A church in Kansas asked me to help them build a cafe, and I told them not to build it. They had the room. Their building could have held the cafe they were describing, and I still recommended against it, because the town was small, their average attendance was about 200, and there was no version of a Sunday there where the thing paid for itself. They were skeptical before they ever called me. I am usually the optimist in that conversation, the one pointing out that a church can put in a cart instead of a cafe and be happier for it, and with them I agreed instead.
People ask me for a number. There is a minimum size for a coffee shop, and I am not going to print one, which is an odd thing for a consultant to say out loud. The reason is that the minimum moves with what the bar has to do, and I would rather tell you what moves it than hand you a figure you would then have to defend to a landlord.
Why doesn't a square footage number answer the question?
Two rooms with identical floor area are not the same shop. Our roastery in Pensacola is 1,000 square feet, and I could draw you four versions of those same 1,000 square feet where one of them works and three of them do not, because the door is in the wrong wall, or the only drain sits under the one thing that cannot move, or the room runs long and narrow while the work runs wide. Floor area is a single number. A room is a pile of constraints wearing a single number.
That is why I ask about the bar before I ask about the building. We published a piece on how a drink actually moves across the bar, and it turns the size question down by name: bar workflow is the path a single drink takes from order to handoff, and it is not a floor plan. Nothing certifies a floor plan. Your plumbing has an inspector and your exits have one, and the arrangement of your bar has none, which makes any layout rule a judgment about somebody else's room rather than a specification for yours.
An average is an honest thing to offer about a building nobody has walked, and a number is genuinely useful to somebody who has never stood behind a bar. My problem with it is narrower than it sounds. A figure travels better than the reasoning underneath it, so it arrives at your lease negotiation on its own, stripped of the menu, the staffing and the drain locations that produced it, and you sign for area when what you needed was a shape.
What decides how small a coffee shop can be?
The busiest fifteen minutes decide it. Not the average hour, and not the day's total — the fifteen minutes when the line is longest, every station is in use, and nobody behind the bar has a spare hand. A room that holds that window holds everything else you will ever ask of it. A room that fails it fails on a Saturday in front of people who will not come back.
So walk the drink before you walk the room. Somebody takes the order and takes payment. Somebody doses and grinds, locks in, pulls the shot. Milk gets steamed and poured. The drink gets called and handed across a counter to a person who is standing somewhere specific. Somewhere in there, a used portafilter gets knocked out, a pitcher gets rinsed, a cup gets pulled off a stack, and a bag of beans that ran out mid-rush gets opened.
Now run it twice at once, because in the busiest fifteen minutes it is running twice at once. The questions that actually size the room are how many bodies are back there, how often two of them need the same square of floor, how many times one of them turns their back on a waiting customer to reach something, and whether the person on espresso can do the whole drink without leaving their position. When those answers are good, the room can be startlingly small. When they are bad, more floor area does not fix it; it just gives everybody further to walk.
It is also the part I cannot answer for you from a distance, because it is downstream of your menu. A shop pouring drip and two espresso drinks needs a different bar from a shop pouring sixteen builds with three syrups and cold foam. Tell me what the bar has to do, and I can tell you what room it needs. Hand me the room first, and the best either of us can do is guess well.
What does the building have to supply?
Power, water in, water out, and air. Those four are the ones that turn a room you like into a room you cannot use, and they are the ones a floor area number cannot see. An espresso machine and a grinder want electricity the building may not have to give. A bar wants water arriving and water leaving, and drains are expensive to move and cheap to design around if you know where they are before you sign. Ventilation depends on what you are cooking, if you are cooking.
Our roastery in Pensacola is 1,000 square feet, and if you were trying to copy that room, the 1,000 is the least useful thing I could give you. What matters in there is what the roaster needs in power and in air, and where the green coffee sits while it waits. It is the same logic that lets the space work as a place where other people rent time on our roasters instead of buying their own: the constraint was never the floor, it was the machine and what the building had to feed it.
Your health department and your planning office will have opinions here that outrank mine, and they will have them about your specific address. That is not a dodge. It is one of the most common reasons a shop's real square footage ends up different from the one in the founder's head, and it is why licensing and permits sit alongside the coffee in the order we work through with a new shop rather than after it.
Here is where we are the wrong people to call. We consult on cafe and church coffee builds, and we are not architects and not a design-build firm. If the constraint is the drawing rather than the coffee — walls, exits, restrooms, a change of use — the person you need is an architect, and hiring one early is cheaper than hiring one after a contractor has already framed something.
How small do real coffee shops go?
Smaller than people tend to picture. Ashley Rodriguez, writing for Barista Magazine in January 2019 under the headline Opening a Coffeeshop in Less Than 200 Square Feet, reported on a shop running in roughly 160 square feet. Valorie Clark, writing for Fresh Cup on 31 May 2023 in The Magic of Small Coffee Shops, put the upper edge of a genuinely small shop at about 400 square feet and landed on rent as the thing that actually kills you.
Read those two together and the useful point is not the range. It is that Fresh Cup's answer to the size question is a rent question, which is the same move I am making from a different direction. The room is a monthly payment with a shape, and the shape decides whether you can serve people fast enough to make the payment.
Our own smallest format is the coffee cart we take out to events, and it sits on an eight by eight foot pad. I give you that because it is a thing you can picture, not because it is a target: Barista Magazine's 160 square feet is about two and a half of those pads set side by side, and a permanent shop has to hold storage, a restroom and a back door that a cart simply does not. What the cart proves is narrower and still worth something. A bar that serves a line of people can be very small when one person can reach everything without moving their feet much.
How much of the space is seating?
More of it than you expect, and it is the part of the room that is genuinely a choice. You can size a dining room by the seat: decide how many people you want sitting, multiply by the area each one takes, add room to walk between them. That arithmetic is real and it works. What it sizes is your dining room. It does not size your shop, and treating the two as one number is how a founder ends up with a beautiful room and a bar that cannot feed it.
A shop with no seats at all is still a coffee shop. The one Barista Magazine wrote about in 2019 is one, and so is every cart, kiosk and walk-up window. So the honest way to ask it is not how many seats fit, it is what the seats are for. Seats buy you time on site, a second drink, a laptop at 2pm and somebody's standing Tuesday habit. They cost you rent on floor that does not brew anything.
Floor area also charges you twice. You pay to build it out once and you pay rent on it every month after that, whether the chairs on it are full or not, which is the same point Fresh Cup lands on from the tenant's side. That is the real argument for deciding what the seats are for before deciding how many of them there are: seats you can justify are cheap, and seats you added because the room looked empty are a bill that arrives on the first of every month.
There is one more thing sizing that room and it rarely gets written down: how the place is meant to feel. If people are supposed to sit and stay, the room has to be able to hold somebody's slow afternoon without them feeling like they are in the way of the line. That feeling has a floor, and it is a higher floor than the same volume of drinks requires. If the plan is that people order and go, that floor drops away and you can build something very tight. Both are real businesses. They are not the same square footage, and the difference between them has nothing to do with how many drinks you sell.
Does having the room mean you should build?
No, and the church in Kansas is why I answer that one fast. They came to us with a space already in the building and a sense that coffee ought to be happening in it. They were right that the room could hold a cafe. They were also skeptical that the whole idea worked, and that is rarer than it sounds.
Usually I am the one arguing the optimistic side of that call. There are options churches have not considered, and the most common one is that a cart does most of what they actually want at a fraction of what a cafe asks of them. A cart does not need its own room, and it does not need a build. Plenty of the churches I talk to want coffee to be a real part of the culture on a Sunday and do not have the space for a cafe.
Kansas was different. Their town was small enough that the doors could not realistically be open on anything but Sunday, and at about 200 average attendance a single morning a week was not going to carry a build and a team. None of that is a space problem. The room they had was the one part of the project that was already solved and already paid for, which is exactly why it was the least informative thing about the decision. We walk through the rest of that math in our piece on what it costs a church to start a coffee shop, and the shape of it is simple enough to state here: a room you already own lowers your build cost and does nothing at all for your revenue.
So I told them not to build. They had the room, and the hardest and most useful thing I did for that church was tell them the room was never the question.
Questions we get about coffee shop space
How many square feet does a small coffee shop need?
There is no single honest figure, because the binding constraint is what your bar has to do at its busiest rather than the floor area around it. Published examples do exist: Ashley Rodriguez wrote for Barista Magazine in January 2019 about a coffeeshop running in roughly 160 square feet. Treat a figure like that as evidence that small is possible, not as a target for your building.
Can you open a coffee shop in 200 square feet?
Yes, if the workflow fits. Barista Magazine's January 2019 piece was headlined Opening a Coffeeshop in Less Than 200 Square Feet, and the shop in it ran in roughly 160. What makes that work is a short menu, a bar that one or two people can run without crossing each other, and a building that already supplies the power, water and ventilation the equipment needs.
How much room do you need behind a coffee bar?
Enough that the drink never stops moving and two people never need the same square of floor at the same moment. No aisle width is worth publishing, because the honest answer depends on how many people work the bar, what the menu asks of them, and where the doors, drains and machines already sit in the room you are looking at.
Does a coffee shop need a seating area?
No. Seating is a separate decision from the shop itself, and per-seat arithmetic sizes a dining room rather than a business. Carts, kiosks and walk-up windows sell coffee with no seats at all. What seating buys is time on site and a reason to come back, and what it costs is rent on floor that does not brew anything.
If you want a second opinion on the room: we consult on cafe and church coffee builds, and you can see how that works on our consulting page, which publishes the list of questions we hand you to take to your health department, planning office, insurance carrier and attorney.
