What The World’s Best Hamburger Taught Me About RevOps
And other thoughts about data.
Once in college I went on a date to see a concert down in Hollywood, I figured if we were going to drive all the way down there we may as well make a day of it and see the city. I’d driven through Los Angeles before, but never really spent any time there, and my date had just moved out from Illinois, and for people who aren’t from Southern California, the City of Angels is incredibly romanticized. The place we grabbed dinner at, Hamburger Haven, had a big sign outside saying “Almost Everyday, people tell us, ‘You have the best burgers in town.’”
“Alright Nic, sounds like a nice date and all, but this is supposed to be about data, not dates.” Yeah, yeah, yeah I’m getting to it okay.
I love quotes like the one on Hamburger Haven's sign. What does this quote tell us? What question is it even answering? It implies some amount of underlying data, but unfortunately, we don’t have access to that data. There are a lot of questions we could ask to pick it apart. Who are the people? How many other burger places in town have they eaten at? Furthermore, what counts as ‘in town’? Just Beverly Hills? The greater Los Angeles area? These are two very different areas. Hamburger Haven is lucky they crossed ‘Almost’ off their sign or I’d be picking that apart too. The more you think about it, the less helpful that quote is. If on 60% of all days, one person tells the workers they have the best burger in Beverly Hills out of the three restaurants they’ve been to, is the sign accurate? And this is even assuming the sign is telling the truth to begin with, but let’s not get into the possibility of lies, it makes things too complicated. All we can really learn from this quote is that some amount of people who eat at the restaurant like it enough to tell the owners or the workers that it’s the best burger in town. We don’t *really* even know if more people make these sorts of comments at Hamburger Haven than at other restaurants, though that’s definitely the implication.
Okay, so what's the point? Most of the time in the world of operations we have access to the underlying data, so it's a different situation, right? Well... that's a good question actually. DO WE have access to ALL the underlying data? Or even all of the IMPORTANT data? I heard a quote while working on this post, “The more you know the more you have to know in order to say anything about almost anything.” A more common formulation of this concept is what a lot of people refer to as the Dunning-Kruger effect, the more you know the more you realize you don’t know (This isn’t actually quite what Dunning & Kruger found but I’ve come to accept I’m not going to change the way society references the term). However, I think there’s a little bit more to be gained in this formulation of the concept. At each level of specificity, you unearth another level of specificity possible. At a certain point, we have to decide when something is accurate *enough*. A great way to evaluate this is to ask yourself what question is being asked/answered.
“Almost Everyday, people tell us, ‘You have the best burgers in town.’”
So what question is this answering? At first glance, it seems like it’s making a claim about how their burger ranks up against others in town, but if that’s the case it’s doing a pretty bad job. Remember when McDonald’s signs would have a count of how many burgers had been served worldwide? These were almost certainly inaccurate / out of date, but even though the sign appeared to be advertising a number, what they’re really advertising is a type of social proof. If they’ve served 1 million burgers, that’s probably a bunch of people who like these burgers. We get to that understanding regardless of our lack of knowledge over how many burgers other companies have sold. Similarly, Hamburger Haven isn’t actually making a statement about whether or not they’re the best burger in town, the statement is about how many people like their food. If people tell them every day that they have the best burger in town, that seems like a lot of people who really like that burger. It doesn’t actually matter their credentials, what part of Los Angeles they’re referring to by “in town”, or how many other burgers they’ve had. All that matters is that they liked the meal.
When we get questions about data we have access to, it’s always important to think through the limitations of our data. There might be gaps preventing us from accurately answering the question as answered, but the question might also be narrower than it needs to be. And as we discussed earlier, the more specific you get the more you have to know. Now sometimes you need to answer specific questions with a high level of data required, but most of the time (in my experience) there is a simpler concept at the core of a question than what you see on its face. Answering that question is often much easier and also more valuable to the asker. When we look at revenue numbers, we (usually) don’t care what the exact number is, we just want to know what category the company should be in. I wasn’t looking for the best burger in Beverly Hills, I was looking for a meal in Beverly Hills, and a lot of people seemed to like Hamburger Haven. You know what? I did too.