How Many Times Do I Have to Tell You My Name?

By on in with 4 Comments

Chris Brogan, one of the leading minds in the world of social media, just started a new job. And, he’s going through the new employee paperwork hassle (I think), which prompted this mini-rant:

http://attentionupgrade.magnify.net/item/BRCM0MCXMK2C3JG9

It’s interesting, as Chris is expressing the same frustration that we’ve all faced. WHY do I have to write in my name, date of birth, spouse’s name, etc. a gazillion times to do one simple thing?

He acknowledges that part of the reason that there isn’t a centralized, global ID that has gained widespread adoption is security concerns. And, that’s certainly part of it. The University of Texas managed to give my social security number up to some hackers several years after I graduated. They’ve now implemented much more stringent security…to the point that it’s a royal pain for me any time I need to conduct a transaction online with them.

Still, I firmly believe that people are overly paranoid on the security front. It’s a valid concern, though.

Another reason to push for a universal ID is from a data maintenance perspective. In the last six months, I’ve changed my home phone number and my home address. My job title, my name, my SSN, my mobile phone, and my work address have not changed (actually — that’s not entirely true; in some cases, my work address has changed). Various studies (none of which are close enough at hand for me to quickly track down), have shown how frequently individual and corporate data changes. Even doing the most simple of analyses of a company’s internal prospect lists will turn up duplication rates that are eye-popping (20% or higher), with no easy/obvious de-duplication options. And this is perfectly valid!

To have a universal ID, you need a universal key, and that is hard. SSN meets the criteria of being unique and unchanging…but only for US citizens. So, hardly universal.

A lot of applications try to use e-mail address, but that is wildly flawed. Chances are, you have at least two e-mail addresses — a personal one and a work one. But, there’s a good chance you have others: email forwarding addresses from your college or professional organization, “junk mail” e-mail addresses for online registrations that you think may spawn a lot of spam, or a gmail address that you need to get to your Blogger or Google Analytics account. And, e-mail addresses can change over time.

You can get people to pick a single e-mail address or ID, which, if memory serves, is what Microsoft tried to do with Passport. But, this was requiring people to remember that ID…and Microsoft was pretty weak when it came to controlling access levels for third parties to use that data. Not to mention, paranoia kicks up a notch in people who have decided that Microsoft is inherently evil (simply because they are large and successful).

In short, it’s a lot more complicated than, “I’m looking at five forms that all want me to fill out the same basic data!” Now, absolutely, an employee onboarding application should be able to make this electronic so, at a minimum, you fill in your name and phone number once. And, maybe that application could have a hook into LinkedIn to do a one-time prepopulation of the data for you to review and confirm. But, as much as companies would like it for that to be a permanent link to LinkedIn…it’s a messy, messy proposition.

And, things get even messier on the company front. Mergers, acquisitions, and spinoffs happen constantly. It is incredibly hard to track a single person through those changes. But, that’s where companies like D&B and Hoover’s, combined with customer data integration companies like Initiate Systems come into play.

Okay, so I’m rambling all over the place. But, I’ve been in this area for long enough and had enough people oversimplify the challenges of maintaining data integrity, that it seemed worth scratching the surface with a quick blog post.

Does all of this say, “There’s got to be a business opportunity here?” Well, yes and no. Anyone who tries to solve the whole problem is going to require a ridiculous level of funding and time…with a limited chance of success. Lots of companies already exist that are trying to carve out pieces of the problem and solve them.

In my mind, the best bet is to focus on standards for storing and transferring that data. I’ve heard rumblings of some attempts to come up with XML standards on that front. But, the W3C is a good example of where global standards for something “simple”…generally aren’t.

4 Comments


  1. Great points, Tim. I mean, I’m worried about how much data leakage would cost me. It’d be really ucky for someone to get all my stuff. But surely, if we can conduct financial transactions online between major banking institutions, there has to be some level of data integrity that most sides agree upon.

    I just want the same for my stock of personal data, and then I want to be able to launch that at various places, upon my will and my consent.

    Is that so hard?

  2. WRT social security numbers being unique, they simply aren’t. Ripley’s has a story of two women with the same name, born on the same date, being issued the same SSN. Other less peculiar instances are out there. They were never intended to be a universal identification number. A more proper vector for attacking the problem would be to use some biometric such as finger printing or retinal scan. Good thoughts all around from you and Chris both.

  3. Chris — with financial data, things are actually more simple. For one thing, you know exactly who it is you’re sharing your data with, and, for another, their reputation is staked on them keeping it secure. And, don’t you still have to keep track of your “account ID” in most cases — that’s a lengthy alphanumeric that you have to keep written down somewhere.

    An interesting exercise: make a list of all of the specific data fields you would want maintained centrally. Flag each one as to how often it changes (monthly, annually, never). Then, list 10 entities that you would want to have access a subset of that information. And, which information you would want them to have and that they would need to have. My guess is that you would come up with 8 different combinations of that data.

    Okay, don’t really go through that exercise. You’ve just given me something I’d like to noodle around with as a thought exercise.

    Jamie — thanks for the anecdote! I hadn’t heard of that. Having a plain name like “Tim Wilson,” I’ve certainly experienced my share of misidentification! Biometrics — that’s coming, too (but I’ve got zilch experience there).

  4. Pingback Gilligan on Data by Tim Wilson » Blog Archive » How Many Times Do I Have to Tell You My Name?…Redux

Leave your Comment


Notify me of followup comments via e-mail. You can also subscribe without commenting.

« »