She couldn’t find the page, which was probably the main oConomy page here which currently shows 248 providers working and 95,545 registered int the system.  I admit that those numbers weren’t calculated to put a new provider in her happy place, but they don’t tell the whole story either.

It isn’t showing how many providers on the network have jobs, but how many people were actively logging time on the oDesk client at the time the numbers were generated.   Given that oDesk is a 24 hour global marketplace those numbers will fluctuate throughout the day and really aren’t anything to worry about.  They certainly don’t reflect the number of providers who are currently employed.  (I expect they’re short by at least one or two orders of magnitude).

Now let’s look at some other numbers from the same page.  There are almost 100,000 providers, and together they’ve earned just under 40 million dollars, which works out to an average of almost U$400/per provider to date.   Since not every provider on oDesk has worked, and that some profiles are so incomplete that the providers are never going to work, the numbers have nowhere to go but up.

When it comes to getting a job, none of those numbers matter.

As I said in the headline:  Chance only matters when it’s random.  oDesk buyers do not select providers by rolling dice. (Well Nelson might, but he’s a special case.)

Buyers select the provider they think is at the intersection of highest quality and lowest price. They go for the most value for money within their budget.

It’s not random at all.  Never has been, never will be.  This is business, not gambling and buyers don’t want to throw their money away.

The end result is that you’re not competing with all 95,544 other providers on the network for any given job.  In fact you’re rarely competing against as many as 10 serious competitors for any given position.  Those still don’t sound like good odds, but that’s just chopped four orders of magnitude off the ratio.

It gets better because they aren’t odds.

It’s not random.

It’s deterministic, and some of the factors are ones you know about and can control and the others are unknowable and uncontrollable.  However even the ones you don’t know and can’t control aren’t random chance.

If you write a great cover letter and your skills fit the job you are very likely to get it.  If you can demonstrate that your skills match your rates you’re even closer.

I’ll go so far as to say that any competent provider with a decent profile (including a picture) and portfolio is guaranteed to be able to find work at or slightly above their category average if they put the time in.  The work is there, and the buyers want it done more than they want their money.

Let me leave you with one other thought about statistics:

While statistics may say only one in ten people can succeed at something, they never tell you whether you’re the one or one of the nine.

I was on another site and someone was hiring members to take and pass the oDesk Readiness Test for them.  Don’t do it.

Just don’t.

oDesk wouldn’t exist if it didn’t make money, and neither would its competitors.  The business model is simple, connect buyers and providers and take a cut off the top.  There’s more to it, but that’s fundamentally how they all work.

It’s much like an agency relationship and it works very well.   Most sites take between ten and fifteen percent and provide various services ranging from escrow and payment through oDesk’s time management software.  The benefit to this method of payment is that it’s performance based and so sites make more money by getting more work for providers.

Some sites have a second revenue stream–

Paid Memberships

Providers pay the site a monthly fee in return for additional benefits over the standard free membership.

On the surface it doesn’t sound like a bad idea, but in practice it does have one major drawback.  It’s a significant change to the business model.  It’s often easier to get a new provider to pay $10-15 per month than to get a buyer to pay $100-150 for a project.  It also extends the site’s revenue stream to include funds form unsuccessful providers as well as successful ones.

Businesses reinforce success.  This means that if they’re getting more money from subscription fees than project fees that’s where they will focus their money-making efforts.

There’s nothing wrong with this from a business perspective, but from a provider’s perspective there is a problem here:  it dilutes the focus on getting new jobs.

Before we go further, let’s look at some examples of membership systems:  eLance and GetaFreelancer both use paid subscriptions, but they take very different approaches.

A free member on eLance gets three bid points per month, and depending on the size of the project it can require anything up to four bid points to put in a bid.  An individual membership is $9.95/month and gives the user 20 bid points and a lower commission percentage rate in addition to other benefits such as a greater number of keywords and more contact options.

A free member on GetAFreelancer has more bid points (I have 25) and pays a 10% commission on all jobs.  The paid membership costs $12 a month, and not only provides more bid points, but also eliminates the commission.  I’ve also never seen a job on GetAFreelancer that required more than one bid point.

Here’s where I see the difference:  GetAFreelancer is rewarding paid members, while eLance is penalizing free members.

I’ve had a paid membership on GetAFreelancer, and only let it lapse because I haven’t been working there lately as my oDesk clients are taking up the majority of my work time.  Should I shift focus back toward GetAFreelancer I will definitely restart my membership.

I haven’t gone for a paid membership on eLance, and part of that is because the free one is so restricting.  My preference is to bootstrap my first job on a site to getting a membership, and with the restrictions on free members, I don’t spend much if any time applying for work on eLance.

That’s the real pitfall of paid memberships, they often make it hard for new providers to get started on a site.

Some might argue that it separates the serious providers from the frivolous but I’m not so sure.  There are other ways to limit bids that don’t involve paid memberships and they seem to prevent frivolous and spam bids just as well as the memberships do.

Alternatives

oDesk uses a combination of skill tests and buyer feedback to limit the number of bid points available to a provider allotting anywhere from five to twenty points a week.  The big benefit of this is that it’s possible to get to the maximum number of bid points right from the start just by taking skill tests, which helps new providers get started.  Once a provider builds up a history it’s all based on feedback, but the skill system really helps get people in the door.

Rentacoder allows people as many bids as they can reasonably make.  I don’t have a lot of experience at the site but I haven’t seen any complaints about spam bidders there.
Personally, I prefer the oDesk approach, but I don’t have a problem with the idea of paid memberships that reward paid members.

Just remember, it’s your time and your money, so be careful.

And for writers remember Yog’s Law.

For some of you it probably works.  Every minute you spend surfing the boards is a minute you aren’t searching for work or actively earning money, and there are some providers who just can’t spare the time.

If you’ve been reading this blog regularly you’ve probably noticed that I am an active member of the oDesk Community.  I not only read the message boards but I also try to post fairly often too.

What I’m going to try to do in this post is explain why I spend my time on the boards.

The first reason is simple:  Most new features are announced and explained on the community boards and by reading them regularly I can keep up with the changes.  That becomes particularly important since I try to post those same changes here too so that those of you who don’t read the community boards can also stay in the loop.  Besides, it gives me something to write about.

For example I discovered the new Message Center by checking the Community Forum.

It provides a single place to view all of your oDesk related communications so you don’t have to go digging through your email.  I have to admit that I’m torn on this one.  I can see definite benefits to it, but at the same time, I already have my email open so do I really want to keep the message center up too?

I’m really not sure.

At least for the moment I think I’m going to use the Message Center as a supplement to email, not a replacement.  It might be different if I could reach it from any tab on the site, but we’ll have to see.

Speaking of email, oDesk has recently stopped sending out job application confirmation emails.  They have also stopped notifiying people they have applied for jobs in the message center.  As of July 15, 2008 if you apply for a job the only confirmation you will receive is that it will show up under “My Candidacies.”

I think it’s a good idea.  You don’t need an email to tell you you just applied for a job.  I don’t know about the rest of you but I generally remember my job applications.

Both of these changes are things I would not have been aware of had I not put some time in on the message boards.  They can be a very good place to go for both assistance and information.

The only thing I would watch out for is the negativity.  Some days you see a lot of it on the boards, and it can really get you down.  The truth is that the people complaining the most are usually, not always but usually, the ones who have not yet figured out the best way to make oDesk work for them.

Many of them don’t know the economics of the site either, and think all jobs pay the very low rates that someone is always complaining about.

In the meantime I recommend you use the boards, but don’t let them use you.

One last bit of news before I close.

Things are always changing, and one of the things that’s changing is the oDesk Insider.  One result of these changes is that you may start seeing fewer updates that you’ve been getting of late.  It’s not a bad thing, but other things are forcing us to put this blog on the back burner for a while.

Anyway until next time (and it won’t be that long) I wish you all happy and profitable freelancing.

You can always reach me at dave(at)odeskinsider(dot)com

Let’s start with what I consider the three most important factors:

oDesk is a free market

oDesk costs money to run

oDesk only makes money when providers make money.

Understanding those three points is vital if you want to understand how oDesk works.  Yes the oDesk Readiness Test is important, as is knowing how to work the oDesk Team, but there’s nothing really all that hard about either.

So, join me on the other side of the fold and we’ll discuss our three factors:

Glad you made it across safely.

Now, where were we?  That’s right, we were discussing how oDesk works, and to start at the beginning let’s talk about a free market.

When I call oDesk a free market I mean that it is free of artificial constraints on the cost of services.  Every provider is free to set their price and every buyer is free to set their budget.  Providers are not required to take a job if they don’t think the pay is worth it and buyers are not required to select a provider if they think they are too expensive.

I’ve seen people work for $0.50/hr and I’ve seen people work for more than $100.00/hr.  I’m not talking about posted pay rates, but about amounts that people have actually been paid for jobs they have done.

That’s a huge range, and the free market is why it’s so large.

We all know that data entry rates are the lowest on oDesk.  There are a number of reasons why that’s the case.  One reason is numbers, there are just under 33,000 providers who list data entry as a specialty.  That’s second only to web programming with just under 34,000.

At first glance you might think that would put the pay rate for web programmers into the same ballpark, but if you do you’re forgetting the other half of the equation:  Job availability.  According to the most recent oConomy figures, there were 4725 web programming jobs, and only 1035 data entry jobs listed.  That means there are seven web programmers for every job, while it’s close to 32 providers for every job in data entry.

Another reason why it’s low-paying is that for the most part any two data entry professionals are pretty much interchangeable.  Yes there are differences in speed and accuracy, but the final product from two top-ranked data-entry people should be identical.

This means that the only things a data entry professional can use to distinguish themselves from the crowd are quantitative factors, not qualitative ones.

If you combine the fierce competition for a limited number of positions with the interchangeability of providers on a qualitative level you have a recipe for low rates.  That’s how the market works.

It may just be a coincidence, but the average rate for web programmers is about five times that for data entry, and they have one fifth as many providers for each opening.  Data entry is very much a buyer’s market.

The above should show why data entry in particular has such low average rates, but what about other fields?

Providers in other fields have something else to distinguish themselves:  Quality.

Some coders are better than others, so are web designers and writers.  This level of ability gives them something else to distinguish themselves.  They can say that the reason they are worth more than their competitors is because they provide a better product.  That gives them something else to set themselves apart and another way to justify higher rates.

Again, it’s the free market at work, this time in the other direction.  Good providers who are in demand can charge more because people want those providers specifically, and they only have a finite number of hours.

The next thing to remember is that oDesk has expenses.  oDesk Team cost money to write, and the servers cost money to keep up.  Then there are other expenses such as paying the staff and interest on financing.  The details and the amount are unimportant for our purposes, all we care about is that they exist.

Finally, we come to to the point that oDesk takes a proportion of what each provider bills.  Their earnings go up at the same proportional rate as each provider’s do.

These last two give oDesk the strongest incentive possible to drive provider rates up rather than down.  They’re in business to make money.

If you’ve followed the community for any length of time you know that people are always complaining about pay rates.  This is usually followed by a request for a minimum wage.  Unfortunately what seems to be forgotten is that this would simply be the first constraint on an otherwise free market.

Constraining a market is never a step to take lightly.

The first problem is that oDesk doesn’t exist in a vacuum.  If oDesk constrains its market, then people looking for a free market will simply go elsewhere.

The second is that it’s impossible to constrain the bottom of the market without constraining the top of the market as well.  Setting a wage floor makes it harder for people to rise above it.  They can still do it, but neither as quickly nor as easily.

Many of the arguments for a minimum wage on oDesk have to do with questions of social justice, saying that people are worth more than some arbitrary minimum.  Other arguments have to do with people saying it’s uneconomical for them to work for less than said minimum.

In the latter case the answer is simple, don’t apply for jobs you cannot afford to do.  Buyers are not responsible for your financial health.

The first case is a little more difficult:  Questions of social justice tend to move the argument outside the boundaries of pure economics.  Most countries do have minimum wage laws in place, and many of them invoke either basic human rights or the social contract to do it.

Let’s start by making one thing clear:

A minimum wage is irrelevant because we are talking prices not wages.  oDesk providers are contractors selling services, not employees earning wages.  So what people are really discussing are price controls, not minimum wages.

Remember that.

Once people accept that the discussion is about price controls rather than minimum wages, it becomes easier to understand why the various issues of social justice don’t apply to oDesk.

People don’t live on oDesk.  They live in cities, towns, states, provinces, and countries.  All of those places have their own jurisdictions and the right to set minimum wages for people working there.  They also collect taxes to pay for services and implement programs of social justice.

Don’t conflate a marketplace with an employer.

You can make money on oDesk.  If you’re good at what you do and know how to sell your services you can make quite a lot of money.

If you understand how it works you’ll make more.

oDesk indicates your reputation in a few ways.  Everyone should know how the feedback system works, and if you don’t just drop me a line or put in a comment and I’ll be happy to expand on it here.  The other aspect of your reputation is how well you rank when buyers do a search.

The basic system works pretty well, it sorts by whatever criteria you specify, and uses the number of oDesk as a weighting factor.  So if you are searching for providers with a perfect feedback rating, you’ll see all the 5.00 rated providers with the one with the most hours listed first.

I like that as someone with a very large amount of hours has demonstrated their reliability.

If you need to you can also filter providers by anything from pay rates to hours worked or even whether they’re affiliated with a provider company.   In all cases your results will show up based on your primary search criteria weighted by oDesk hours.

It’s a good system, but it does have flaws:

Here’s an example of what I’m talking about.

I recently raised my rates and so I wanted to see where I was sitting among the writers working on oDesk.   I think I’m one of the better writers in the system and wanted to make sure my rates were in the same range as my peers.  So I went to the provider search, set my filters to exclude anyone with zero hours and did a search for writers based on hourly rate, highest first.

The first four pages were all programmers.

That’s right, none of the top forty providers that come up when you do a search for writers by pay rate are actually writers.

If you search by feedback it appears to get a little better, the fourth ranked candidate has writer listed in their job description.  However, this person is primarily a programmer and has not taken any writing tests or done any writing work on oDesk.  Going over their feedback you can see that the person is a very good communicator with excellent English skills, however that does not matter if someone is looking for a candidate for a writing job since this person doesn’t do those kind of jobs.

It’s not until number eight that we see anyone who has actually done writing work, and even then it’s only a fraction of their work history, which is mostly administrative support.

The next two entries in my search are people who are primarily working jobs from oDesk’s writing category.  One’s primarily a transcriptionist and the other’s doing editing work.  At least they’re on the first page.

The sad thing is that this is the best result I’ve ever seen for writers when searching the writing category.

If you search by portfolio items there is only one writer who comes up on the first page, and that’s only because the person in question spent an entire afternoon doing nothing but add items to her portfolio.  That got her 78 items, and second place.  It also gives her the distinction of being the only one on the first page with any writing samples in her portfolio.

I don’t know what the other categories are like, but from my perspective, and that of many other writers this is a serious problem.  If a buyer is searching for a writer, they should be able to find them without wading through dozens of providers in other categories.

The problem is that complaining alone doesn’t do any good, we need to find a way to fix the problem.

I think part of the problem may lie in the fact that many providers may list multiple job categories on their profile and primarily only work in one of them.   One possibility could be to get providers to list one category as their primary, and others as secondary and then provide the ability to filter providers by primary or all categories.

That way if you filter by primary category only you should be able to get providers who focus in that field, whereas if you don’t filter at all you could get the present results.

It’s a tough question, but that’s the best idea I’ve been able to come up with.

It’s not really changing the basic algorithm, which appears to work fairly well, just allowing for better filtering for relevant providers.  Having said that I have no idea how hard this would be to implement, but I think it might go a long way toward resolving this issue.