Frilans Best Sites

Introduction

oDesk is the world ' s largest (at least as desk itself) international freelancer exchange. The desk itself on the Habre has been written quite a few times, for example here or here, where, in my view, it's almost burnt out why and for whom the resource is created, especially what it is. The description phase and the website ' s operating principle could therefore be deleted. I'll try to analyse data on both the Frilancers themselves and the orders, customers and their demands on the Frilancers, saying that you need to learn and know to be more less about modern technologies. I will also analyse demand and supply based on data on Frilancer skills and customer demands. And, of course, some statistics and some pretty pictures as examples (who works for the Desk, where mostly orders come from, who earns more and who works better, etc.). And all of this is based on self-collected information, so openly and generously provided by the Desk’s own through API. It should be noted that the article, with a small number of statistics on the desk, had previously been able to blemish in the harbour, but in it, in contrast to the current article, the Desk himself had praised the results. In general, I would like to say at the outset that the review is not intended to be complete, but the data collection process will be briefly described below, so that your own analysis can be conducted as necessary.

Data collected

In creating a new job, the client usually adds a list of the necessary skills to perform it. Skills may vary widely (both knowledge of programming languages and knowledge of operating systems or conventional languages, such as Russian or English). A full list in alphabetical order can be found here, just at the time of writing about 250,000 skills. In addition, their skills are added to the profile and Frilancers. So I've been doing my data collection on the basis of their skills and skills. Well, first, I got a list of all the skills with API. Frilancers♪ In fact, this method is missing work and Frilancers who have no skills, but we're just going to think that we're not interested in doing this because the client doesn't know what he wants, and Frilancers with no skills or deer to add them to their profiles are unlikely to work well. The reference code of the consumable application that collects data and keeps them in the database, I've put it on GitHub.

A little bit about the size of the base that was collected. A total of about 500,000 Frilancers, of which over 200,000 did at least one job and some 150,000 worked for at least one hour. More than 50,000 open works at the time of writing, with an average of about 5 skills.

Regrettably, with API, I did not have access to financial information, but rather how much customers actually spend and how many freelancers earn, but on an indirect basis, some estimates could be made, for example, on the number of hours worked and the hourly rate of Frilancers (although not always true, as sometimes the hourly rate indicated in the profile might not be appropriate).

Country

For starters, rather than countries where more work is being done and where more Frilancers come from, and a little about how well these Frilancers work and how much they ask for money.

The next picture shows countries with the largest number of jobs (more intensive colour means greater importance and signatures are absolute). The top 10 countries include the USA, Australia, the United Kingdom, Canada, India, the Philippines, Germany, Israel, Singapore, Pakistan, in order to reduce the number of jobs.

The next picture shows the countries with the largest number of freelancers. Top 10 countries, in order to reduce the total number of employees: Philippines, India, USA, Bangladesh, Pakistan, Ukraine, Russia, Great Britain, Canada, Romania.