This week, we received a pretty routine request from a client: track down the address and phone number of a former employee of a company. The client wanted to interview this person in relation to an ongoing lawsuit. Another employee had provided the person’s name, believing that he had some insightful information that would help our client’s case.
The name seemed rather uncommon, and although the person worked for a large multinational conglomerate, it was believed that he worked in a relatively small town in the Midwest.
Seemed pretty straightforward.
We did a couple of initial searches through a few of our investigative databases, which found that there was exactly one person in the country with that spelling of his name.
But on first glance, it looked like this person had been in New Hampshire for the past 20 years.
Stumped, we went back to the client to get more information, just to make sure we didn’t end up on a wild goose chase. The client didn’t have anything else, unfortunately, but acknowledged that the spelling of the last name might not be accurate.
The client also said in no uncertain terms that it was essential that we find this person.
So we went back to the drawing board, and after some digging around, we found the right guy.
Here’s how we did it.
Eliminate the Wrong Name
The first thing we needed to do was to confirm that the original name we were provided—let’s use Joshua Simmens (not his real name)—was not accurate. We searched multiple databases and confirmed that there was only one person with the name Josh or Joshua Simmens in the country, with the exact spelling that we had been provided. We confirmed—as our initial research had suggested—that the person had been born and raised in New England and did not appear to have ever worked for the target company.
Search engines revealed no results either.
Certain that this was a dead end, we moved on to other avenues.
We then conducted some research with the name Joshua Simmons within 60 miles of the Midwest town. It seemed to us when we had first received it that the name Simmens was an obvious misspelling of Simmons, which is certainly a more common form.
So we set out to see how many people lived within commuting distance of this Midwest town. We were lucky in the sense that this was a pretty small town; if the person was in New York, we probably would have come up with dozens of results. Luckily for us, our search gave us two possible leads: a person who had died in 1988 and a 31-year-old who lived in the town in question.
Although the 31-year-old seemed like a decent candidate, he was probably too young to be an executive. Further research found that the 31-year-old did not have much of an online presence—another yellow flag that he might not be the person we were looking for.
So we pressed on.
Sure, having a list of employees at the target company at that particular location would be helpful. But despite what binge-watching CSI might have taught you, there really is no way to get a list of employees that worked at a company. The human resources department probably has that list handy, and the IRS may know, but neither of those things are readily accessible.
Nevertheless, we spent some time mining LinkedIn, resume databases, social media, news media, and investigative databases to see if anyone with a similar name had ever worked for the target company. That got us nowhere either.
A nice and handy little trick is to search for everyone who has ever been affiliated with a particular address. Investigative databases derive their information from credit header information, so if a person had applied for credit using their business address, there may be something linking the person to that address. So we searched for anyone who has ever reported the address of the target company.
Unfortunately, there was nobody at that address with even the right first name.
It was a long shot, but we’ve found people that way before.
Convinced that we had the wrong spelling, we took it a step further. We could have gone through every name variation that we could think of, but that would be costly and time-consuming.
So we did the next best thing.
Many investigative databases, such as TLO, IDI, and IRB, have a “sounds like” feature, where you can input a person’s name and it will find names that are similar in sound. So we put in “Joshua Simmens” and got a bunch of wildly different results. Clearly, each of these databases had a different algorithms for its “sounds like” feature.
None of the options were jumping out, though. Ages didn’t match up; location didn’t seem right.
So, using the “sounds like” feature, we changed the search name to “Joshua Simens” and came up with a bunch more results. This time, one name jumped out: Joshua Simenz, who lived in the right town during the right time and was about the right age (early 60s).
A quick Google search confirmed that we had the right guy.
Some good reminders…
- “Garbage in, garbage out” is a saying often used in computer science. In other words, bad input will result in bad output. Computers operate using strict logic, so inputting bad data will generally get you a bad result.
- Never underestimate the power of human error.
- Computers are not a substitute for thinking.