Bad Math

[matthewschenker]

Hi Everyone,

There have been many discussions about real and fake jobs, and the whole process of bidding and boosting. In some of these discussions, I have posted my view that Upwork makes more money on freelancers attempting to get jobs than actually getting jobs, I have argued that this situation is ripe for abuse.

In some discussions, people have disagreed with me and said that Upwork does not make more money from attempts. In this discussion, I want to offer my arguments for why it is obvious that Upwork does make more money from attempts.

The math is pretty simple actually.

Take a typical job, with some modest assumptions for the sake of argument:
- $500 offer
- 50 proposals submitted

- 16 connects needed to apply for the job
- 20 proposals "boosted" with an average of 5 connects spent on boosting

There is only going to be 1 hire for this job, no matter how many freelancers apply. That means Upwork earns $50 commission after the freelancer is hired.

Now let's look at the rest of the picture, which is the attemps to get hired:

- $120 from submitted proposals ($0.15/connect x 16 connects x 50 proposals)
- $15 from boosting ($0.15/connect x 5 connects x 20 proposals)
- $135 total earned from attempts

Upwork earns $135 from attempts and $50 from actual hire. That means Upwork earns $85 more from attempts.

This is a modest example.

Let's go a bit further. Prices for jobs is driving downward, meaning that Upwork is making less and less from commissions. Which means that they are making more and more on attempts as time goes forward. It's simple corporate logic to say that they would build a profit system around attempts instead of hires.

Consider the implications of this.

Anyone trying to tell me that Upwork does not make more money on attempts must offer a counter argument, based on numbers, not feelings.

As long as Upwork is making more money on attempts than hires, we have a problem.

[natashaabadash]

Oh Ely... hell no.

Anyways, I think I've found it here.

RIP UW. 

[spectralua]

Yep, this one.

[natashaabadash]

It's a featured job post fee. Just realised that. It shouldn't be alarming, and it shouldn't work by suppressing new applications and/or unlisting the job from the marketplace. So, nothing new basically. 

[spectralua]

Shouldn't, but look like suppress. Who know? Except Upwork. 😊

I wont be surprized when some fired Upwork team member will share all cheating going here. Many strange things happening after Upwork changed management 2 years ago.

My proposals i paid for should be shown to client. No: must be shown. But won't. And such topics fully explaining why.

[natashaabadash]

100%. I believe too that UW with their algo messes up all job posts and proposals you didn't pay extra. Actually no, they mess up those paid extra as well... 

[yofazza]

If a proposal is pushed out of the top-4 boosted position without the client noticing, connects are returned. If the client has noticed it, promise fulfilled, connects are not returned.

Just letting you know that they have thought about many things thoroughly so they can counter all arguments about their 'cheating'.

We can stop joining their game unless we know we can still profit on it. Just don't be the one who is gamed.

[nyonomega0]

---

 wrote:

We also need a thumbs down vote icon!

---

That wouldn't help and only move things in an unhealthy direction.

[2297e2bc]

I dissagree. Negative balances positive!  

Hot water balances cold do you can shower!

[2297e2bc]

4 thumbs up......Oh wait, we dont have a analog star rating system.

Just digital True/False

[yofazza]

It's like this: if you have a profitable, legal, but morally questionable business, and then you suddenly become a good person, you'd rather close the business instead of replacing it with something that can only incur losses.

Upwork operated at a loss for nearly a decade, and they just experienced profitable months after doing "some changes". They will not return to the previous failed model, nor will they care about people wanting "fairness for freelancers".

Assuming this current model is really unsustainable, it's better to stop participating in their game. The more people do it, the faster the "saturation point" will be reached, where Upwork will reveal their next/contingency plan, where I'm kind of curious to see it.

[2297e2bc]

As I say - if you dnt like your partner, leave the dance hall!

[atlinguist]

'...build a profit system around attempts'

-> Yes, that's exactly what it would be in a buyers' market. 

But I would rather have the freedom to spend my connects as I see fit than have to shell out for a plan or subscription that I would not be able to 'use' with money I could not recoup. Call it timid investing if you like, but my money has to go into my work tools and education first and foremost, and advertising comes second.

[matthewschenker]

Alexandra H,

But wouldn't it be even better if "spending connects" was not a factor? Wouldn't it be better if Upwork made money only when we made money? That would give you ultimate freedom.

[feed_my_eyes]

---

 wrote:

Alexandra H,

But wouldn't it be even better if "spending connects" was not a factor? Wouldn't it be better if Upwork made money only when we made money?

---

And what about the 90% of freelancers who don't make any money? Why do you think that Upwork should provide free services to people who contribute little or nothing in fees?  

[natashaabadash]

And what about the 90% of clients who don't give any money? Why do you think that Upwork should provide free services to people who contribute little or nothing in fees? Oh wait...

[feed_my_eyes]

---

 wrote:

And what about the 90% of clients who don't give any money? Why do you think that Upwork should provide free services to people who contribute little or nothing in fees? 

---

For the simple reason that Upwork needs more clients, whereas they don't need more freelancers. If the higher prices cause unsuccessful freelancers to give up and leave, that would be a good thing, not a bad thing.

[alexandernovikov]

Well it does need more freelancers - it's very hard to hire competent people here. But what it doesn't need is the yeah, 90% freelancers (and probably not many of the next 9% - let's face it, under 1% of people here consistently find good volumes of work sufficient to sustain them). It needs to try to actively get rid of those masses of people because they create clutter and put off clients - just try creating a project and see how much of (well-boosted) spam you'll get.

One way to do so is by creating conditions for making boosting unaffordable to them so they declare "upwork sucks" and give up. I see nothing wrong with that.

[the-right-writer]

The absolute last thing this platform needs is more freelancers. Do you want 30 million? Or 40? Is it not difficult enough to find decent clients who are not disgusted with the enormous numbers of irrelevant and garbage proposals, burying the few good ones?

It is not hard to find competent people here. If you throw out a job, and hire whoever comes along, then that is your error as a client. There are as many scammer freelancers as scammer clients. If you vet the freelancer, have a chat and a video, set up the contract with limited hours/milestones until the freelancer is proven, and pay a fee that will buy quality - there is no problem. Yes, many great freelancers have left, and more are leaving, but there are quality freelancers who remain.

If Upwork wanted to reduce the number of freelancers, they would reinstate all of the rules and limits and tests they threw away, because they were "too difficult." They are actively advertising for new freelancers with the, come make money with no skills, mantra. Upwork sees something wrong with raising the connects too high - they would lose income.

[matthewschenker]

You are blaming the people looking for work and paying Upwork for this platform? Upwork is the one responsible for that.

And as for clients, they are the ones racing this whole concept to the bottom by only looking for freelancers willing to work for abusively low rates.

Here's another idea: maybe make this platform based on hiring, not fees. Maybe have some standards for what work in various fields is actually worth.

[2297e2bc]

Good suggestion - and logial too!

We need a USER workgroup to march  on down to the Principals office and get some reaction.

[the-right-writer]

You are blaming the people looking for work and paying Upwork for this platform? Upwork is the one responsible for that.

No, Upwork is not responsible. While they can throw ads all over, only the people that don't understand freelancing, and how to work, will throw the enormous amounts of connects. No one should apply for jobs they aren't qualified for, and never boost unless you know you are one of the best.

Freelancers will complain they boost and boost and are never hired. Then you look at the jobs, and half of them are for $5.00, they have an incomplete profile, and no skills. And no vetting of the client. While this certainly doesn't cover everyone, it definitely covers a segment of the Upwork freelancer population.

And as for clients, they are the ones racing this whole concept to the bottom by only looking for freelancers willing to work for abusively low rates.

The clients can't do what you accuse them of, without the volunteer assistance from freelancers that don't use the platform properly. Why do freelancers break best practices by going for the cheapest of jobs? Why do freelancers volunteer for abusively low wages? Because some "guru" tells them that's how you build a career. Or Upwork tells them. So?

Here's another idea: maybe make this platform based on hiring, not fees. Maybe have some standards for what work in various fields is actually worth.

Upwork removed all limits and all testing for every aspect of the platform. They literally let in anyone and everyone, regardless of skills, experience, or history. Does that sound like they give a fig about quality? They purposely removed any and all barriers, to allow everyone in the world to be a freelancer. Standards are not of interest to Upwork. So, the solution? As I said, earlier, the freelancers need to stop doing the same thing over and over, while expecting a different outcome. Stop applying and boosting, unless you have done the work. Freelancers do have power here, but they prefer to continue to do the same thing, over and over, and then blame Upwork.

[2297e2bc]

What we need is a hoydy-toydy site that only invites elite members to join. No more of this open the flood gate metality and "prove yourself" to the world startegies.
Segregate the diciplines and only include "the best", and don't be open to evey bum on the street.
Yes-In my case, I'm talking to all the vidiots that jump out  YouTube and think they are a pro!
Pro means PROFESSIONAL, and by definition equals ones livelyhood - not side hustle!

So one metric of our model is duration at your trade!   I have 30+ years in! How about you?

I know---unfare to the young ones. OH then, prove participation in a school or university specialized course for your trade. I started with TV cameras in 9th grade HS, plus a university training for media, and also trade school for Radion/TV engineering! How about you?

Now, as far as payment to web owners (UW). for most of my carrer I paid DUES. A proportion of my saleries. BUT also, the Union put some of that away for a pention for me!!!!!!

NOTE: Hey UW, I'm not starting a protest here...just slugging out ideas that you may consider adopting! (So dont blow off my post)

All of these issues can redily be compared to AZon , E-boy and others.
Seriously it is time for UW to update the game. Without that, distrust , depletion and EOL will occur.

See whare I'm going with this folks? 

[matthewschenker]

Wow, the volume of illogical posts is amazing.

So your idea is that all the boosting creates "clutter" and "spam," and the way to reduce that clutter is to... have more boosting?

[2297e2bc]

Obviously the volume of posts logocal or not  matters  little to management!

[natashaabadash]

And why did they open the flood gates then? Now everyone can join and become freelancer. I have a feeling you are going in cicles... 

[alexandernovikov]

Exactly to sift through people to pick the tiny number of those who can get jobs!

That's how it works:

- let everyone try. because bidding is no longer free, they won't be bidding in truly insane amounts and won't create all that much clutter.

- let those few who can win, win

- make sure those who stand no chance, give up, if simply for running out of money

I see it as the only workable plan and it seems to be working. Unless GSV starts falling, which it probably won't.

[natashaabadash]

Alexander, what jobs? To try what? What is left to win? What chance? When was the last time you actively looked for a new project here, when was the last time you actualy looked through marketplace offerings or applied for new project with or without boosting system? 

[alexandernovikov]

I'm not looking for projects but continuously getting invites, some of the offers i'm discussing but in general there's too much incoming from non-Upwork sources to take up any Upwork ones, for now. Also i see plenty of relevant projects in my feed. I don't think anything has changed or there's less work. Upwork GSV stats are also worst case, flat - they aren't falling.

What changed is inequality. It appears that the competitive connects bidding to get visible solved the problem of freelancer spam and thus inequality in earnings now reflects actual value - which means putting vast majority of freelancers irrecoverably in red and directing most of the earnings to maybe top 1000-2000 most successful freelancers/agencies (say those who make over $500K-1M a year). With many millions of freelancers, from a standpoint of an average one, it will naturally look like "all real work has completely disappeared".

[natashaabadash]

Oh wow ok so you have no personal experience with the marketplace lately, but sure, you know everything about it naturally. Also, boosting system contributed to solving the spamming issues and served as freelancer quality filter and yet they need more freelancers because it is hard to hire good ones, and not to mention "just try creating a project and see how much of (well-boosted) spam you'll get."? All your talking points. I just feel like they are pretty inconsistent and majorly contradictory. 

[matthewschenker]

What a strange comment. What you are calling "unsuccessful" is often talented people who just don't want to pay lots of money to gamble on all the fake jobs.

Or maybe you define "unsuccessful" differently?

[matthewschenker]

Actually, yes. Upwork should be making their money by providing a platform for freelancers to be hired and for employers to find freelancers.

Your question assumes that Upwork should be making money on fees, rather than doing their stated job.

[natashaabadash]

I'm not sure where you got that assumption from, but I kind of agree with it. I don't see anything wrong with making a profit mostly from fees (is there really any other way?) - the question is how are those fees distributed among all customers and WHEN are those fees charged? And because UW charges many fees, unreasonably high (connects) and they charge them mostly before their work is actually successfully done, that tells me they failed their purpose (they can't make money along with us making money), so the only thing left to be done is to milk their customers dry before they are all gone. 

[atreglia]

---

 Christine A wrote:

And what about the 90% of freelancers who don't make any money? Why do you think that Upwork should provide free services to people who contribute little or nothing in fees?  

---

Good point, but from what I can see Upwork is no longer providing free services to that 90% because they have placed that burden squarely on the remaining 10%.  That said, I'm curious to understand why you think that arrangement is OK.

[marc_compte]

I wouldn't call it bad math. Your math is pretty simple and sound. That is the problem I see, it is an oversimplification of the entire universe of jobs and proposals. Drawing conclusions from an oversimplification is never wise. With so many assumptions and reductionism there is plenty of room for error and for argumentation against the main thesis.

But besides that, the real issue I see is not maths, but logic. Let's assume you are correct in both your assertions:

Assertion 1. UW earns more in connects than fees.

Assertion 2. UW is not interested in us getting hired.

The thesis that 2 happens because of 1 is not a valid thesis, in my opinion.

To explain assertion 2 based on an analysis of connects and fees, what you should be comparing is the revenue they get from a job with a hire compared to a job without a hire. Not connects with fees as if they were alternate sources of income, they are not.

To exemplify, taking your assumptions, a job without a hire earns them $135, whereas a job with a hire wil earn them that much plus the fee, totalling $185.

The difference between one and the other is that the job with a hire will probably imply one less proposal in the next job post (because someone is busy working and paying fees). So they will loose $3.15 there ($2.4 from the proposal plus $0.75 from the boosting, assuming the one getting hired was one of the regular boosters). But those are largely compensated by the $50 fee you are paying. The net gain is $46.85.

To make the jobs without hire more profitable than the jobs with a hire, a $500 job would need to keep me from submitting 16 proposals. I'm not sure that is realistic.

Let's looks at two other extreme cases to see another perspective:

A. Let's say I get jobs (fixed or hourly) that keep me busy every month the entire month for a full year, so I spend 0 on connects. And let's assume that this is not extremeley well paid and I earn a total of $1000 every month. I would be paying $1200 in fees for the entire year.

B. On the other extreme, let's say I get no job and I only spend money on connects. To make it more profitable for Upwork than the previous case I would need to spend more than $100 every month on connects (that is, buying at least 667 connects every single month).

How many people spend $1200 on connects for an entire year? Many, I'm sure. But they probably end up earning much more than $1000 a month. If I aim at earning $1000/month would I be paying $100 every single month without having any income? I seriously doubt it.

This brings another point, which I believe relevant to the analysis. Case A is self-sustainable. As long as I have jobs from Upwork, they will have my fees for granted. Case B is not self-sustainable.

They don't know if next month, next year or tomorrow I will decide to try another platform or look for jobs locally, or even look for employment and give up on freelancing altogether. What they do know is that if I don't get any job, sooner or later they will stop earning my money.

If my ROI as a freelancer is not enough, I will definitely quit UW. Surely, they don't rely on me entirely. If I quit upwork, someone else will join and pay on connects. But if that is the general trend and it is really difficult to find jobs here (because they prefer us not having them) sooner rather than later people will stop investing on Upwork and spend less and less on connects. The business does not hold up.

From a math/financial point of view, it seems reasonable to believe that UW would be interested in everyone getting a long-term job that guarantees a steady revenue, rather than relying on people deciding to invest blindly on connects until the end of time.

So, going back to the top, if assertion 2 is right (which I may be inclined to believe), I don't think it is because assertion 1. In my opinion, their lack of interest is probably due to other reasons I'm not fully aware of. It may be their enterprise program, or their flirting with the employment market. What I do think is that there is some issue, other than the freelancing gigs business, that gets their attention.

The logic could also be reversed. Meaning, assertion 2 does not happen because of assertion 1, but the other way around. They may have given up on the feelancing gig business altogether for other reasons and they are just trying to squeeze it dry before they finally let it die.

[matthewschenker]

Marc C,

Your details make many more assumptions than my simple examples!

The bottom line is this: Just examine any particular job posting. It's so painfully clear that Upwork makes more money on attempts  to get that job than they do on commissions from that job. Commissions happen once for a job. Attempts can happen 50 or 100 times for a job.

My example was very favorable to Upwork. I did not even factor in the duplicate jobs, and the fake jobs that have no intention of hiring anyone. In those cases, obviously Upwork makes more money from connects and boosts, because there is no money from hires! Surely, you agree that anything is more than $0, right? Or are you going to argue that I still have no proof?

The effort you make to defend Upwork in this situation is... odd. Let's leave it at that.

[feed_my_eyes]

You're making a lot of assumptions yourself, because you're not factoring in repeat business or invite-only projects. A lot of good, high-budget clients - I would even say most good, high-budget clients - will seek out freelancers in the search results instead of posting their projects, in order to avoid getting swamped by dozens of low-quality, time-wasting bids. For example, one of my clients has hired me dozens of times for a total of $35,000 so far; they initially contacted me directly, and since then, they just hire me over and over - they have never posted a single project. So Upwork has made $3,500 in fees from just that one client, and zero money from connects. I have a 90% repeat client rate, so Upwork doesn't make any money from connects on any of those projects, either. You're simply dead wrong when you say that "commissions happen once for a job". I'm not the only freelancer who gets lots of repeat business, and in fact, my Upwork income is very modest compared with the most successful freelancers here.

Upwork themselves said in their last quarterly results that they earn more from fees than they do from connect sales, so you'd have to accuse them of falsifying data if you're going to continue to make your claim. Anyway, let's pretend for the sake of argument that you're correct and Upwork does make more money from connects than from fees. And? If they're making a good profit from this, and they never made a profit from charging fees only, why would they change their business model? Just because you think that you know how to run a freelancing website better than anyone else? I don't understand what you hope to accomplish with these comments.

[matthewschenker]

Christine A,

Your response, like Marc C, is weird in how you defend Upwork.

Of course there are situations when clients hire freelancers on a repeat basis. But the vast majority of the jobs obviously have a long list of freelancers using connects and "boosts" to get the job! And yes, there is only 1 commission for a job, maximum! Your point that I am wrong when I say that "commissions happen once for a job" is very odd. Are there ever jobs where the commission is paid more than once?

If Upwork was really focused on building a freelance community, they would earn all their money from jobs that get freelancers hired, they would crack down on the growing number of fake jobs, and they would eliminate boosting (the concept of "boosting" is so flawed, I don't know how anyone with high-school level logical capabilities can fail to see it). What I'm saying is so inherently obvious, I'm suspicious of anyone who argues against it.

Instead of improving the hiring of freelancers, with each passing week, Upwork is squeezing more and more profits from the non-commission parts of the freelance situation.

[yofazza]

(the concept of "boosting" is so flawed, I don't know how anyone with high-school level logical capabilities can fail to see it). What I'm saying is so inherently obvious, I'm suspicious of anyone who argues against it.

It is flawed from your (some freelancers) point of view only.

The concept of creating competition for visibility among members was introduced (I believe) by Google around 2 decades ago with AdWords and AdSense, where the ads are the main business. This is where the terms CPC, CTR, comes from.

I then saw it adopted by different businesses such as product marketplaces, a few years ago. Where sellers compete with CPC etc. to have their products and store shown in multiple places, inside the marketplace! It seemed strange to me the first time I saw it, but I soon realized the potential for profits of the marketplaces. They have big member base, they use it.

And now, even a freelance marketplace adopted it as well.

Just open your mind, see from above, but don't be gamed by them. Although I don't like it, I do see it as "bad" or "wicked", but I understand it when I know that they never profit for years. I'll be the one who's "wicked" if I tell them to keep losing every month, just so I can make some profit, isn't it?

And is my 2k earnings makes things more suspicious or not? 😁

[feed_my_eyes]

Yes, there are jobs where Upwork earns commission more than once, as in cases where a freelancer is hired multiple times or the client keeps adding milestones to their original project. I've had some jobs that started off as $300 projects and ended up in the thousands. What part don't you understand?

But I'm not defending Upwork, I'm fact-checking you. It's interesting that you haven't even been here for a year, yet you think that you know everything about how they should run a website. I've been on Elance/Upwork for almost 24 years, and believe me, they've continually made changes over the years in their quest to make a profit. The website used to be higher quality and attract higher quality clients because they didn't use to let swarms of unqualified freelancers join; it used to be free to send proposals and they would take a 20% commission off the top. I made a lot more money, and I hate the way the website is run now. But, Upwork continually lost money, year after year, until they hit upon this idea of monetizing everything. I don't expect them to care more about my business than they do about their own. If I'm making less money, then that's my problem, not theirs, so I need to look for work elsewhere, which is exactly what I've been doing. I don't know why anyone with high-school level logical capabilities would continue beating their head against a wall if Upwork isn't worth their time and money.

As for boosting, it's nothing more than advertising. Radia summed it all up pretty nicely.

[matthewschenker]

Your response makes me wonder about your position here. I'm talking about how much money Upwork makes on attempts for particular jobs. It's inherently obvious that only 1 freelancer gets hired for a job, but multiple freelancers attempt to get that job. Are you saying this is not true? On top of the connects, any number of freelancers are also spending money on boosts for that job. But again, only 1 freelancer gets hired for that job. Even more, the jobs are routinely paying embarrassingly low rates of $5/hr or less, meaning that the commission is tiny. I don't think I need to be here more than a year to observe these facts! Are you saying that it would take more than a year to observe all this? Tell me anywhere I am wrong.

You are shifting the argument to a discussion of possible subsequent hires after that 1 freelancer is hired for that job. That's a whole different discussion, and has nothing to do with my points.

You are using a classic technique to deflect: change the discussion into something different than the stated parameters, then use that new discussion as proof against the original -- unrelated -- argument.

Please tell me where I am wrong about the argument I was actually making.