Throw a rock in Hong Kong in any direction. Chances are, you’d hit a stock broker. Or his accountant. It stands to reason then that someone would finally open a stock market themed bar modeled after the kind seen in Barcelona, New York City, and London. The one that’s opened in Hong Kong is called Wolf Market, and follows along the same line of thought that governs the other stock market bars; prices for spirits start off at a low, but the more guests order a particular drink, the higher the price rises. According to the press, the drink prices rise until they hit a certain limit, at which point the ‘market’ crashes and brings all the drinks back down to a low.
By allowing the customers some input, the bar owners are applying a popular marketing philosophy that has been the core of many tech start-ups as of late. Referred to as ‘gamefication’, the concept taps into people’s inherent desire to win at a game so as to drive them to consume in a way that suits your needs. The age-old happy hour model is the most basic example of this: people don’t drink as much before 6pm, so by offering, say, a 20% discount within the classic happy hour time slot, you encourage guests to ‘win’ by drinking during a quiet business period and getting a good deal on drinks. A slightly more nuanced version is the happy hour offered by Stone Nullah Tavern, where the price of a drink start at $1 and double every 20 minutes between 5-7pm. In this case, a guest can ‘win’ by coming (and drinking) early, but ‘win’ bigger by coming even earlier (and drinking even more).
Wolf Market’s gimmick differs little from Stone Nullah’s Beat-the-Clock happy hour model, with of course a few differences. Of course, the big jackpot reads ‘If you and your friends drink hard enough, we’ll reward you by crashing the market and resetting the prices’, but there’s more to it then that. The players, through their consumption, collectively influence the price of drinks at an individual level. A rational drinker (though one might ask if there was such a thing) would then drink his favourite spirit until the price rose to a point where the savings from switching to his second favourite beverage would outweigh the dollar premium of his first choice of drink.
To put it simply, this system has the potential to quantify the dollar value for how much more some people prefer one brand over another.
Just to be sure, read that line again.
Do you know what that kind of data is worth? For a spirit brand or a market research agency, reliable data of that kind would be priceless, offering brands a way to quantify their brand market value, especially in comparison to other competitors
But in order to extract that kind of data, Wolf Market, or any bar looking to cash in on the secret potential of this gimmick, needs to fulfill certain conditions.
- the bar must stock at least three different varieties of spirits per category, per price bracket
- the pricing algorithim must be sensitive enough to vary at least $5, and preferably $10, within the duration of one customer’s visit
- the system must be able to track the sales of items to each table in particular
- the pricing algorithm must be able to affect the price of both individual items, as well as broader categories based on demand
- the pricing algorithim must be able to bring prices of low volume items closer to high volume items
- the bar must limit one order of drinks per guest
To go through each point, the 1st point is required to draw fair conclusions about a brand versus a brand of similar stature. For example, if the menu consisted only of one type of cheap, medium, and premium vodka each, the only conclusions that can be drawn from this set of data is how much value a customer puts into a brand versus a cheaper option; i.e. how price sensitive the customers are. However, once you place multiple options within the same price bracket, an analyst would then be able to tell how much more a brand is worth compared to competitors in the same market. An example would be to place Belvedere, Grey Goose, and Stoli Elit at the same initial price point to compare brand values. If all start at $100, and a guest starts off with Belvedere until it reaches $130, switches to Grey Goose until Grey Goose reaches $115, and then completely ignores Stoli Elit and moves to Absolut Blue at $80, you can see that to that one guest places the relative brand values of Belvedere, Grey Goose, Stoli Ellit at +$30, +$15, and -$20, respectively
The second point enables an analyst to be sure that the data both bears useful comparisons, and is representative of price sensitivity and brand preference of individuals. For example, if one guest sees only $2 differences in the prices of their drink within the duration of their stay, it may not be enough to sway them to switch to a different drink, thus making this data sample useless. $5-10 on the other hand, may be enough to cause a guests behaviour to change, creating a useful data sample. On a similar note, if the overall change in price swings $15 or more, but happens long after a particular table has left, then you are not tracking a single person’s perception of a brands value, but looking at those of different individuals, which is not as useful.
This leads on to point no. 3, as you would need to be able to look at sales data within the same group in order to make valid comparisons. By looking at purchasing patterns of group by group, you ensure you are not comparing the purchasing patterns of two groups of people that could be from wildly different demographics
Points 4 is intended to maximize the possible comparisons that can be made. For example, if a mid-tier bourbon is increasing in price due to high demand, so too should the price of both the cheap and premium bourbon (though of course the latter can increase at a smaller rate than the item that’s actually being driven up by demand) What this allows is for the following question to be asked: when the price of a guests drink becomes higher than what they are willing to pay for it, do they switch to another category, or do they switch to a cheaper item in the same category. By varying the rate at which the prices of items in the same category increase due to high demand in another, you can create various situations from which you can draw many different conclusions.
Low volume items, typically super premium spirits that cost more than most people are willing to spend casually, will see little useful data as there will be fewer data points. in order to increase the number of data samples, items that don’t move very much should gradually decrease in price to increase the chance a guest will purchase it. The reason why this is required is because there is a lot of upward movement in price, but there will be a certain price ceiling that many guests will not want to surpass, thus reducing the number of data points for items priced above that ceiling.
The last should be fairly straightforward. Guests ‘hoarding’ drinks when they reach low prices completely ruin data collection, on top of being a revenue sinkhole for managers.
Ultimately, from what I could see, Wolf Market possesses none of the features I just mentioned, which seems to me a real waste of potential. The possibilities of real hard data on brands could have raised huge interests from various parties and created a real revenue stream that far outweighs that of the bar itself. Whats left is a fairly transparent attempt to gouge money out of LKF party-goers while dangling the prospect of a ‘market crash’ before them.