Facebook showing Traces of Crowd Sourcing in Risk Management (?)

Picture by Matthew Filed/Creative Commons

If you're following the blog, you know I'm a big advocate of using the "wisdom of the masses" (well... at least their accumulated computational ability) to crowd-source complex tasks that cannot be easily automated. The way I see it, it's not that users merely "don't mind", they actually expect that to happen. This is the reason I'm pro offer walls (well, at least some of them) and like the concept of “jobs” or “tasks” incorporated into these walls. There's a lot to be done in the area of engaging users around various complex decisions, risk management being one of them (see other ideas on gwap). Now, I don't think that we cracked the code of making financials and risk interesting – whether it’s because financials are less “sexy” or because or more elusive reasons - but I do enjoy seeing interesting attempts.

That's why I liked the feature I discovered in a TC post: 

Yeah, I know, you’re wondering what I am so excited about. Well, for me it goes back to the dynamics that help establish and nurture communities. Online communities are here to stay, from Habbo hotel to SL to social networks. Communities like Facebook are growing by mere network effect; every day, people are pouring into the platform to interact, share, play. And at the same time, you can’t help but hear the murmur: Facebook did this, Facebook did that, I don’t like the new layout, I hate the privacy policy. This might means that we have (potentially) passed the docile stage of throwing sheep at other users, to the involvement period. What’s that? Basically, creating a real, lasting online community requires more than a news feed and a constant unedited stream of brain farts (dad, I actually like yours. Really). It requires users’ engagement, their involvement in regulating their environment, in setting its rules and in actively helping to make it better. It requires some kind of ownership, a sense of responsibility. This is what creates a healthy community that can be actually leveraged as more than a collection of unrelated, though somewhat connected, individuals. And that’s the reason why I like the potential of this nascent form of crowd-sourcing risk management: from my point of view, it’s a fair attempt at starting to enable users to assume that kind of responsibility. It’s a call to action where Facebook’s Risk team, effectively the police in a network that’s around 1.5X the size of US population, is asking you to join the neighborhood guard. If it’s really your neighborhood, won’t you act to keep it peaceful?

That’s why I like it. Or, at least, that’s the potential I’m loading on one poor notification feature… The other reason is, of course, the poetic justice of using the same type of resources fraudsters are using to overcome standard risk controls to actually deter fraud. Gotta love that.

What is your take on crowd sourcing risk-related process in your system?

PS

In case you’ve never seen it, catch this remarkable piece of the performing arts.

Lyrics are here.

And, of course: what goes around comes around. 

New York under zero: some thoughts on the Engage! Expo

"If there are any Mattel engineers in the audience, the astronaut Barby's space suit is not crash proof" (loose paraphrasing on Will Wright's keynote)

Yep, the keynote was entertaining and Engage brought a lot of vendors to snowy New York's Javits center. The two day event, though a bit low on developers, had a few interesting sessions and some interesting chances to share opinions. So what did I pick up from these two full days?

Payments and mobile

This Engage was heavy on payments companies, and by payments I mean mostly - if not exclusively - mobile payments focusing on SMS billing through carriers (obviously Paypal was there - a few of my colleagues and me - and additional sponsors). While the value of mobile payments for a streamlined, high conversion purchasing experience is clear (on the verge of overstated), the abundance of these companies over such a small space only served to emphasize how not-that-different these companies are from one another. Better coverage, low fraud and a promise for lower fees in 2011 were the value propositions.

Now, while I think mobile payments are clearly an avenue the industry must pursue, it was clear to me that until operators make a big leap of faith to embrace mobile payments, this field will not move much unless the companies themselves move to a Zong+ like, account based system that allows users to add a financial instrument and for the mobile payments company to charge it directly. And, as you are soon to find out, account based systems are a whole new world of pain - while with direct billing you charge a prepaid or underwritten balance an operator is liable for, accounts are a much more complicated structure. Plainly put, you start writing big fat checks directly to fraudsters' pockets. Looking at chargebacks in hindsight, as at least two of the participants suggested, just doesn't cut it. So mobile payments are looking for the next big breakthrough, and if fees don't drop soon (and they probably won't), I'm expecting some M&A work as competition heats up.

Offers and tasks

I'm a long time advocate of offers. Yes, offers have their "dark side", when misused, however they have a huge potential for creating incremental volume - something I personally love. When at the conference I heard that Offerpal are integrating tasks from Amazon's Mechanical Turk, and have been hearing assertions that competitors are going to follow suit (also heard it on stage from IMVU.com's CEO). Why is this good? I think that using social gaming to crowdsource simple but human intensive tasks is good for user education - do something good instead of just signing up for Netflix (nothing bad about Netflix, though); plus, it's good for the potential work providers - ideally, research institutes, advanced OCR services and others. In short, tasks are the new "green". Two caveats in this optimistic view, though: the first is that there is a serious chance of shortage of tasks, at least until this market picks up; the second is that abusing this model is still doable, maybe even easier than standard offers - if I were a fraudster, I'd immediately outsource my CAPTCHA operation to Amazon. Oops! Better read previous posts and do some risk analytics, guys, or you'll find you're breeding an ecosystem of thieves.

Zero cost of goods

I had this feeling in the past, but the conference reassured me: the "zero cost of goods produced" concept is both a blessing and a curse. Why a blessing? Because developers, bathing in the sensational bliss of high margins, were keen on trying new things - new business models, new payment options (30% take for mobile payments? come on) and various experiments in user interaction (offers, vanity items and many other really cool stuff). Why a curse? Because the notion has outgrown its proper boundaries, actually harming some of the developers. Assuming that if you just auto-refund your zero-cost virtual good, the problem of chargebacks goes away is a mistake, and not checking operational costs related to this "zero cost" work will make your bottom line look pretty bad eventually. Additionally, zero cost of goods got many developers focused on solely growing their user base and ARPU - both important but, as a few speakers noted, shifted attention from a few other very important stuff. Like fraud, like going international, but also like pricing - when the third pretty senior person suggested to developers that going all-in on a freemium model just isn't a good idea, I started to understand that the problem transcends risk management and controls; it's starting to detach companies from sound business judgment. So this is probably time to reconsider - it's all a part of growing up as an industry.

P.S. One last thing

I was delighted to meet a few young and talented entrepreneurs working exactly on the things I find exciting - namely p2p trade and new, great ways to engage users. It's fun to see how ideas evolve, and I'm looking forward to hearing more about them and others like them. Well done, guys!

Offer walls and marketplaces: the real alternative to "scamville"

Let me just say one thing up front: well done, Mr. Arrington! From the first clash with Offerpal (former, it seems) CEO Anu Shukla, through this post and others, there's been quite a stir around offer walls and the big question of the legitimacy of their offers (some news sites in Israel literally copied the post's words. But that's another type of scam). Beyond the provocation, there are a few actual issues here, that I think are left out since "scamville" and CEOs being replaced are much more sexy.

Here's the thing: if the social gaming industry is a viable industry (which I think it is) it should, at one point, start to mature as one. Maturing doesn't mean moving slower or becoming less appealing to users, on the contrary, there's still huge potential and a momentum so strong can't just be stopped by a few posts. But what it does mean is that you start getting attention for your mishaps and you need to start addressing this attention in a tone that is way, WAY milder and more responsible than just saying "this is sh*t and bullshi*t" (look here for some current thoughts of industry leaders and how I'd respond to them).

Fraud Fighting 2.0

“Wow, I've been a victim of fraud for 10 days and didn't even know it until now. Holy crap.” (A random Twitter user reporting)

During FraudSciences’ fraud operations days I was never keen on letting analysts and agents call people who were defrauded. Old school credit card users, who have had their details stolen, were never too happy hearing about it from someone they didn’t know, calling from another country and sounding like the fraudster himself - with a thick accent and all of their personal data at hand. It didn’t help that the company was called FraudSciences either, but that’s a completely different story. As time went on it became clear that most users we encountered preferred that fraud be dealt with out of their sight. They didn’t want to know about, or be involved in, any process regarding their identity being stolen. Sure, we’ve had the occasional angry customer calling back to understand whether we know the person’s name, who they were and their whereabouts to get even (and even had one person explaining that she always suspected her next-cube neighbor at the office), but generally speaking – no involvement. And we were completely fine continuing to work, undisturbed.