Archive for the ‘Technology’ Category

Contract Negotiation: Case Study in Contingent Thinking, Behavioral Economics

December 27, 2012 Leave a comment

Whether in sales or strategic partnerships, a deal closes once a contract has been agreed upon. As expected, both sides are going to haggle over the nuances surrounding some of the greater business terms. At this point in time, negotiators are brought in to hammer out the final terms. Many would argue that agreeing on the final decisions is an art itself–this is indeed the opinion of many a “closer,” those always-in-high-demand negotiators who are able to harangue the best terms all while tying off the sales cycle quickly. Just as it is very difficult to quantify the distinction between mediocre and great salespeople, it is very difficult to pinpoint a great closer. Most people will say the distinction is intuitive–you just know when you see one.

From experience, I can say that the secret to negotiations is simple (but difficult to practice). The secret sauce of successful contract negotiation lies in understanding your opposing negotiator’s opposing point of view. In other words, the person who most completely understands their equivalent’s point of view–or to paraphrase Jerry Weissman, their WIIFY (what’s in it for you) – will best be able to broker acceptable compromise. Understanding one’s opponents’ WIIFY is difficult to do, but not impossible. What “closers” call intuitive is simply an organized plan of attack in haggling over terms.  In fact, a systematic framework for contract negotiation can be established by understanding the opposite negotiator’s key business goals; once one has understood the other side’s best alternative to a negotiated agreement, one can be in a better position to ink favorable business terms that seems acceptable to the other side.

The framework one applies in optimizing the best terms resulting from contract negotiation can be exemplified in a light case study of behavioral game theory, especially one that draws from the realm of contingent thinking. According to the main tenets of behavioral economics, people’s actions under certain time constrains can be categorized. Behavioral game theory details what people do when playing games (i.e. puzzles, prisoner’s dilemma scenarios, etc.). Normative analysis postulates what people should do (i.e. what people should do to maximize positive externalities). In contrast, positive analysis predicts what people actually do. So in scenarios where two parties are faced with two alternative decisions (in the trite prisoner’s dilemma scenario, “rat on partner” or “stay silent”), the best decision to take can usually be summed up by the Nash Equilibrium. In this situation, the negotiator, having fully understood his choices and those of his opposing party, has no incentive to deviate from the decision he ultimately chooses.  Nash Equilibriums are possible in certain situations, but they are rare, especially if the game is not zero-sum.  For instance, it may be difficult to pinpoint Nash Equilibrium if one plays the game over multiple rounds; in this new case, one’s choice can instead turn into a strategy.  As a matter of fact, a dominant strategy is dominant if it is the best response to any feasible strategy that others might play. In mathematical terms, displaying the utility function of dominant strategy is as follows:

If si = dominant strategy, then Ui(si*, s-i) > ui(si’, s-i) for all s-i, si’ si*

In a game where you can do multiple iterations, your choice of action providing the best result can be difficult to pinpoint unless one keeps in mind what one’s opponent decides. This is contingent thinking in its full form: one has to make a decision based on what one thinks one’s opponent will do. This is exactly what contract negotiation is: one has to hedge one’s bets by understanding what is valuable to the person one is in negotiation with, consequently choosing what to give in to and what to fight over.

The p-beauty game repeatedly tried in behavioral economic case studies best exemplifies the applications of contingent thinking to contract negotiation.  In the p-beauty game (Moulin 1986), all participants are asked to simultaneously pick a number between 0 and 100. The winner of the contest is the person(s) whose number is closest to 2/3 (an arbitrary p value usually picked by the mediator that falls between 0 and 1*) times the average of all numbers submitted. How do you win this game? By choosing a number contingent of what one thinks other people will choose. In this case, one knows that the highest possible mean is 100, so choosing below (2/3)* 100 is a dominated strategy (best strategy one can take considering what you think other people will do). One can do better by choosing as low as (2/3)* (2/3)* 100–in this case one assumes that everybody else had the same reasoning as you and picked (2/3)*100 as the average, so one has to pick (2/3) * (2/3) * 100 to win the game. The same logic applies when one considers if other people think of the aforementioned strategy a step further (as you just did), so one decides to take one’s iterated decision a step further and choose between (2/3)*(2/3)*(2/3)*100 and (2/3)*(2/3)*100. Extending this process N steps results in the following conclusion: the highest possible valuation of the mean is 100*(2/3)^N, so one can do better by choosing at least as low as 100*(2/3)^(N+1). If everyone practices this sort of contingent thinking, then the best possible number to choose is 0. However, we know in the real world, not everybody will be practicing this sort of contingent thinking (I sure didn’t when I played this game for the first time). So one has to take that in consideration, and hedge one’s bets by choosing a number a little more than 0 rather than exactly 0. That gives one the highest realistic probability that one’s chosen number will win.

In the p-beauty game, one has to make a realistic decision by contingently thinking of what one’s opponent(s) may decide to do.  This happens every day when one negotiates contracts. As Roger Fisher and William Ury explain in their book Getting to Yes, arriving at the best business terms comes from negotiating on merits rather than on positions. They argued that negotiators need to focus on the issues and not the people, generate a variety of creative possibilities that can be mutually agreed to, and insist that final decisions be made on a rational basis rather than arbitrary or emotional ones. Ultimately, when one understands the opposing side’s best alternative to a negotiated agreement–BATNA as Fisher and Ury write–then one can avoid negotiating from a bottom line, allowing for flexibility in negotiated terms that are favorable to both sides. Just as in the p-beauty game where one has to make decisions based on what other people will decide, the contract negotiator has to ask for terms that they think can be agreed to by both sides.

Ultimately a contract is a formal approval for a relationship to progress between two sides. The one lesson that I’ve taken away from my own experiences is the following: sometimes it’s worth losing some battles for the sake of preserving the relationship with your business partner. Especially if you are inking the beginning of a partnership, whether distribution or channel, the potential upsell will be more expedient if the relationship begins on good terms. One must  understand what your partner’s BATNA is for the contract negotiation to be fruitful.

Sadly, sometimes each side’s BATNAs is more attractive than any compromise, and one side walks. Sometimes preserving the relationships means discounting the potential utility from future endeavors in hope that the future discounted utility is greater than the present BATNA’s. The relationship is the most important factor to keep in consideration when contracts are being negotiation–sometimes not having that relationship in place is an unacceptable BATNA. This is a card that needs to be played close to the chest when dealing in hostile conversations.

The truly skillful negotiator will prepare for his séances with the other side by meticulously researching what the partner’s BATNA is, and strategize his thinking so that the resulting compromises is better than his own BATNA. Indeed, planning one step ahead–or multiple steps, as exemplified by the p-beauty game–separates the best “closers” from the mediocre.

[*if p=1 then there are many possible Nash equilibria, as there are multiple dominant strategies. ]


Why Harvard Did Not Get Schooled: Defense Against BetaBeat’s article, and Ramifications

July 14, 2012 Leave a comment

On July 10th, Kelly Faircloth wrote an article entitled “Harvard Gets Schooled: As Techies Flock to Stanford, MIT, Even Penn, Crimson Goes Green With Envy.” She argues that Harvard does not prepare its students to spearhead any sort of entrepreneurial venture. Faircloth continues that your typical Harvard student–who, according to Kelly, has never had to live a “disruptive” life–is ill-prepared for the vicissitudes of start-up life.

Faircloth argues that Harvard will continue to lag behind MIT, Penn, and Stanford in preparing students for start-up life. She concedes Harvard will not be ousted as a beacon of excellence any time of soon. Indeed, the exit opportunities for the graduating class of 2012 reflects that the Crimson keeps sending its graduates to the top consulting and I-banking institutions; many more enroll in top PhD programs and law schools. Yet Faircloth argues that Harvard students aren’t suited for start-ups. So to speak, they’ve been taught to color within the lines, not outside of them. The creativity and willingness to break the established norms of technology and society are traits common to start-up visionaries. These qualities are not usually associated with the over-achieving, type-A, extra-credit-on-homework-seeking Harvard students. Certainly, the institution has redeeming qualities, but Faircloth wonders whether a cultural shift  in the university’ framework can foster entrepeneurial spirits that seem to have sprouted more easily at Stanford. But while Faircloth mentions that computer science is becoming more popular in the College, and that working for start-ups post graduation is becoming sexier, her tone remains skeptical. Can Harvard really keep up with Stanford, MIT, and Penn in the quest to prepare students to become the next entrepreneurial visionaries?

While this argument is trivial, almost inconsequential in the larger scheme of themes (scroll down to my second last paragraph if you don’t want to hear about another trivial Ivy League vs Stanford argument), it still merits discussion for the sake of argument.Having joined a start-up immediately post graduation, and knowing a fair number of other grads who have done so, and others who have started their own companies, I believe that Faircloth’s argument needs to be qualified and refined.

I disagree first and foremost with her implication that Harvard kids are not ready for start-up life. Harvard teaches you how to compete with some of the hungriest, most ambitious young adults. You have to adapt quickly, and work in groups to improve your chances to score better on that problem set or exam. In the end the material you learn may not be the most useful–my focus field was intellectual history, and I built a background in microeconomics and game theory–but the methods by which you learn impact you later during your career. Four years at school taught me how to be resourceful and scrappy. This means that I have no problems doing anything and everything. And I am not the only one who has this mentality–conversations with some of my graduating friends agree that to succeed you have to do anything (within moral guidelines of course). This mentality, or this desire to do everything you can to put yourself in a position of success, transfers directly to start-up life. If you’re on the technical side, you will be willing to code all day, every day. If you’re on the business side, you will help your director of bd and do your own competitive research to locate potential partners, or you’ll hit the pavement and source your own deals. In either case, Harvard grads will be creative enough to adapt their analytical skills to the grind of start-up life. In essence, Harvard kids are ready for start-up life because they already have this inner motivation that drives them to go the extra mile.

Faircloth also argued that Harvard does not have the same types of connections to Silicon Valley that are available to Stanford students; therefore, in a couple of years it will be difficult to locate Harvard grads in the up and coming companies of the Valley. I believe Faircloth’s comment should be qualified. From my experience, it was a bit more difficult to connect with entrepreneurs on the west coast; I had to use my own network to find out which start-ups were hiring. As someone with a background in business, not engineering, it was even more difficult to find smaller companies looking to expand their bd or strategic partnerships departments. Harvard’s network is a great resource, do not get me wrong. When you’re in Silicon Valley competing against Stanford and Berkeley kids, you can raise an eyebrow or two with Harvard written on your resume. Nonetheless, drawing from my experience, I would agree with Faircloth’s argument that Stanford is at an advantage because it is so close proximity to Silicon Valley’s human capital–that tight-nit community is hard to come by unless you are within at the epicenter of it all. Harvard just doesn’t have that.

Let’s address one of Faircloth’s more salient points–can Harvard adequately prepare their students to understand technical concepts and prepare them to build software products? Certainly, MIT and Stanford are consistently ranked higher than Harvard in engineering school rankings. However, Harvard still has all the resources in place to train the new generation of engineers and product managers. There is no question about that. I want to qualify Faircloth’s argument because from my inside perspective, I realized that Harvard does a great job training generalists. The beauty at Harvard lies in the way a student can major in a liberal art and get a secondary field in a technical field, thus balancing a broad analytical skill set with a technical know-how. The counter-argument might edict that any of Havard’s rivals could do the same. To a certain extent, yes, but I think that this concentration-secondary field combination is more popular at Harvard than at the other schools. Product managers in particular need to have a fine business acumen all while being versed in a technical training–the dual study that Harvard enables is a formidable preparation for that. (again this train of thought comes from my own personal experience…if you have any statistics saying differently I’d love to see them).

At the end of the day, this discussion is trivial. Does anyone really care if Harvard is better than Stanford or vice versa? And let’s not forget that they’re not the only schools that send qualified talent to Silicon Valley. Let’s stop pretending that these institutions are the jewels for preparing businessmen and women and engineers to enter the real world. I wanted to qualify Faircloth’s argument. Things are never as black and white as they seem, and Faircloth does a great job in nuancing her points. I’ve tried taking that a step further. But at the end of our discussions, we have to realize–and by we, I mean you, the reader, and me, the writer–that there is more to life than the institution that prepared you. These types of meta arguments are trivial, and make us roll our eyes.

But I hope that this discussion shows us that Silicon Valley highlights that we should look past the brand on the resume, as in the end, it doesn’t matter where you went to school, but whether you can deliver. Whether you’re a startup or an established software company in the valley, you’re only as good as the product or services you deliver. My argument might be inconsequential, but its ramifications are not.

Categories: Harvard, Technology

It’s All About the Parties: Making Location-Based Service Relevant for College Kids (Part II)

January 9, 2012 Leave a comment

Let me touch on some comments made regarding the first post:

1-I received a lot of comments in regards to the way I sized the college student market. I sized it deductively not empirically. The point of this exercise–and this blog–is to flush out ideas and thinking. I could have simply gone to US Census and picked up the numbers of college students in 2011-2012. So to cross-check my numbers, I did. There is about 19.7 million college students attending two- to four- year institutions (I’ve embedded the excel document downloaded from the US census website). That’s not too far from my estimate of 22.8 million…

US Census Data, 2011-2012 Americans attending two- to four- year institutions

2-I’ve been asked: why should Foursquare care about college students? I discussed the subject with Adam Bordow, co-founder of Lasso, a creative mobile app startup based in LA. At first glance, Foursquare actually excluded college students from their overall strategy. He mentioned that Dennis Crowley had originally developed Foursquare for his own reality of NYC, where there is a need for real-time stream of location-based interactions. There simply hasn’t been a need for one on college campuses. “There is simply less to be discovered on a college campus which would render Foursquare less potent, and even less irrelevant if that distribution channel was pursued,” Adam emailed me. Agreed. For small colleges located in the middle of nowhere, there is no incentive for Foursquare to extend their network so that the small student body can check into the two coffee shops on their campus.

But think of larger schools located in prime cities. Harvard-Cambridge/Boston. Stanford-downtown Palo Alto and its plethora of stores, restaurants, coffee shops. Think of the larger metropolis: SF State-San Francisco. NYU or Columbia-NYC. UCLA or USC-Los Angeles. For the students attending those schools, there  is plenty of opportunity for those college students to check in at bars with their friends and get free new deals for food and beverages (what I would do for a Foursquare deal for John Harvard’s in the Square to get a discount for pitchers of beers…especially if the Niners go to the Super Bowl…it’s not like I’ve been living in Boston for the last four years getting bullied by my roommates, aka the Patriots/Tom Brady uber-fans…what better way to rub it in their faces while buying discounted pitchers of beer and batches of hot wings? Just saying).

Foursquare could go in a different direction with college students. Foursquare already has had a difficult time incentivizing check-ins. That’s not the case for merchants. Foursquare can easily flash their 15 million users when they’re trying to garner SMBs’ interest.  These mom-and-pop shops are in the early stage of their businesses, so if they can direct potential customers to their shop at a very low price, they’ll partner with Foursquare. But for younger users, Foursquare is having a more difficult time incentivizing check-ins. When you first start using Foursquare (i.e when you check into Quedoba for the first time), you can get a newbie discount for the good being sold. But perpetual users seem to be having a more difficult time unearthing discounts. For instance, Adam has checked in 445 times and has only redeemed 2 deals. I’ve checked in over 110 times and I haven’t received notice of any sort of deal. I think Foursquare needs to offer a different sort of incentive for college students–or, as a matter of fact, any of their users, but that’s a different subject altogether–if they are ever to start using Foursquare. So how does Foursquare reel them in?

Offer real-time check-ins for parties.

Fact: college students go out. Even at Harvard  (trust me, it happens more frequently than you would imagine.) More than once, on a Saturday night, one of my friends asks/emails/group texts “Where’s the party tonight? Where’s everyone meeting up?” If Foursquare could leverage its real-time check-ins, college students could know where their friends are partying. Whether it’s a bar in Boston, a final club on Mt. Auburn street (, or a party in Eliot House, college students will check-into the parties, write in comments (“sweet party”, “too many bros”, “rough scene”), and allow for their Foursquare friends to see where they are. Even if it’s a random Monday night and you want to go Karoeke at a bar in Cambridge, Foursquare can alert your friends that you’ll be there. Moreover, if those places haven’t been created on Foursquare, you can create them yourself–just as you can put your house on the map in Foursquare, enabling you to check-in (and become mayor right away), you can create new locations where you might go out and party.

Remember the the end goal isn’t to offer college students rebate deals: we want them to start using Foursquare. Let college students party, and let them brag digitally about how sweet their time is–and once their friends see that,  they will hop on the bandwagon, and hopefully become new Foursquare users.

Indeed, there is potential for a location-based service to streamline one’s social life. Putting it like that sounds sad, but how would we describe what Facebook and Twitter have done in the recent past? They’ve made social interaction online basically seamless and transparent…why can’t Foursquare do the same when it comes to connecting people when they go out?

I’ve been operating under that hypothesis for these last two posts because I feel that if Foursquare can get more users, they will have more data available to them regarding the tendencies of the people checking-in. From there they can figure out how to use that to their advantage and monetize their product. But it doesn’t seem like Foursquare wants to go in the direction I’ve outlined in these last two posts. As Adam pointed out, Crowley recently said in, “People know us [Foursquare] for check-ins, but with the data, we way to push people more toward the recommendation engine and the way to do that is make that prominent and a big part of the app.”

Yet…when is data convincing enough? When you have truck loads. How do you get truck loads of data? You get more users.

Plus it wouldn’t hurt if I were able to check into John Harvard’s, order a cheaper batch of buffalo wings, and feel good about myself when the 49ers win the Super Bowl. Just saying.

PS Many thanks to Adam Bordow, co-founder of Lasso. Follow him on Twitter @bordow or check out his creative website at More on Lasso soon–stay tuned for more about that!

Categories: Mobile, Technology

“You’re on Foursquare, bro?”: Why Foursquare isn’t relevant to college kids, but should be (part I)

December 22, 2011 Leave a comment

“You’re on Foursquare?!?” My buddy gasped the other night. “What’s the point of using Foursquare, bro?”

He had a point. Out of pure curiosity, I recently started using Foursquare a couple of weeks ago. For so long I had heard great things about the mobile platform…but it had yet to be relevant to me. So I decided to check it out a couple of weeks ago, and started using Foursquare as often as possible. I checked into “Chateau Popp” as often as I could; checked into every library at Harvard every day; lined up some sweet badges (I unlocked the Bookworm badge in no time); and next thing you know, I had racked up a couple of mayorships, and even unlocked a deal at Chipotle. So now that I had become a pretty avid user (can you say 143 points in 5 days?), I asked myself the same question my buddy asked me: What’s the point of Foursquare?

Foursquare is an incredibly simple to use product with great design: right there we already have the great first step of a successful startup. And Foursquare knows that: with 15 million users checking in around 1 billion times, it has already garnered a lot of success. Step #1 for startups: build a easy-to-use, well-designed product that gains traction quickly. Step # 2: worry about monetizing it. So far Foursquare has done a formidable job with Step # 1. As for Step #2…well, if my buddy’s comment is any indication, the fact that Foursquare appears to be irrelevant–I don’t think we could say the same thing about Facebook or Twitter–makes Step # 2 a far reach (at least for now). Foursquare should be worried by Facebook’s recent integration of Places into its platform. And considering that location-based services is becoming a more crowded field with the likes of Gowolla, aka-aki, Loopt, and Rumble (to name of few of the hundreds), Foursquare needs to realize that while it has a great product, it needs to position it more strategically.

I’m not saying that Foursquare needs to rethink its business development strategy. Instead, I’m saying that in the short-term, the location-based service provider needs to make themselves more relevant day-to-day. Foursquare has a great product because it adds another dimension to social media.  For Foursquare, Location, location, location is key–at least that’s the paradigm the New York based startup is trying to inculcate within the larger matrix of social media technologies. Just as Facebook made wall-to-wall posts and photo sharing the first pillars of social media websites, Foursquare is trying to make location another cornerstone of  web-based interaction. I hope that Foursquare succeeds in this endeavor; while I know that location-based social media can be construed as simply another voyeuristic dimension to already narcissistic set of technologies, I think Foursquare can make a meaningful addition to this field. The billion dollar question is: how exactly to they do that? Have they already done it but they simply haven’t gained traction yet? Or do they need to change their outlook on the market?

So where should they start? My answer: the college-age demographic. In a series of posts, I will be looking at how big that market is, why Foursquare should more aggressively pursue that demographic, and what exactly Foursquare should do to gain more traction in that untapped market.

How many students attending 2 to 4 year institutions are there in the US? Let’s make some assumptions. If there are 320 million people in the US, and assuming that the life span for Americans is 80 years old, then there are 4 million people per year (aka 4 million 21-one year old, 4 million 5-year olds, etc). If we assume that the ages of 18-24 constitute the ages of college-level students, that’s 28 million people. However, since we know that not everyone in each year goes to college, let’s assume a 60 percent penetration rate for each year (aka 60 percent of 20 year olds go to college, 60 percent of 21 year olds go to college, etc), that’s 2.4 million * 7 = 16.8 million people between the ages of 18 to 24 who are in enrolled in either 2 or 4 year college. Let’s also not forget those who are returning to school-out of the 60 million people between the ages of 35 to 50, 10 percent are going back to school (6 million). so 16.8 million plus 6 million=22.8 million students who are attending 2 to 4 year colleges in the US. 

Foursquare has 15 million users…and there is a potential market of close to 23 million? Out of the 15 million users that Foursquare already has, even if we assume that 30 percent are college students constitute that 15 million already using Foursquare, that’s still 23 million-5 million=18 million potential users that Foursquare can tap into. Even if they only capture 20 percent of that market, 3.6 million new uses increases its user base by close to 25 percent…that’s not a bad showing. If Foursquare wants to become more relevant, it can very simply attract more users. With an even stronger user base, who knows what will happen here?

Great product, great design, strong user base…so why isn’t Foursquare as relevant as Facebook or Twitter?

Categories: Mobile, Technology