The Data Whisperer

by Jinryu

[Zanshin] came up with that name for me, and I kinda like it.

My part-time job, which helps me get through my law degree without starving, consists of data management for a major Sydney company. The official title is something like “Data Collection Officer” or something like that, but there’s actually a fair amount of work that we do that doesn’t involve calling people up. Mostly it has to do with fixing bugs in the data that are introduced due to human error. In theory though, aside from error correction, most of the raw data comes from manually calling up clients and obtaining it from them. It’s not as bad a call centre job as you can imagine, because the people we’re calling are expecting our calls.

It’s kinda strange that in this day and age the job like this still exists. You’d think that there could be automated systems set up so that the necessary data would automatically be reported electronically between our departments, but somehow, we’re still doing this the old fashioned way.

Our progress through the day tends to be measured by how much raw data we can obtain, in the form of entries. I found out that the average for an experienced peer, through calling, is somewhere in the neighborhood of about 70 to 80 entries per working day. I found out that I was getting about 80 entires as well, but it was a pretty non-stop day of extremely repetitive work.

I’ve found a bit of a backdoor to the raw data though. You see, I don’t really like calling. Despite that the people we’re calling mostly expect us to be calling, they’re not very efficient at providing us with the answers we need when we actually do ring them up. So, we get all sorts of runarounds: we’re put on hold; told to leave messages; told we’ll get a call back; told to send a formal written request for the data; told to call someone on their mobile; etc. All sorts of things that, really, shouldn’t be happening in an efficient environment– and it all happens because people are basically the weakest links. By that, I mean, humans are the weakest links.

I’ve found a way to get around that though. I’ve found it’s possible to get to a lot of the data through just looking at concurrent data gathered by other automated sources– basically, electronic clues that, if you did some basic math in your head, you could arrive at the same conclusions that you would if you managed to talk to a real human. The raw data isn’t always available through reading the concurrent sources, but when it is, it’s much faster. Instead of spending on average about 1-3 minutes per call (on hold, mostly), I might simply be able to luck out and read interpret the concurrent data, which gives lets me have a result in about…. 13 seconds.

You do the math. It means that I can get the same amount of results in about a tenth or twentieth of the amount of time it should take me.

It means I have more time to do other things.


I think to some degree we all do the same thing when it comes to dealing with people. What humans are willing to communicate to me is usually the weakest link of information. If, instead, we find objective projections of their actions, their situations, their positions, etc– oftentimes, it gives us shortcuts as to figuring out what they’re about. Sherlock Holmes’ ability to size people up simply based on superficial clues is an extreme example.

All I’m saying, people can say what they want– but some things speak louder and more accurately than what they say. There are other ways to be informed.