ROI for a small business is net profit from an investment divided by the cost of that investment, as a percentage. Spend £10,000 on equipment that generates £13,000 in additional profit and your ROI is 30%.
The formula is the easy part. What makes the exercise worth doing is being honest about both halves of it: counting every cost rather than just the purchase price, and counting only the profit caused by the investment rather than everything that happened afterwards. Done loosely, an ROI calculation will justify almost anything you already wanted to buy.
The Bottom Line
- ROI = (net profit from the investment ÷ cost of the investment) × 100.
- Use profit, not revenue. A £20,000 revenue increase on a 30% margin is £6,000 of profit, and the difference decides most borderline cases.
- Annualise before comparing. A 40% return over three years is worse than 20% in one.
- Common thresholds: investors look for 5% or more, while owners generally want 15% to 30% on operational spending.
- Count the total cost: installation, training, insurance, maintenance and your own time.
Save 80% of delivery management time
We handle everything:
- Dedicated operations manager
- Real-time tracking dashboard
- Automated customer notifications
- Urgent issue resolution
The ROI formula
ROI = (net profit from the investment ÷ cost of the investment) × 100
Some write the numerator as gain minus cost, which gives the same answer:
ROI = ((total return − total cost) ÷ total cost) × 100
Both halves need care.
Net profit means profit, not revenue. This is the single most common error and it inflates results by whatever your margin is not. A new van that brings in £40,000 of additional sales has not returned £40,000. On a 25% profit margin, it returned £10,000, and if the van cost £28,000 that is a −64% first-year ROI rather than the 43% the revenue figure suggests.
Cost means total cost. What you want is the incremental cost of the decision, and how to calculate incremental cost, with real examples works through the method. Purchase price, delivery, installation, training, insurance, servicing, finance interest and licences. For a hire, the cost is salary plus employer contributions plus equipment plus the hours you spend recruiting and training. Understating cost is the second way these calculations become fiction.
Worked example one: a delivery vehicle
An owner is considering a second van at £26,000, plus £3,200 a year in insurance, tax and servicing, and £4,800 a year in fuel.
The van allows roughly 320 additional deliveries a year. Average order value is £85 and gross margin is 35%, so each delivery contributes about £29.75 in gross profit. That is £9,520 a year in additional gross profit, and it is where the driver cost has to come in: a part-time driver at £11,000 a year leaves the van contributing −£9,480 in year one once the purchase is counted.
Over three years, with the purchase amortised:
| Line | Three-year figure |
|---|---|
| Additional gross profit (320 × £29.75 × 3) | £28,560 |
| Van purchase | −£26,000 |
| Running costs (£8,000 × 3) | −£24,000 |
| Driver | −£33,000 |
| Net | −£54,440 |
Negative, decisively. The calculation says this van only works at much higher utilisation, a higher average order, or both.
That is the useful outcome. It redirects the decision from “can we afford a van” to “what would have to be true for the van to pay”, and the answer here is roughly double the delivery volume or a materially larger order size. Raising average order value is often the cheaper half of that equation, because the delivery cost per stop barely changes when the order on the van gets bigger.
Worked example two: a marketing campaign
£2,400 on local advertising over three months. It produces 48 enquiries, of which 12 become customers. Each spends an average of £420 in the first year at a 35% margin, giving £147 of gross profit per customer.
First-year return: 12 × £147 = £1,764 against £2,400 spent. ROI is (1,764 − 2,400) ÷ 2,400 = −26.5%.
On a first-purchase basis the campaign lost money. But most of these customers order again, and that is where the calculation has to extend. If the average customer stays two and a half years and orders three times a year, lifetime gross profit is closer to £1,100, and 12 customers return £13,200 against £2,400, an ROI of 450%.
Both numbers are correct and they answer different questions. The first-purchase figure tells you about cash timing: you are out of pocket for months before the campaign pays. The lifetime figure tells you whether to repeat it. A business with tight working capital can be entirely right to decline a campaign with excellent lifetime ROI, because the cash gap is unfinanceable, which is why this calculation belongs next to the cash flow metrics rather than in isolation.
Annualising ROI so comparisons work
A raw ROI figure with no time period attached cannot be compared with anything.
Annualised ROI = ((1 + ROI)^(1 ÷ years)) − 1
A 60% return over four years annualises to roughly 12.5% a year. A 25% return in a single year beats it comfortably, despite looking smaller. Whenever you are choosing between two investments with different horizons, annualise first.
For a quick sanity check, payback period often serves better in a small business: cost divided by annual profit contribution, giving the number of years before you are whole. Payback under two years is generally comfortable; beyond four, the forecast is doing more work than the evidence supports.
What counts as a good ROI for a small business
There is no universal threshold, but the published ranges cluster usefully.
| Context | Typical expectation |
|---|---|
| Minimum investors look for in a small business | 5% or more |
| Buyers assessing a small company | 15–30% |
| Owners on equipment and operational spend | 15–30% a year |
| Marketing, measured on lifetime value | Often 200%+ |
Marketing looks wildly better than equipment because it is usually measured over a customer lifetime rather than a single year, and because the cost base is smaller. The comparison is not like for like.
Two adjustments to any threshold you set. Subtract what the money would earn doing nothing, since a 4% return is worse than leaving the cash on deposit. And weight by certainty: a reliable 15% on a piece of equipment whose output you can predict is worth more than a speculative 40% on a campaign in a channel you have never tested.
The mistakes that inflate ROI calculations
Using revenue in place of profit. Covered above, and responsible for more bad purchases than any other error.
Ignoring your own time. Owner hours are a real cost even when unpaid. An investment that returns 20% while consuming ten hours a week of your attention may be worse than one returning 12% that runs itself.
Claiming credit for everything that followed. If sales rose 15% in the quarter you bought new equipment and also hired a salesperson and entered a busy season, the equipment did not cause 15%. Isolate what you can and be explicit about what you cannot.
Omitting the ongoing costs. Purchase price is the beginning. Insurance, maintenance, consumables, software renewals and training all belong in the denominator.
Forecasting from the best case. Run the numbers at the volume you are confident of, not the volume you hope for. If it only works in the optimistic case, that is the finding.
Ignoring opportunity cost. The same money has other uses, including paying down debt. Compare against the realistic alternative, not against zero.
Where ROI fits with your other numbers
ROI is a decision tool rather than a tracking metric. You run it before committing money and again afterwards to see whether the forecast held, and that second pass is what makes future forecasts better.
It depends on other metrics being right. The margin figure in the numerator comes from your profit analysis; the lifetime value figure comes from retention and order frequency. If those inputs are wrong, the ROI calculation inherits the error and presents it with false precision.
It also connects to growth. An investment that raises capacity only pays if demand materialises, so the forecast should be grounded in observed revenue growth rather than ambition. Where the demand is not there yet, the cheaper move is usually to generate it first, and five smart ways to grow a small business covers options needing less capital than new capacity does. And where an investment is meant to improve service rather than capacity, net promoter score is often the only measurable result in the first year, with the revenue effect arriving later.
ROI belongs in the financial group of the small business metrics worth tracking.
Frequently asked questions
How do you calculate ROI for a small business?
Divide the net profit generated by the investment by the total cost of that investment, then multiply by 100. Use profit rather than revenue, and include every cost associated with the investment rather than the purchase price alone.
What is a good ROI for a small business?
Most owners treat 15% to 30% a year as a solid return on operational investments, and investors generally look for at least 5% before putting money into a small business. Marketing is usually judged on a higher bar because it is measured over a customer lifetime.
What is the difference between ROI and payback period?
ROI expresses the return as a percentage of the cost. Payback period expresses it as the time until you recover the cost. Payback is often the more intuitive figure for a small business because it speaks directly to cash risk.
Should ROI use gross or net profit?
Use the profit measure that reflects the costs the investment actually affects. Gross profit is appropriate when the investment drives additional sales and you are counting the direct cost of those sales; net profit is appropriate when the investment also changes overhead.
How do you calculate ROI on a marketing campaign?
Divide the gross profit from customers the campaign acquired by what the campaign cost. Calculate it twice, once on first purchase and once on expected lifetime value, because the two answer different questions about cash and about whether to repeat the spend.
Before you commit
Write the calculation down before the decision, including the volume assumption it depends on. That assumption is where these forecasts fail, and recording it means you can check it against reality in six months rather than discovering the gap by feel.
One input that is easy to get wrong on a vehicle or capacity decision is what delivery actually costs where you operate, since labour and distance vary considerably between markets. Our breakdown for San Bernardino, California shows the kind of local detail worth checking before you build it into a forecast.
Then run the pessimistic case. If the investment still clears your threshold at 70% of the volume you expect, it is a sound decision. If it only works at the number you are hoping for, you have a hope rather than a return.
