Corporate Matchmaking

Recently I've been thinking about an idea that was jokingly brought up at lunch at work a few months ago. At the time it sounded stupid, but now for some reason it sounds a tiny bit less so. Maybe I'm just crazy, so I decided to write up the idea to see if anybody can point out obvious gaping holes that I missed.

The basic idea is to write some simple "matchmaking" website software, specifically designed for internal deployment at large corporations, for use by employees who are looking to meet other single people in the area. The software itself would be simple to write since it's basically just a web front end for a database. The difficult part is to determine which features it should have (and whether it's even a good idea in the first place).

Before you laugh your way away from this page, allow me to list a few reasons why this might be a potentially interesting idea.

It might actually be feasible
Let's make up some assumptions out of thin air and assume that the target company has 50,000 employees, two-fifths of which are single, and 50% of those might actually sign up for the service if it's well-advertised and becomes popular. If it's just a simple casual profile-listing site, I think that last number might not be too unreasonable, given the success of similar sites like Facebook or MySpace on college campuses. So we're left with 10,000 potential users, or 5,000 potential matches for each person if for the sake of the argument we make some egregious assumptions about everybody's gender expression & orientation. Now let's cut out 90% of those matches to account for things like age differences, physical distance, personal preferences and any gross errors in the preceding assumptions. We're at 500 potential matches, which is roughly the number of matches returned from a match.com search for single females aged 20-26 in Seattle. So the basic idea isn't too far out there.

Most of the infrastructure is already in place
One of the main advantages of the fact that the software would be restricted to internal corporate employees is that we get tons of infrastructure for free. Namely, identity & authentication. Everybody already has a corporate username & password, so nobody has to sign up for an account. Depending on what kind of information we can pull out of the directory, the only thing a user might have to input would be their gender preferences & age. If the IT infrastructure is properly managed, the users just need to go to the page and the NLA will take care of authenticating them and signing them in automatically. We also get trust and verification for free: assuming security policies are enforced rigorously, you know that the person you're looking at is really who they say they are and that they work for the company. Since it's impossible to create new accounts (except by getting hired) you don't have to worry about scamming & fake profiles, among other things. It also becomes possible to ban misbehaving users much more effectively. The main downside to this is the loss of anonymity (see discussion below).

No financial pressure
The other main advantage I see is that unlike other matchmaking sites, money isn't involved and there's no pressure that it needs to make any money (from advertising or subscription fees). It's free for all users, which is a huge advantage since the one common thing among all people is that they don't want to pay for anything. It also allows for interesting possibilities on the management side since there's no pressure to maximize the number of active accounts at a given time. You want to remove your profile? Great: one click and it's gone. No "are you sure?" games. We could also do things which commercial dating sites would never have the guts to do, such as enforce gender load-balancing across the system by temporarily preventing members of the over-represented side from signing up until balance is restored. One of the biggest problems with regular internet matchmaking sites is that there are usually way more men than women, which isn't good for anybody. This sort of enforcement would be difficult to get right without killing the whole thing before it can grow due to over-restriction. Perhaps a 'voucher' system where you can get a free pass to sign up if you convince one of your female friends to sign up would work.

Good for the company (?)
This may be the weakest argument, so it comes last. It could be argued that if a company's single employees were provided with a matchmaking site that actually worked, they would all get matched up and would be happier. Their newfound happiness would lead to increased satisfaction and work productivity. The higher employee productivity would lead to rises in revenues and profits, which is what the owners (shareholders) want. Thus it would be good for the company, and it wouldn't be motivated to squash the service. Admittedly, the argument has a tenuous chain of inferences, and who knows if the time wasted by employees using the site would actually be greater than any savings due to increased happiness. I think this one could really only be determined through experimental testing.

Now let's take a look at some of the downsides:

Matches are restricted to the same company
This one's obvious due to the nature of of the thing, but it could be a big disadvantage that you couldn't get matched with people outside the company. Out of all the people in the world, what are the odds that the one person you will eventually marry works for the same company you do? Those odds might not be so good. Either way, it might still be a good way to meet new friends nearby, and you could grow your network out from there.

Loss of anonymity
Since the system works on the basis of corporate username, anonymity is lost from the beginning. Depending on your point of view, this could be a good thing since you gain trust and verification that people aren't pretending to be somebody else. But anonymity is important to many people, at least in the initial stages of dating, so this could be a deal-breaker. The system could be configured to present profiles anonymously, but then the users have to trust the administrator of the site, because he/she will have access to everything. Perhaps a hybrid model of selectable levels of anonymity would work.

The company itself might be gender-imbalanced
Even with a gender-balancing algorithm in place, it's possible that we would hit the problem that the input data (that is, the company's employees) is imbalanced in the first place, and thus we could never make everybody happy. For example, consider your prototypical high-tech software company, whose gender balance is 75% male, 25% female (and that's including sales, marketing, HR, etc). In this scenario, even if every eligible woman signed up, two-thirds of the men would be prevented from signing up at all if a gender-balancing limit was in place. This, combined with the fact that in such a situation many of the females might not be single in the first place due to the existing overabundance of males, might kill the idea. I haven't been able to think of any solutions or workarounds for this problem yet. Bad ideas that come to mind might involve some sort of Hot-or-Not ranking system where the 'top' 33% of the males are bubbled up to the top and allowed to join, while the rest are not (obviously, that would not work).

The whole idea is lawsuit-bait
I can easily envision a scenario where the Legal or HR department finds out about the system, squashes it like a bug, and fires the employee who hosted the server in his office. It would probably be a good idea to make sure to get buyoff from the appropriate management, HR & IT people before attempting to do something like this.

---

So those are some of my thoughts. It'd be interesting to see what other people think. Is this just another case of me being a programmer, so my first solution to any problem is to write an algorithm or program to try to solve it? (probably... see: hammer & nail analogy)