When I start thinking seriously about a new coffee business, I usually open Excel fairly early.
Not because I already know the numbers. Usually I do not.
I use the spreadsheet because I need to start seeing the shape of the business. If I only think about the idea in my head, it is easy for everything to sound possible at the same time.
But I also do not begin by saying, “This is my budget, so what can I buy with it?”
First I need some idea of what I am actually trying to sell.
For me, that comes before most of the cost model. The product affects how I make it. How I make it affects the equipment. The equipment affects the working space, utilities and workflow. If there is a customer-facing space, that adds another part to the footprint.
Once those pieces start appearing, I can begin to estimate what kind of investment the idea is asking for.
The first model is there to give me a feel for the business
When I planned larger coffee businesses, I did not build one start-up-cost sheet and stop there.
I would rough out the product, equipment and space, then look at what a business of that size might have to sell. From there I could start thinking about staffing, ingredient cost, utilities and the other operating expenses. I wanted to see whether the rough economics made any sense at all.
Then I would change things.
Maybe the space was too large. Maybe the equipment plan was heavier than the product required. Maybe the sales needed to support the investment looked too optimistic. Then the model had to come back down and I had to look again at the assumptions that created it.
That is still how I think about it.
The first spreadsheet is not a prediction. I am trying to get a feel for the relationship between the pieces.
Only after I can see that rough picture do I want to put a real cap around the investment.
I think the cap is important because otherwise cost has a way of moving in one direction.
There is always a better machine. There is always a nicer finish, another feature or something that would be useful to have. If I keep choosing the better option one decision at a time without coming back to the whole number, eventually I can end up with a business that costs far more than I intended to build.
So my sequence is a little different from simply starting with a fixed budget. I first want to understand what the concept seems to need. Then I look at what I can actually support and start tightening the plan inside that boundary.
A piece of equipment has to arrive and actually work
The large equipment prices are the easiest numbers to notice.
They are not always the numbers that worry me most.
A roaster or another machine has a purchase price, but then it has to get to the site and become usable. There may be freight, installation, commissioning or accessories. The space may need enough electrical capacity, gas, water, drainage or ventilation for the equipment to work properly.
Some of those costs belong directly to the machine. Others make more sense as facility costs. I do not care very much which spreadsheet line they sit on as long as they do not disappear.
The premises side can be especially difficult to judge too casually.
Electrical upgrades, ventilation, plumbing and drainage are not exciting parts of opening a coffee business, but they can become substantial. They can also depend heavily on the condition of the space you eventually choose, which means an early estimate may still have a lot of uncertainty in it.
Timing creates another problem.
If construction, equipment delivery or an approval takes longer than expected, the effect is not always just an extra invoice. The opening date may move too. That means the point when I expected revenue to start can move with it.
When I looked again at my old 2024 Filgap financial workbook, I noticed that I had already spread pre-opening expenses and capital purchases across different months. I had not put everything into one opening-day total.
The business in that workbook is no longer the business I am building now, but I still agree with that part of the method. A start-up cost has an amount, but often it has a timing problem attached to it too.
Some things are worth buying once
Filgap is much smaller now, and the staged approach changes the way I think about equipment.
I can separate what I need for the Farmers’ Market from what I may need later in a permanent roastery and tasting room.
That does not mean I want the cheapest version of everything for the first stage.
I am looking at this as something I hope to keep doing for twenty years or more. Of course there is no guarantee that a business will last that long. The point is that I am not making current decisions on the assumption that I will quit soon.
So if a piece of roasting, production or quality-control equipment is something I can realistically keep using as Filgap grows, I am willing to think differently about that purchase. I would rather choose it properly once than intentionally buy something inadequate and replace the same capability again later.
A Farmers’ Market tent is not the same kind of decision.
I need a tent and display equipment that work well enough for the market. But those things may have very little use inside a future permanent roastery. I do not need to turn a temporary-stage asset into a long-term investment just because I want the market setup to look complete.
That distinction is becoming useful for me now: what should travel with the business, what only belongs to the current stage, and what should not be purchased until the next stage is real.
I try to make the individual numbers as real as I can
There will always be uncertainty in an early cost model.
My answer is not to put a large buffer into every single line.
If I can get a supplier quote, I would rather get the quote. If I need to compare options, call somebody or research an installation cost, I would rather do that and improve the estimate.
One expression probably explains how I see it: points become a line, and lines become a surface.
Each estimate is one of those points.
If every point has an oversized safety margin built into it, eventually I may be looking at a total that is very conservative but not very informative. I have lost the sense of what the individual things actually appear to cost.
So I prefer to make the line items reasonably accurate and keep contingency visible separately for the uncertainty that remains.
The numbers can then improve over time. An online estimate may become a supplier quote. A quote may become an invoice. A general installation allowance may become a real contractor estimate once the site is known.
I expect the spreadsheet to change. That is part of its job.
My first version does not need a complicated template. For each line, I want enough information to understand what stage it belongs to, what the item is, the best base cost I have, what freight or setup still needs to be added, where the estimate came from, how confident I am in it, when I expect to pay it, and whether I need it now or can wait. If the site itself creates a cost—electrical, ventilation, plumbing—I want that visible too.
The point is not to create more columns. It is to be able to trace the number.
I do not want to spend the last dollar getting ready to open
There is one part of the required capital that I do not want to leave until the end of the calculation.
The business still needs money after the equipment has been purchased and the doors can open.
I will deal with cash flow more closely later because it deserves its own discussion. The important point here is simpler: I do not consider a start-up fully funded just because I can pay for the physical setup.
The business needs time to establish itself.
My own preference is to have enough strength to withstand roughly a year if the early period is slower than I hope. I am not presenting that as a rule for every coffee business. It is the level of caution I prefer for myself.
If I start with very little room and immediately need the sales forecast to come true, I know what can happen psychologically too. The owner becomes impatient. Every weak week starts to feel like an emergency, and that pressure can push decisions in directions that were never part of the plan.
I would rather account for that possibility before I spend the start-up money.
The spreadsheet can look precise while the assumption underneath it is weak
The thing I am most careful about is optimism.
It is surprisingly easy to make an optimistic spreadsheet look professional.
I can enter a sales number for a future month, connect the formulas, and suddenly the whole business appears to work. The labour percentage fits. The expenses fit. The investment starts paying itself back.
The formulas may all be correct. The weak part may simply be the number I typed into the sales cell.
That is why I do not want the start-up cost model to become more convincing than the information behind it. Later I need to test the sales assumptions, the cash timing and what happens when the plan is wrong. Those are separate questions, and I will get to them next.
For now, when I build the first cost model, I want to be able to trace where the number came from.
- What am I selling?
- What does it take to produce it?
- What equipment and space follow from that?
- What does it really cost to get those things operating?
- Which purchases are for this stage and which can wait?
- When will the money probably have to be paid?
- How certain is the estimate?
Then I can look at the total, put a limit around it, and keep changing the model as the business becomes less theoretical.
That is what I want from the spreadsheet.
Not an answer that tells me exactly what the future will cost. Just a picture good enough that I can make the next decision with my eyes open.
