Notes from a Data Entry Gig

Large pool of cheap resources sometimes is enough of a motivation to outsource tasks. Sometime even those that you might not have done in the first place ;) It also is very tempting to engage manual labor rather than create, debug and use tools. Those reasons along with some business drivers were behind a data entry project I started a few weeks ago. While small and fairly simple the project offered a few interesting lessons to learn and a couple of interesting points to share.

  1. There are many places where you can find freelancers. Most of those places offer offshore labor. Even local resources such as craigslist will generate more response from offshore than from locals, even if you specify “locals only”. In my case I was specifically looking for offshore resources and the rock bottom rates. I knew that every site has its own community of freelancers, what was somewhat surprising is how substantial the difference in response would be. Response to my ad from 5 sites I tried in the first 3 days was 0, 2, 3, 6, and 78. The last figure was the response from oDesk community. It’s no surprise that the best candidates also came from oDesk. As a matter of fact I ended up to picking all providers from oDesk (I was looking for 5 people).
  2. The rates diversity was quite surprising as well. My project which was a basic internet research and data entry attracted freelancers from all over the world with majority of applicants from India, Pakistan and Philippines. There were a couple bids from USA (I frankly doubt that the work was planned to be performed by USA resources though). The lowest bid was $0.78 an hour (Bangladesh), the highest was $26 an hour (India).
  3. The quality of responses varied greatly from thoughtful and professional to “Need a job!”, the last one incidentally was one of the highest bids as well.
  4. Fit between the job and skill set was decent with a few exceptions even though I had somewhat of a difficult time categorizing my project – fitting it into one of the categories / subcategories provided by the sites.
  5. Each of the sites has its own idiosyncrasies and proprietary conventions; that makes search for freelancers across several sites rather cumbersome. In this case I did not have to work across the sites – the difference in response clearly made oDesk a better place to seek for my resources. That is not always the case though. In particular many type of projects such as web design, graphical arts, etc. would find equally strong support on many sites.
  6. For this project pruning candidates was not complex – I cut off everyone with rate above $5 an hour and those who did not appeared to put any efforts into their bid. That still gave me about 25 candidates, at that point ratings and hours worked helped me quickly pick top ten.
  7. I did not put a lot of efforts in the “Interviewing”; a quick email exchange quickly showed whether the person appeared professional and responsive enough. A few of candidates requested Skype conversations, that was a bit more time consuming and I am not sure whether for this kind of project the time is justified.
  8. I picked 7 suppliers (my target was 5). Can you guess why? Of course the quality of suppliers, especially when you scrape the bottom of the rate barrel is a hit or miss. One of them “did not show up for work” after the bid was accepted, one turned out so dense that I had to stop working with her after two days into the project.
  9. I now have only three suppliers left. All three are from Philippines and all are doing a decent job. The rates are 1.11, 2.78 and 3.33 an hour. The communications are sufficient. Productivity as expected or even better. I think so far I can call this project a success.

If you are facing a data entry, web scraping, email response, etc. project here are a couple tips I suggest for you to consider:

  1. Using freelancing sites saves time of sourcing candidates, simplifies management, and helps with payment aspects.
  2. Today the rate target could be $3 an hour plus / minus a buck.
  3. Have a very simple, concise and unambiguous project description. A step by step operating procedure should be developed. (remember the 3rd fundamental rule of outsourcing?)
  4. Do not invest too much effort in selection of the candidates; it’s easier and faster to start another project and get a bunch of new candidates than try to pick just the right ones. Using the project above as example – the candidates I thought were the best are no longer on my team, one of them was the no-show.
  5. Use the site communication methodology rather than your own email. That reduces the clutter in your own inbox and helps with categorization of email and follow up.

I guess that’s as much as this project deserves. I am kicking off  a SEO/SEM project shortly. It will be a bit different will see how it pans out and whether there is much to learn from it.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to Ma.gnoliaAdd to TechnoratiAdd to FurlAdd to Newsvine

One thought on “Notes from a Data Entry Gig

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s