Did you know that you can navigate the posts by swiping left and right?
Food plays a huge part in every person’s life. For me, it plays as big of a role as, and even possibly more than, friends and family. It brings us together. It tears us apart. Given its importance for every human on Earth, I thought I’d share some of my thoughts surrounding it.
We all spend inordinate amounts of time pondering about or interacting with food (“Hey Mr. Avocado, what’s goin’ on pal?”). In the middle of the afternoon, for instance, it may not be an uncommon occurence to find yourself asking, “what in the Lord’s name am I going to make for dinner tonight?” Sounds familiar, doesn’t it? If not, perhaps this next one’ll resonate with you. “I wonder how this ranch sauce is gonna taste on this curry chicken leg?…Damnnn, that hit the spot.”
The issue is, I really don’t enjoy spending my time thinking about finding or preparing food; I’d much rather be spending my time writing essays like this, coding, learning or doing something seemingly creative or productive. The only reason I think about food is in anticipation of how it is going to feel in my mouth and my belly. Well, I guess that isn’t the only reason, but it probably constitutes like 95% of my time spent thinking about it. The other 5% of the time I’ll admit to getting curious about the chemistry of food, how it all fits into the bigger picture in the world, etc. I desperately want to optimize my food intake strategies though, without sinking as low as ordering soylent. Basically what I want to do is maximize taste, health, entertainment, and minimize price. It’s a dynamic optimization problem that differs from person to person. Most of this post is simply going to be about ordering food and frequenting restaurants, since preparing food is a different kind of beast that I’m not prepared to tackle at the moment.
I’ve always thought that it’s far easier to order food when you know what it looks like in advance – and here’s where pictures come into play. I can remember a gruesome moment from my childhood when I was in Saskatchewan at a family restaurant (it doesn’t get much gruesomer than that), and an item on the menu called an open-faced sandwich tickled my fancy. Only one slice of bread, I thought to myself – now that would be the perfect amount of bread for me right now. Little did I know it, I was on the verge of being presented a gravy-soaked loaf, and if you know me well, you know I’m far from being an advocate of soggy bread.
Come to think of it, I think restaurants would be more successful if they used pictures for each of their food items, and here’s my reasoning. When people read a menu, they convert the descriptions into pictures in their head, and then decide whether or not those pictures in their head are desirable. Given that there is added error in the process of converting a menu item description into a mental picture, to reduce the error, why not cut out this mental process entirely? Then there would be a tighter mapping between people’s desires and what they actually wind up ordering. Customers would be more likely to order what they want, leave the restaurant feeling satisfied, thus increasing the likelihood of them returning. So there you have it restaurants. You have no excuse to exclude pictures for every one of your nontrivial food/drink items on the menu, because including pictures should, in theory, increase customer retention.
Admittedly, restaurants may have other (illegitimate) reasons for not including pictures on their menu. One potential reason I can think of is that at the time of crafting their menu they didn’t want to invest the money in taking quality pictures and spending the bucks to print some quality pics. Another reason is that maybe they’re an upscale restaurant, and for some reason holding such a status precludes them from any obligation to showcase pictures of their food; as a consumer you’re just supposed to trust that a classy establishment is going to serve you right even though every person’s gustatory needs differ. One last reason I can think of is that there may be some fun involved in painting a picture of the food in your head based on the description, which contributes to the fun of going to a restaurant. Out of the 3 reasons listed, I really only think the last reason has any merit; that said, I think most of the fun is not due to forming a mental picture of the food in one’s head, but rather can be chalked up to reading the descriptions of the food aloud with friends – “Jerry, we all know your mouth is watering over this maple-glazed salmon on the menu right here. Don’t try to hide it Jerry, you’re not fooling even the biggest airhead at the table – yeah, that’s right, I centred you out Winston.”
Okay, let’s summarize so far. I’ve conceded that yes, there is some fun that food descriptions bring to the table. But having a cute description for a menu item shouldn’t preclude it from also having a tantalizing picture. Going back to my open-faced sandwich example, there’s simply too large of a cost that can come with not showing a picture alongside a menu item. Another example where expectations can sometimes misalign with reality, resulting in resolute chaos and destruction, lay bare in the dimension of quantity. Imagine you order something on the menu which sounds incredible but when it arrives at the table there isn’t enough food there to feed you, let alone a 12 year old girl from Saskatchewan. Nobody wants to go into a restaurant and leave still feeling hungry. I guess on rare occasions you’re pleasantly surprised with the quantity – good gracious, I never knew a boatload of nachos literally meant they would serve us an aluminum boat filled to the brim with nachos – and bonus, we get to keep the boat! Most of the time, a mishap in the dimension of quantity leaves you with a bitter taste in your mouth, and possibly even stewing in anger upon the realization that you’ve been taken for all your marbles. You get into your car, both you and the exhaust just fuming, you engage in some good ol’ fashioned road rage, ultimately making the world a more hazardous place to live in. This – all because a restaurant took a simple shortcut, deluding themselves into thinking it was okay not to include pictures on their insignificant menu. I guess the bottom line, and what’s clear here is, I don’t like to mess around when it comes to food. When I’m hungry, I don’t care to paint pictures in my head. I just want some good grub hot damnit. The other take home message is that food pictures on menus save lives by preventing traffic incidents.
Moving from one hood to another in Toronto this past year has kind of changed my perspective on food. Every time I move to a new neighborhood, there’s such a huge number of restaurants nearby it’s imperative that I discover the best deals and tastes near me as quick as possible. Otherwise, I’m missing out on good times, a thicker wallet, and the local foodie culture. People often use restaurant deal nights as an excuse to hang out, so therefore, knowing the deals results in a richer social life too. This leads me into a discussion about current software out there that can be used to discover restaurants, and, if it were socially acceptable, find others with similar tastebuds to enjoy a meal with.
I’m an avid Yelper. I generally write either scathing reviews or glowing reviews, but I pour my heart and soul into every one I write to try to communicate the most meaningful information possible to other users out there. I’d rather sample restaurants several times before writing Yelp reviews, but I simply don’t have the luxury of that kind of time and money, so the data I’m contributing is ostensibly a little noisy. Besides, it’s Yelp’s job to aggregate over many users and smooth out the noise contributed by its users. Without doing any data mining I can safely say that the majority of average restaurant scores on Yelp are between 3.5 to 4.5, which makes me question how meaningful Yelp’s 5 point scale really is. It would be nice if there was some way to apply rescaling methods to average restaurant reviews on Yelp so that they map closer to human perception, but I’m not quite sure how that would be done. I mean, how is one supposed to look at restaurant A which has a score of 3.6 and restaurant B which has a score of 3.8 and, at a glance, conclusively say that they should invest their night in going to restaurant B? Where’s the margin of error? Maybe listing the variance would be meaningful. Actually, in this case simply plotting variance probably wouldn’t do too much, but seeing the distribution of ratings would certainly be informative.
I think Yelp would be a lot better if, instead of it being centred around restaurants, it were centred around food items. They could even add a feature for swiping left or right on various food items within a given region (I’m only slightly being sarcastic here). I rarely go to restaurants because I’m fond of its environment and the entire menu, but rather, I go there because I’m craving just a few things on the menu. I guess it really depends on the social situation, because if you’re going to eat with lots of people, then it’s more important that the menu is diverse and has a high overall quality to cater to everyone’s gustatory needs. Another reason I go to a restaurant, as previously touched upon in this article, are the deals that are offered by select restaurants. It seems silly in this information age that people should have to call up restaurants to find out their specials before visiting them. Both consumers’ and restaurateurs’ time are being wasted whenever such an exchange takes place. At the same time, if a website existed that posted up to date information about all deals offered by every restaurant, that could add additional maintenance cost to the restaurants in order to stay competitive, while putting more power in the hands of the consumers.
You’ll notice in the previous paragraph that I started listing attributes that motivate me to go to a restaurant (deals, good food, the ambiance). These reasons are obviously highly idiosyncratic and so restaurant recommendations should be based upon a person’s preference for these different attributes. I have a spreadsheet myself where I grade each restaurant on its taste, health, classiness, proximity (how close it is to me), price, and whether take out is an option. I assigned myself a weight for each of these attributes on how important each one is to me, and with these I can compute an overall score for each restaurant by taking all attributes and their weights into account. There are lots of issues with this method, so I generally find the spreadsheet to be unmanageable. For example, the weight I assign to each attribute varies wildly throughout the day for each meal and it depends greatly on the context of the social situation. If I just want to stuff my face and don’t care about health, I’ll ramp up the weighting of taste and ramp down health; if I want to impress a colleague or I have a hot date, I’ll ramp up classiness, and suddenly proximity becomes less of an issue because exploring the city becomes a talking point. The spreadsheet method clearly has its issues, but it was fun working on and maybe there’s something useful there.
I could probably go on and on about this topic, but the goal here was sort of a stream of consciousness exercise to get these ideas out of my head and into the wild. I haven’t fleshed out many of these ideas as much as I’d like to, but I don’t want to spend my whole day doing so. In summary, here are the take home messages:
Extra Bits
While writing this article, I had a funny idea for a website/app. Imagine an app where you selectively broadcast to your friends your desire to get food. You would propose a few restaurant ideas that people can vote on, then a countdown begins with a time limit, encouraging replies from friends who are on the fence. If people don’t like your restaurant suggestion(s), they are given the option of firing off a counter-proposal, making a game of it. A restaurant would automatically be decided upon in this way, which would be a fun game to play in and of itself.