Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: How good are freelance programming sites?
39 points by gnosis on Jan 3, 2010 | hide | past | favorite | 14 comments
Does anyone have any particularly good/bad experiences to share about using freelance programming sites like elance, rentacoder, scriptlance, odesk, guru.com, etc?

Would you recommend using them? Does any freelancing site stand out from the rest or are they all pretty much the same?

I've heard horror stories about people not getting paid, projects never getting finished, etc.. how representative are they?

Scanning through the listings on some of the sites, it seems a lot of ads are looking people with many years experience and a buzzword-compliant resume to work for them for next to nothing. Are there some good jobs to be had through these sites, or are they pretty much a waste of time?




I've had great luck with rentacoder, and have only good things to say about the site itself and the controls and mechanisms it has in place to help ensure the process goes smoothly (ratings, on-site communication so there's a paper trail, escrow, etc.).

However, there are caveats:

1) Best used with very small, very well-defined jobs.

2) You must put in a fair amount of careful work to specify what you want done, in a way that leaves no room for interpretation.

3) Best used with low-risk (experimental / proof-of-concept) type projects.

4) Don't bet your company on getting good quality. You'll usually get something usable, but it often won't be exemplary code.

If "usually get something usable" sounds scary, see point #3.

5) Don't expect to save yourself time. Instead you spend the same amount of time as you would have spent had you already known how to do what needed to be done but you instead spend the time on communication, and you get your job done by someone who knows how to do it.

A corollary to #5 is for some small jobs, you might be able to learn how to do the task and then do it yourself, all in the same amount of time you spent directing the freelancer.

So why do it at all? Because of #6:

6) Using such a service can move an item off your to-do list, because having a person at the other end of a job bid (and having your reputation on the site as a buyer on the line) you get forced to finish the item. So effectively it's an anti-procrastination tool.

This is my experience, which is biased toward small jobs. But then, I wouldn't use one of these sites for a big job.


Good tips. I must add that I would not count on them knowing what they are doing. They may get you something usable but it usually won't be good practices. See the comment above about unfiltered SQL statements in the GET requests. If you don't know why that is a bad thing then when you get the code back from the people and it has this type of huge security error you will not know what to look for.

That being said I think there is a fair bit of work that can be safely outsourced to these people without too much fear of security. PSD to HTML slicing, etc.... But the old adage applies that you get what you pay for.


My vote: complete waste of time. I experimented very briefly with rentacoder before giving up on the idea of sites like this altogether. My experiences with services like this (and to a lesser extent craigslist) have been unpleasant to say the least. These days I only take on gigs that come to me through my network and even then I'm careful to vet potential clients before accepting a gig.


How do you vet your clients?


Next to simply being familiar with the marketplace and being able to tap informed individuals within it to learn about a potential client, LinkedIn is a GREAT way to vet clients. Networking in various forms is the best way not only to promote yourself, but to build a body of knowledge and a network of knowledgeable individuals who can help you to make determinations in these situations.

There are also various signs you become familiar with during the pre-sale/sales process. They shouldn't be completely inflexible. They should be willing and able to pay a reasonable deposit, or open to an early-payment discount mechanism that ensures you are pre-paid.

People skills. Networking.


I've used it for small web coding projects and had decent luck.

I always lay out the items which need to be completed with excessive specificity. That's the stage where I've seen a lot of people get let down, disparity in the expectations due to assumptions from each party involved. Never had any issues with payment, either.

If you're very plain and up front things will work out. Make sure you ask any questions so everybody understands what they need and what they're getting prior to any work or payment being completed.


From a buyer's point of view: odesk.com and elance.com are pretty good. All the others - not so much.

As a service provider - you will always get paid a bit less though - because of the real time competitive bids. When 15 coders bid for the same project, no matter how skilled you are - you will have to make competitively priced bids to win the projects.


I have been using RentaCoder for about 2 years now and find it pretty good. Once you establish a reputation, its upto you to pick up projects that are interesting and there are many prototype projects. Gives you an opportunity to try out new technologies that you may never do otherwise.

But as I get more experience on the site, I stopped bidding for low bid projects as the back and forth between buyer and seller takes a lot of time (for e.g. some sellers expect that you do a prototype before accepting the bid) and it does not work out for the amount bid for.

As for payment, Ive never had an issue with RentaCoder, Ive tried both snail mail and PayPal and they both have been fantastic.

My experience has been this: I only work for 2 or 3 clients now on Rentacoder because that takes all the time I have. They have been repeat buyers for me and keep giving more projects as I finish the current ones.


"Getting the best out of Freelancing Site" could be a decent selling book.


Is this area too fast moving for a "book"?

Perhaps a newsletter might be a better bet?


It looks more like you're asking from a providers point of view... I'm usually a buyer so I'll tell you about my experiences as a buyer:

I've been working with numerous freelancers for quite a while. It took me a long time to figure out how to choose my freelancers and I've had several failures along the way. The first things I outsourced were website PSD to HTML conversion ("slicing") and php scripts. These projects were usually somewhere between 50$ and 200$. While I did check their references and their ratings, I always tried to take one the cheapest providers. The projects grew and my average project size currently is 1000$, with the smallest ones still 100$ and the bigger ones 15k USD. I have changed my criteria since then and now look for the following things:

1. If you need a special skill (i.e. I once had a project that involved a lot of work with the google maps api): I most often search these sites my self and usually invite only the people that actually have that skill mentioned on their profile page. That way you can be sure that they actually know what they're doing and that they don't have to learn it while working for you.

2. Average hourly wage & price is the most reliable predictor about the quality. Usually providers will try to be as cheap as possible, providers who have enough work and who are confident enough that their work is really valuable however won't play this game and usually charge more per hour. I usually pick developers who charge more than the average.

3. It may look like you'll get a great deal but most of the time you won't. If someone is bidding 200$ on a project that even a non techie like you can say that the project won't be just a 3hour job but weeks of work, then don't pick this bidder. Most of the developers quickly loose their motivation to do anything after they have exceeded their own budget for this project, start to do sloppy work, or argue about not implementing required features.

4. Don't trust the ratings. There are some providers that go out of their way to get good ratings even after they screwed up big time. I've had people begging me not to give them bad ratings even offering me a refund. When looking at their ratings the valuable information is usually in the buyers comments not in the 5-star "grades"

5. Always check the code they're writing. I've had freelancers who screwed up you couldn't even imagine. Like putting and SQL query into a GET variable and then execute it on the server unfiltered.

This is when buying from one of these sites. By now I more or less figured it out but especially in the beginning I burnt my fingers quite a few times when working with freelancers. It's something that takes time to master.

I agree with ankeshk, I've had the best experience with odesk because their workforce is better qualified and the hourly billing is ok for me. I also oftentimes use elance for some projects.

Another helpful tip: I was working on a project where I needed a twisted proxy to translate something for me. I started reading the documentation of twisted and after a few minutes I decided that I'll try my luck in their IRC chat. I went in there and basically just said: I need some help with twisted, would anyone be interested in a small half day project and earn a few bucks? and worked with the first guy to answer in the chat. It worked out quite well.

I've also hired people from djangogigs.com and it worked out ok however you have to watch out that these people really are freelancers that want to do freelancing, I've worked with a few who said after a couple of weeks that they couldn't do it after all because they don't have enough time after their normal workday to do freelancing.

There are outsourcing companies like Itteco.com and x-minds.org which I can really recommend for bigger projects and longer lasting partnerships. I have worked with both of them, I don't know all of their developers but they certainly have good developers. If you decide to work with a bigger company I would usually only do that with a personal introduction and/or careful due diligence.

Generally speaking: if you're looking for freelancers, ask people in your network, if they know good freelancers that's the best way to find them.


I have worked through Guru.com and enjoy it. The web site provides a lot of nice tools, including escrow payments, work rooms, and private message boards. You pay for it though -- around 7% fees on each project.

The one complaint is that you have to lower your rates substantially at the start until you build up return clients. At the point, they like your work and give you private projects.


check out lancemonster.com. It's a new freelance site but it's really easy to use. I used them already. its free to join, bid, and post projects. Great site!


Independent tech consultant here. I did a healthy five figures in part-time indie consulting in 2008. I went fulltime indie in June 2009, and in that last half-ish of the year, did more in Twitter-related work in '09 than all work in '08, and billed something like 200-250% total in '09 over '08.

I've subscribed to various freelance sites for years (since 01 or so) ... guru.com, elance, rentacoder ... I've never once landed work off of any of those sites. You have to spend too much time competing with:

a) cheap offshorers b) cheap "clients" uneducated in project management/software development and are therefore unwilling to pay appropriately for it c) noise, noise, noise d) general wage pressure of the "marketplace"

... to make it worthwhile. Some people who got in and got big early on made their reputations there, but individuals on these sites tend to top out at maybe 40k/year -- and those are rare. The only people making any greater money on them tend to be people who have turned around and co-opted teams of other developers via the site.

You're better off networking, both online and in-person. Make sure you have a solid skillset and a unique value or unique sales proposition. Focus in one area to begin with -- new consultants who try to generalize too soon tend to fail before they have a safety net of money or reputation. Once you're really, really good at something, consider branching out.

Attend and present at local and regional user groups and unconferences. Offer webinars and seminars. Maintain a regularly-updated blog and newsletter that focuses on that initial niche. Write (useful, not overly markety) whitepapers. Join your local Chamber of Commerce. Attend relevant Meetup.com meetups; start relevant meetups. Check out your local SBA office. Always carry a business card, and manage your personal brand and reputation -- it's your finest and most valuable tool.

All of these things are far more worthy uses of your time than piddling it away on noisy, garbage-filled, flake-filled "freelance marketplaces."




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: