Business operations are everything your company does between taking money from a customer and handing that customer what they paid for. For a bakery, a florist, a caterer or a small wholesaler, that stretch of ground is wider than it looks, because it ends at somebody’s door rather than at your counter. You are not just making the product. You are buying for it, staffing it, scheduling it, routing it, invoicing it, and answering the phone when it shows up late.
Most owners don’t have an operations problem in the abstract. They have one function quietly leaking into the other five, and it shows up as a bad week rather than as a bad number. This guide maps the functions, shows you how to find the one causing the mess, and points you to the two areas deep enough to need their own treatment: the day-to-day management of the work itself and the money side of a delivery operation.
The Bottom Line
- Business operations break down into six functions for a business that delivers: sourcing, production, order handling, delivery, money, and people. Problems almost always start in one and surface in another.
- Delivery is the most expensive of the six and the easiest to underprice. Capgemini Research Institute puts last-mile costs at 41% of total supply chain cost, and found that running a suboptimal delivery model can cut retail profitability by up to 26%.
- Rising costs are the top financial pressure on small firms. In the Federal Reserve’s 2026 Report on Employer Firms, 75% of firms named rising costs of goods, services or wages as a challenge, 56% named paying operating expenses, and 51% named uneven cash flow.
- Fix the constraint, not the symptom. Track where orders wait rather than where people are busy, and you will usually find one function creating work for the other five.
Save 80% of delivery management time
We handle everything:
- Dedicated operations manager
- Real-time tracking dashboard
- Automated customer notifications
- Urgent issue resolution
The six functions that make up business operations
Every order you fulfill passes through the same six stations, whether or not you have anyone assigned to them. Writing them down is the fastest way to see which one you have never managed.
| Function | What it covers | The usual failure |
|---|---|---|
| Sourcing | Buying ingredients, stems, packaging, fuel; supplier terms and lead times | A single supplier with no backup, and terms nobody has renegotiated in three years |
| Production | Making the thing, to a standard, in the window it has to be made in | Batch sizes set by habit rather than by the delivery schedule |
| Order handling | Taking the order, confirming it, getting it onto the right day and the right vehicle | Orders living in three places at once: phone, email, and someone’s memory |
| Delivery | Routing, loading, drivers, proof of delivery, exceptions | Priced as a flat fee that stopped covering the real cost years ago |
| Money | Invoicing, collections, cost per order, margins | Revenue looks fine, cash does not |
| People | Hiring, training, scheduling, who covers what when someone calls out | Everything routes through the owner |
Each function hands off to the next, and the handoffs are where the day gets lost. A late supplier delivery pushes production, which pushes loading, which pushes the route, which turns a 10 a.m. drop into a 1 p.m. apology. None of those are delivery problems, though the customer will describe it as one.
Why delivery changes the shape of your operations
A shop that delivers is running on a clock that a shop with a counter is not. The customer picked a window. That window sets your production deadline, which sets your prep start, which sets what time you needed the supplier’s truck to show up. Delivery does not sit at the end of your operations. It reaches backward and sets the schedule for everything else.
It is also the most expensive part. The Capgemini Research Institute’s last-mile study, based on a survey of 500 supply chain executives across nine countries, put last-mile delivery at 41% of total supply chain costs and found that retailers running suboptimal delivery models risk up to 26% of their profitability (Capgemini Research Institute). Those are enterprise numbers, but the shape holds at any size: the last few miles cost more per unit than anything upstream of them, and they are the part most often priced from a gut feeling.
The practical consequence is that you cannot plan production and plan delivery separately. Scheduling a bake or a build without the route in front of you produces work that is finished at the wrong time, which is the most expensive kind of finished. That interaction is where operations management for a shop that runs its own deliveries does its work: the scheduling, the batch sizes, the inventory buffers, the quality checks that have to happen before something goes on a truck rather than after.
How to find the function that is actually breaking your week
Busyness is a terrible diagnostic. Everybody in a small operation is busy, including in the parts that are working fine. What you want is the place where orders wait.
Pick ten recent orders and write down four timestamps for each: when the order came in, when production started, when it was ready, and when it went out the door. You don’t need software. A notebook for two weeks will do it.
Now look at the gaps rather than the work:
- A long gap between order received and production started means order handling is the constraint. Orders are sitting somewhere before anyone can act on them.
- A long gap between production started and ready means production is the constraint, and usually means capacity or materials, not effort.
- A long gap between ready and out the door means the delivery side is the constraint. Product is finished and waiting on a vehicle, a route, or a driver.
- Short gaps everywhere, but still chaos means the problem is people and coverage. The work flows until one person is out, and then nothing does.
The point of the exercise is that only one of these can be your biggest constraint at a time. Fixing anything else first will feel productive and change nothing, because the queue just re-forms at the real bottleneck. Owners routinely buy equipment to speed up a step that was never the slow one.
Two more things to measure once you have the timestamps, both of which reliably surprise people: the share of orders that needed at least one phone call to resolve, and the share that went out on a different day than originally promised. Those two ratios describe your operation more honestly than revenue does.
Fixing operational bottlenecks without spending money
Once you know which function is the constraint, the first round of fixes is almost always procedural rather than capital.
If order handling is the constraint, consolidate intake into one place. The expensive part of orders arriving by text, phone, email and walk-in is the reconciling, plus the one that gets missed. One intake list that everyone writes to, with a required delivery date on every line, removes an entire category of failure.
If production is the constraint, stop sizing batches by tradition and start sizing them by the delivery schedule. Grouping production by route rather than by product type is often the single highest-yield change a delivery-heavy shop can make, because it ends the practice of finishing Tuesday’s east-side order before Monday’s west-side one.
If delivery is the constraint, sequence before you buy. Most routes built by hand are built in the order the orders came in, which is not a route, it is a list. Grouping drops by geography and by time window, and loading the vehicle in reverse drop order, costs nothing and shortens the day.
If people are the constraint, write down the two or three procedures that only live in one person’s head. Not a manual, a single page each. The test of whether you have written enough is whether someone else could run that station on a Saturday without calling you.
If money is the constraint, that has its own mechanics, and they are not intuitive. A shop can be profitable per order and still run out of cash, particularly when it invoices business customers. Cost per order and order-to-cash timing are the numbers that settle it.
The money function most owners under-manage
Money deserves its own note here, because it is the function most likely to be doing damage while every other station looks healthy.
The Federal Reserve’s 2026 Report on Employer Firms, drawing on the 2025 Small Business Credit Survey, found that 75% of firms cited rising costs of goods, services or wages as a financial challenge, 56% cited paying operating expenses, and 51% cited uneven cash flows (Fed Small Business). Uneven cash flow is an operations symptom as much as a finance one. It is what invoicing terms, delivery costs and order timing add up to.
The buffer most small businesses run on is thinner than the planning horizon they use. Analyzing more than 470 million transactions across 597,000 small businesses, the JPMorgan Chase Institute found the median small business holds 27 cash buffer days (the number of days it could cover outflows if inflows stopped), and that small restaurants hold the fewest, at 16 days (JPMorgan Chase Institute). A supplier price increase or a slow-paying wholesale account eats into that quickly.
The operational reading of those numbers: your delivery pricing, your payment terms and your production schedule are the three levers that move cash, and only one of them is a finance decision.
What a business operations plan looks like at this size
A business operations plan for a shop with a delivery van does not need to look like a consulting deliverable. It needs to answer six questions in writing, one per function, and be short enough that someone actually reads it.
- Sourcing: who supplies each critical input, what the lead time is, and who the backup is.
- Production: what gets made when, sized against the delivery schedule rather than against the week.
- Order handling: where an order lives from the moment it arrives, and what information is required before it counts as booked.
- Delivery: how routes get built, what the cutoff time is, and what happens when a drop fails.
- Money: what an order costs to fulfill including delivery, what the terms are, and who chases invoices.
- People: who covers each station, and which procedures are written down.
Review it when something breaks rather than on a calendar. A plan that gets revised the week after a bad Saturday is worth more than one revised every quarter out of habit.
Where to start this week
Take the timestamps. Ten orders, four times each, two weeks. It is the cheapest diagnostic available and it settles arguments that otherwise run on opinion, including the argument about whether you need to hire.
Then fix one constraint, and only one. Re-measure after two weeks, because the constraint moves once you relieve it. That movement is what success looks like, not a permanent fix. The operation that improves steadily is the one whose owner always knows which of the six functions is currently the worst, and that answer is supposed to keep changing.
