the system which might otherwise seem goofy or counter-intuitive.

One of the most helpful tools in my money-making arsenal is the understanding that Upwork's customers are the clients - not the freelancers.

[ri3dviz]

Upworks money comes from the ability to have both clients and freelancers, I'm a business man and have been for nearly 40 years I'm not naive.

Tell me why a cap and and an ability to have the client allow more bids is a bad idea? Unless Upwork is just trying to maximize their dollars from people buying connests? Now they wouldn't do that would they? LOL

[lysis10]

---

Robert I wrote:

Upworks money comes from the ability to have both clients and freelancers, I'm a business man and have been for nearly 40 years I'm not naive.

Tell me why a cap and and an ability to have the client allow more bids is a bad idea? Unless Upwork is just trying to maximize their dollars from people buying connests? Now they wouldn't do that would they? LOL

---

They tried this and it was terrible. Their high-end moneymakers aren't sitting around waiting to bid. So you cut off the very freelancers you *want* to win jobs.

[ri3dviz]

I've been here since the beginning I never saw that implemented? Refresh my memory when did that happen?

[lysis10]

---

Robert I wrote:

I've been here since the beginning I never saw that implemented? Refresh my memory when did that happen?

---

Maybe about a year ago? It was announced in the announcement forum.

[alexandernovikov]

People are paying to apply. If they do, they think it's worth it. Only real fix for that is increasing the price of connects. Or better, improving the formula that determines how many each job needs to require to apply - a simple ML thing can do it, like it will figure that it's okay to charge 100 or even 500 connecs to apply to a full-time job offer from a client who's been on the platform for 10 years, paid 5 million dollars and paid $60 an hour average, but applying to an unverified, noob client for a one-off small fix can be entirely free...

[prestonhunter]

re: "...it's okay to charge 100 or even 500 connects to apply..."

Let's not give them any ideas.

[ri3dviz]

I'm also a client here, believe me I don't want to go through 50 proposals when I posts a job.

[prestonhunter]

That is why Upwork explained that they started charging for connects: to help maximize the quality of proposals that clients receive... and also to cut down on the number of proposals that clients receive.

For a very long time, connects were free. Upwork only started charging for them last year. Trying to make money by charging freelancers to apply to jobs was previously never part of their business model (obviously) because they weren't charging.

If Upwork now sees this as a potential additional source of revenue in addition to a way to help enhance the client experience, I wouldn't be surprised.

As for these jobs that we get messages about, asking us to apply... Others here have already explained that those are "bargain bin" jobs, being advertised to freelancers in the hope that some people will apply to yet-unfilled jobs. These are not great prospects, but there is little or no cost to Upwork to advertise these jobs to freelancers in a last-ditch effort to get some money out of them.

These lists of yet-unfilled jobs are being sent out primarily to help Upwork earn more money. Don't think of it as a benevolent service being provided to freelancers or clients.

Clients who act the way Upwork wants them to act - posting a job and hiring for that job relatively quickly, and then closing the job posting - do not have their jobs included in these batch messages.

[ri3dviz]

Well, in my field it has not worked there are 50 plus applicants on jobs open to the entire globe and over 20 nomally to US freelancers only. Just this week 4 jobs I applied to were shelved with no hires by the client - those cost me. Things are getting worse not better in my experience, phishing schemes by freelancers to see what others are charging, clients not posting enough info to bid and when you request it not reponding, and the list goes on.

[ri3dviz]

I gave you a solution, let the client determine if they want more applicants but set it to default of 20. The client is still in charge. I suspect it won't be implemented because Upwork wants freelancers to "buy" connects. I've been here since day one, I see how they slowly have creeped towards more profitability with no decernable advantage for freelancers.

[prestonhunter]

I think your suggestion (your solution) is an excellent idea.

It is an idea that would benefit both clients and freelancers.

If I was a decision-maker at Upwork, I might push your idea if I wanted to make the user experience at Upwork even better. I might squash your idea if I thought it would decrease revenue in the short term.

[ri3dviz]

I'm not sure how it would decrease revenue by much except the aforementioned buying connects. If buying connects was solely to have fewer applicants and more qualified ones it has not worked. But I suspect that is partially spin to sell the idea to freelancers. As I said I'm not naive.

[lysis10]

---

Robert I wrote:

I'm not sure how it would decrease revenue by much except the aforementioned buying connects. If buying connects was solely to have fewer applicants and more qualified ones it has not worked. But I suspect that is partially spin to sell the idea to freelancers. As I said I'm not naive.

---

No, I believe them when they say the point of paid connects is to reduce bids. Yes, they did try to spin it and the way they did wasn't all that far off, but yes Upwork has a way of spinning stuff sometimes.

Before paid connects, they were actively trying out different ways to reduce bids, so I think it makes sense that this was another effort to do it. Their problem is that the connects aren't expensive enough, but there are people here who say they can't afford connects so I guess they are killing off the extreme low end.

eta: I also think paid connects was (in theory) supposed to kill off people who can't sell a job so instead of the overhead of banning them, they just starve them out.

[ri3dviz]

Nothing has changed in my field, still over 50 applicants on global jobs and well over 20 on US only. I only bid on US jobs as I live there and can't compete with India, China etc. on fees.

[tlsanders]

---

Robert I wrote:

I'm not sure how it would decrease revenue by much except the aforementioned buying connects. If buying connects was solely to have fewer applicants and more qualified ones it has not worked. But I suspect that is partially spin to sell the idea to freelancers. As I said I'm not naive.

---

It would decrease revenue because the clients would leave after the initial flurry of garbage proposals and never see the high-end freelancers they would have been likely to hire. 

You seem to feel like someone is trying to "sell" you on something, but that's just silly. No one cares what you think, and there's no need to "sell" you. The system is what it is. Upwork has spent a lot of time and money and tested different things and this is what they picked. 

In any case, it appears that, despite your objections, you've been "sold," because you're here looking for jobs instead of taking or finding an approach that's more lucrative for you.

[alexandernovikov]

Problem with this solution is that it doesn't let the *best* 20 applicants to apply. But the 20 most desperate ones: those who can't get hired and go through jobs all the time and see the ad first. No way it helps client satisfaction: all he will see is 20 "please hire me", unqualified desperate cover letters in broken English. Neither does it help you: all you will see is a job closed for bidding, with 20 applications. Because you came in 5 minutes later.

[ri3dviz]

Having it open ended does not get you the 50 best applicants either so your argument doesn't hold up.

[prestonhunter]

There are many different ways to hire, and many different types of jobs.

I have been very pleased in my hiring experience as a client.

I often post a job, and hire the first person who applies, and then close the job.

I can post a job and hire somebody often within five or ten minutes.

So I don't experience "20 desperate freelancers" and I don't experience the need to go through 50 applications. I typically use a very fast hiring process.

Why is this?

Maybe because I post jobs with very specific details about the type of help I need, and these are technical in nature. So the people who end up seeing and applying to the jobs are people with specialized skills who know how to do the work.

When I want to hire quickly, I make this clear in the job posting. I point out that I want to hire someone who is ready to work "now" or "immediately." I may point out that I want to work with a freelancer using screen sharing, such as with TeamViewer. Only certain freelancers are looking for jobs like that. I find that the people I hire this way routinely look for this type of work. They are freelancers who already have the screensharing software ready and they are actively looking for immediate-hire short-term jobs.

I also post jobs for creative freelancers such as artists, illustrators, graphic designers. For these jobs I will allow more applicants to submit proposals before I close the job posting. I am able to look at their portfolios in order to make decisions.

[ri3dviz]

Okay I'm out, I expressed my views, I won't hold my breath for changes.

[yitwail]

Robert, Jennifer M is correct. It was announced one or two years ago that once a certain number — I don’t recall exactly how many — of top rated freelancers applied to a job, it would be removed from the job feed. I don’t recall any announcement that the policy has changed. Perhaps a moderator could update us on the status of this initiative.

[lysis10]

---

John K wrote:
Robert, Jennifer M is correct. It was announced one or two years ago that once a certain number — I don’t recall exactly how many — of top rated freelancers applied to a job, it would be removed from the job feed. I don’t recall any announcement that the policy has changed. Perhaps a moderator could update us on the status of this initiative.

---

I thought they announced in the thread that they yanked it? I think they have because I'm not getting jobs in my feed that I can't apply to just because too many applicants.

[wescowley]

Clients are already free to pause applicants at any point by putting the job on invite-only.  There's no need for an artificial cap, and putting one on takes control away from the clients and limits their applicant pool to the first random people who click submit. Let's not go there.

[tlsanders]

---

Robert I wrote:

It's rediculous, I get an email from Upwork with on average 15 jobs that might interest me but when I get the email there are 50 applicants already. Upwork you should cap applicants on job proposals at a reasonable number like 20. Then if the client hasn't found what they need in those 20 they can tick a box to allow more. I'm not going to apply to a job with 50 applicants.

---

I know I've probably said "This is the worst idea I've ever seen posted here" before, but this is definitely top three. 

Capping proposals means the client gets an initial flood of robo-bids or cut and paste proposals, mixed in with maybe a few personalized proposals from freelancers whose services are not in demand and so are sitting around refreshing the feed. Then, proposals close before freelancers who are busy serving clients ever see the posting, meaning that the chances the client will see a decent proposal are near zero.

Sure, the client could tick the box to see more proposals, but why would he/she? If I'd gotten 20 proposals and all or the vast majority of them were garbage, I'd just assume Upwork was a garbage site and move on.

[alexandernovikov]

Very true. In fact the whole "applying" thing is kind of moot. Most good freelancers mostly get jobs by invitation and they don't bother sifting through the feed at all (i wouldn't be - i just hired 2 people to do that for me and made them compete to fill me a Google Sheet with leads... this is the only reason i ever bid on jobs at all)

[tta192]

If a job has 50+ applicants you ignore it, simple. If applications were capped at 20 that same job you ignored previously would now be locked so you still wouldn't apply - what's the difference?

If the client is only interested in the first 20 applicants that information is already available to them - again, no difference.

If you want to be in the top 20 fastest applicants to a job you can do that too by bidding early.

In reality clients have their own criteria for determining the best candidate, and only need about 10 seconds per candidate to short list from 100 bids to 10 - 15. That's less than 20 minutes. 

No point in freezing the job post.

[prestonhunter]

A client who uses skill, or thoughtfulness, or has experience on the platform... or a combination of all of these... will be more successful in posting jobs and getting the response they are looking for.

I question how often any type of job is going to be worth going through 50 job proposals in order to hire somebody.

Wouldn't time be much better spent crafting a job proposal that will attract a small number of applicants, but they'll be right for the job?

Time could be better spent talking to a small number of applicants and hiring them and working with them on the actual project.

Letting a job proposal garner 50 applicants when one is planning to only hire one or two freelancers? That doesn't seem like an effective use of the client's time, much less the time and connects of the freelancers.

This morning I posted a job and hired the first freelancer who applied, and then set the job to private. So only ONE freelancer submitted a proposal and used his connects. This freelancer did a great job! 90 minutes later I closed the job. I got the help I needed. The freelancer had a good first-time Upwork work experience.

Effective use of Upwork can involve very few proposals.

50 proposals on a job makes me think the client is posting on multiple platforms and not serious about hiring on Upwork.

[tlsanders]

---

Preston H wrote:

Wouldn't time be much better spent crafting a job proposal that will attract a small number of applicants, but they'll be right for the job?

Of course. But, we're not all posting for jobs that require very specific technical skills. And, we're not all posting in areas where there's a demonstrated measure of whether or not the person is qualified to do the job.

When I have posted jobs for writers in the past, or calls for submissions, I have typically received about 200 responses within the first two days. The requirements are very detailed. In the case of soliciting submissions for a publication, the posting included the topics we accepted submissions on. When I advertised (actual example) for news-style opinion pieces on legal issues that yielded outrageous results, I got everything from love poems to humor columns to erotica in response.

In any given batch, fewer than 25% of applicants/submitters bore even a few seconds' consideration.

[prestonhunter]

Tiffany: excellent points. You have added a much needed perspective.

I wanted to be clear about what kind of hiring I was doing. Your observations about why the results are different make sense.

[tlsanders]

---

Alexander N wrote:

Very true. In fact the whole "applying" thing is kind of moot. Most good freelancers mostly get jobs by invitation and they don't bother sifting through the feed at all (i wouldn't be - i just hired 2 people to do that for me and made them compete to fill me a Google Sheet with leads... this is the only reason i ever bid on jobs at all)

---

I get a fair number of invitations, but I turn most of them down because they're not strictly within my niche (though perhaps related enough that I can see why the client invited me). Most of my good, ongoing Upwork clients have come from bidding.

[alexandernovikov]

In fact, the best solution will actually be some AI system that will calculate the cost to apply to each job with a goal function of having about the same (around 20 looks best) number of applications for each. It will probably never work perfectly but at least will keep applicants to between 5 and 50 with 20 being the average. And also analyze hiring rates on bids made platform-wise vs time from project being posted, with a goal function to make it a flat line, by adjusting cost to apply as a function of time, too. The algorithm will find by itself, but the way i  see it is making the price high in the first minutes (because that's when the copy-paste spammers are mostly doing the bidding), then gradually decrease to probably bottoming out some like 12-48 hours from project being posted, then rising again because of the likelihood that a client gave up on the project already/hired outside of Upwork.

A good idea will not be optimizing to a given number of applicants in each project, but to optimize to charging the same total amount for bids on every project, and replacing the initial 20% commission with these money (so people don't complain), i.e. charge ~$100 or so total (because now connects are also not free) for biddinng on every contract, counting only contracts that hire.

[alexandernovikov]

Come to think of it, what needs to be optimized, is the hiring rate. Basically a system can do an A/B testing of a sort (for example picking the cohort by last bits of a salted hash of a job posting id), to try bringing different randomly selected groups of projects to different average numbers of applicants, until reliable statistics with a high p-values can be obtained tell which one results in highest hire rate, then just try to pull all projects to that average number of applicants by varying cost to apply by both time from posting, and other project and client parameters. Keeping a small subset of projects in the A/B testing mode to make sure to react properly if market conditions change. 

No guesswork, just a normal machine learning task.

(my assumption is that there is some optimal number of applicants: make it too low and it means that bidding is too expensive/difficult and clients are missing out many qualified applicants, make it too high and clients are buried in spam)

This is the way to go. No emotional approach is needed, just science. A monetary barrier to apply should be just high enough to ensure the maximum hire rate.

[alexandernovikov]

One possible failure in my thinking is that this will optimize hiring rates, but not necessarily the amounts being charged. Upwork (and the freelance community either) isn't so interested in fulfilling the most jobs, but more in fulfilling them at the highest price possible...

[researchediting]

---

Alexander N wrote:

One possible failure in my thinking is that this will optimize hiring rates, but not necessarily the amounts being charged. Upwork (and the freelance community either) isn't so interested in fulfilling the most jobs, but more in fulfilling them at the highest price possible...

---

Thanks for sharing your thoughts, Alexander. I do not believe Upwork has ever invested a penny in AI/machine learning, and I would be surprised at their doing so. Their algorithms all seem to be constructed of flowcharts and surprisingly unintegrated lists. They have an enormous amount of data available to them, but seem unstrategic in mining it and unreliable in its interpretation. Hence the periodic policy lurches from on high, with IT and users scrambling to assimilate and adapt.