Why Some Restaurants With Huge Menus Run Smoothly, While Small Menu Places Keep Running Out of Everything
You would think the math here is simple. Fewer menu items should mean fewer ingredients to track and fewer things that can go wrong. And yet anyone who has eaten at a small neighborhood spot has probably heard "sorry, we are out of that" more than once, while the place down the street with a menu the size of a small novel somehow never misses a beat. That contradiction is basically the whole story, and it turns out the answer has less to do with menu size and a lot more to do with what is actually happening behind the kitchen door.
A big menu is not really as big as it looks Here is the thing restaurant operations researchers keep pointing out. Menu length and operational complexity are not the same measurement, even though people tend to treat them like they are. Roy, Spiliotopoulou, and de Vries (2022) describe a case where a restaurant kitchen appeared to be juggling 260 unique dishes, a number that sounds like an operational nightmare, until researchers grouped the dishes by shared prep steps, shared ingredients, and shared resource usage. Once grouped, those 260 dishes collapsed down to around 30 real families of items (Roy et al., 2022). Same menu on paper, completely different level of actual complexity underneath it. That is the trick a lot of large menu restaurants are quietly running. A hundred item menu built around fifteen core proteins, a handful of sauces, and a few starches is not a hundred separate operational problems. It is more like a dozen. Every dish is a remix of the same underlying inventory, so ordering, prepping, and storage stay manageable even as the printed menu keeps growing. The kitchen is not really tracking a hundred things. It is tracking a dozen things and combining them a hundred different ways. Small menus do not automatically mean simple operations Meanwhile, a small menu feels like it should be the safer bet, right up until you look at what is actually driving stockouts, and it is rarely the number of dishes. Azharudin (2024) documents a real case at a small Indonesian restaurant with a genuinely limited menu, just three main dishes and four condiments, that still struggled constantly with inventory mismatches. The problem was not complexity, it was forecasting. The restaurant was essentially guessing at how much to order each week, without any system tying purchases to actual demand patterns (Azharudin, 2024). Some weeks they were sitting on nearly double the inventory they needed. Other weeks they would run dry with no cushion left at all. What is telling is where the money was actually getting lost. The study found that certain low cost condiments were seeing potential savings of over 50 percent simply by ordering smarter, not less, just smarter, based on real demand data instead of habit (Azharudin, 2024). That is the opposite of what you would expect from a supposedly simple menu. Having fewer ingredients does not protect a restaurant from bad forecasting. It just means each mistake stands out more, since there is no buffer of dozens of other dishes to quietly absorb the miscalculation. The real dividing line is engineering versus guessing This is where menu engineering research becomes useful, because it reframes the whole question. Restaurants that treat their menu as something to be actively analyzed, tracking which dishes are profitable, which are popular, and which are quietly draining resources, tend to run leaner even at scale. Fang, Peng, and Pan (2013) apply this kind of analysis to entrées, using cost and efficiency data to figure out which menu items should stay and which should be cut, rather than relying on a gut feeling about what is probably working. That is a very different mindset than just shrinking the menu and hoping the math sorts itself out on its own. Put differently, a big menu that has been engineered around shared ingredients behaves like a small, tightly controlled operation with extra flexibility bolted on top. A small menu that has never had its demand patterns actually studied behaves like chaos wearing a smaller costume. Menu size was never really the variable that mattered. What matters is whether anyone is doing the unglamorous work of tracking what is shared, what is popular, and what demand actually looks like week to week. So what is really going on with the restaurant that is always out of something Put these two pieces together and the constantly out of stock small restaurant starts to make a lot more sense. It is usually not managing too much. It is managing too little, in the sense that nobody ever built a system to translate past sales into future orders. Meanwhile the sprawling menu place is not secretly simple either. It has just been engineered so that complexity on the customer facing side does not translate into complexity in the walk in cooler. One restaurant scaled its systems up along with its menu. The other never really built systems in the first place.