I like this idea, I had it myself. I was even discussing this yesterday with a friend of mine. Two years ago I even put together a prototype and started experimenting with it.
Our conclusions? Yes, freelance sites suck. Do providers want something else? Hell yeah! Do buyers want something else? Hell no! Simply put; being able to first try to get your project done for $100 while you know the real price is at least $20k+ is compelling apparently. I know companies who had stuff we quoted for $50k done for $1k. And it didn't even suck too badly. Why? Because there is SO much competition on those sites that 'newer' providers have to compete on price to get anywhere. So you can get occasional solid work for 1/50th of the price. Especially if you know what to look for.
Then if it doesn't work you just go 5x that, doing it for $5k and most likely you'll have your $50k project done nicely. Note that most companies under $10k price point are willing to go no-cure-no-pay, meaning if it failed so far, you didn't pay a cent yet.
If you get farther than this, yes a different method would be better, however, most don't get to that online or they already have their preferred teams, meaning this won't help them. I already know that if I pay a Ukraine team with 400 5 star projects $50k I WILL have high quality; why do I need these moderators?
I'm not against this at all mind you; like I said, we thought about this seriously ourselves but we just didn't see enough interest from companies. Basically most companies on freelance sites who want to buy something still suffer from that Elance-Digg-syndrome; pay very little for top-notch stuff. And often you do get that.
Also for that $50k the site is worth, you can for instance put 6 teams for $5k on it, having a quite high (from experience) chance of success in at least 1 or 2 of them and banking $20k 'profit'.
I wish you luck but it's a very much uphill battle!
You have too many generalizations in your post with no mention of the countless variables which make a project a success or not. Dollar figures are almost meaningless as this is just one small part of the whole picture.
One of the best pieces of advice I have ever heard regarding offering development services is that the technical skills are easy to pick up, it's the people skills which are difficult. It's the people skills that buyers are willing to pay for if they have the budget. These cheap developers aren't able to connect with the people skills, so they aren't able to break out of their low cost offerings. Instead, they might work for a middle-man who handles the client for a mint and hands off the work for cheap.
Also, if developers on Odesk racking up 5 star ratings are so great, then why don't companies such as Github and 37Signals with mostly remote teams just go snap up all the 5 star developers they can find on Odesk? The reason is that there is a big gap between A-level developers and C-level developers and that gap is measured in ROI. Dollar amounts mean nothing, ROI is the most important number. A 5 star developer working cheaply on Odesk might get the job done with passable code, but the A-level developer who isn't on Odesk (no need because this person has no problem finding work) will deliver a much higher ROI.
Sure, the A-level developer won't be a good fit for every situation (maybe you really just need someone really cheap to get the job done ugly) but this person doesn't have much competition.
As for the service from the OP goes, just get going. You never know until you at least start the initial probing.
Scoop: Contractors will not be quantified by 400 tasks each worth $5.
Contractors will be able to include their stackoverflow profile, github, blogs, articles written, personal or paid projects, endorsements, along with usual reviews. This provides a much better assessment on the type of person, and yes, communication and people skills.
I've spoken with countless people complaining that they tried going with a freelance and they gave up when the individual disappeared. This is the problem we have to fix and I'm willing to bet there are enough buyers looking to buy real service for their money.
But i'm using generalizations just because the OP does that too; freelancer services suck. Yes they do, for certain jobs. The fact that there ARE countless variables gives me pause about the viability of this project. You either try to provide a service which github/37signals would hire which they will never do (they would try to steal all the people away but they wouldn't outsource) or you are trying to give less high brow companies 'worth for their money'.
I think in GENERAL the market is big enough for this, it's just an uphill battle as most companies won't listen. They are price focused EVEN though they had very bad experiences. We have people calling: 'I have had very bad experiences on freelancer.com, so now I want a local person, I found you. I paid $2000 on freelancer.com and that was shit, so i'm thinking $5k max here.'. This is just very normal behavior. Without seeing the project I'm 100% sure, after reading that, that this project will be FAR over $20k to build, not $5k. This is, however, not what the buyer thinks. Issues like that are by far the biggest part of the market. And enterprises probably won't use the services as they want to be pampered locally (so it can be a frontoffice service in front of this, but not just this).
Also; 'technical skills are easy to pick up'. Right. I agree with the people skill part, but technical skills are easy to pick up is complete rubbish. It's just simply not true. 'Cheap developers' do not only lack people skills as you are implying; they are mostly just really bad and not capable of 'picking up technical skills easily'. I can copy interviews with CS Master graduates with 3+ years experiences showing that after roughly 10 years of study + experience 'nothing' was mastered. And that's not an exception. I understand that your experiences are different; I would like to see what you mean in that case. We have managed to turn every promising/talented devs into very valuable developers with people skills; there are a few (really a few, like 1 or 2 on over 500) who where so Rainman that it didn't work, but the rest did. So what kind of 'cheap developers' do you mean, because I think we mean the same thing but you just call that people skills?
But yes, right here is; people skills and especially quite advanced people skills and project management skills are needed to do large projects where this pesky money thing hardly ever appears; these people are very hard to get on this site. If you have those skills you are worth A LOT of money and if you have those skills you know you are because you learned those skills somewhere.
If OP gets this going i'm willing to help (for free) giving advice if wanted (I have lead and done 1000s of projects, direct or indirect, for big and small companies) and just talk about the problems in this market. I'm not looking for money or a % or a job; i'm just very interested in the subject. That's why I recognize the problems and I would like to brainstorm about the possibilities as I find the challenges here very interesting. So if you want to mail me, please do.
oh, I (almost) agree! The thing that I don't agree with is that _often_ you get $50k worth of work for $1k. It _may_ seem this way at a first glance, but a lot of projects end up being cancelled (wasting everybody's time) and if you do get something for your money, on the medium/long run it will cost you the same and more: for lost time, bad choices, no support, difficulty to update/upgrade the product, etc, etc. Lately, the entire "Development" field has gotten a bad vibe, of mis-understood, unreliable and difficult to control wizards.
And I think it's time for buyers to be educated on how this should work.
I'm not even going for these guys anyway, I don't need the buyer looking to spend $200 for a Google clone (yes, I've seen this), I don't need the guy selling crap for $5/hour, crap that needs to be re-written completely for it to work, I don't need arrogant asshole buyers, or low-bid cheap spammers.
We want to attract the enterprise market. That means A LOT less projects, paid submissions, reviews every step, guaranteed support from contractors, bug-free warranties, higher costs and significantly better service.
As you said, it's a very much uphill battle, but I'm willing to try (or else I'll go back to feeling sorry for myself).
So, I need all the support I can get, sign-up.
I agree with you on that point. I'm just telling how it looks from the outside. Once you are really doing projects, it all looks very different. That's why we tried to make something like that in the first place.
Our conclusions? Yes, freelance sites suck. Do providers want something else? Hell yeah! Do buyers want something else? Hell no! Simply put; being able to first try to get your project done for $100 while you know the real price is at least $20k+ is compelling apparently. I know companies who had stuff we quoted for $50k done for $1k. And it didn't even suck too badly. Why? Because there is SO much competition on those sites that 'newer' providers have to compete on price to get anywhere. So you can get occasional solid work for 1/50th of the price. Especially if you know what to look for.
Then if it doesn't work you just go 5x that, doing it for $5k and most likely you'll have your $50k project done nicely. Note that most companies under $10k price point are willing to go no-cure-no-pay, meaning if it failed so far, you didn't pay a cent yet.
If you get farther than this, yes a different method would be better, however, most don't get to that online or they already have their preferred teams, meaning this won't help them. I already know that if I pay a Ukraine team with 400 5 star projects $50k I WILL have high quality; why do I need these moderators?
I'm not against this at all mind you; like I said, we thought about this seriously ourselves but we just didn't see enough interest from companies. Basically most companies on freelance sites who want to buy something still suffer from that Elance-Digg-syndrome; pay very little for top-notch stuff. And often you do get that.
Also for that $50k the site is worth, you can for instance put 6 teams for $5k on it, having a quite high (from experience) chance of success in at least 1 or 2 of them and banking $20k 'profit'.
I wish you luck but it's a very much uphill battle!