A business has enough money and management time for one or two significant investments this year. Five proposals are on the table: new equipment, an online ordering system, a second location, a solar installation and a replacement inventory system. Each has a sensible case. Each pays back. Each has a champion who believes in it. If the owner judges them one at a time, against the question “is this a good idea?”, all five pass.
That is the problem. Businesses rarely choose between a good project and doing nothing. They choose between competing claims on the same money, the same few capable people and the same owner’s attention. A proposal can be profitable, well aligned and technically sound and still be the wrong choice, because something else would do more with the same resources. The question that matters is not “does this pay back?” but “what are we not doing because we fund this?”
This article explains how to bring opportunity cost into investment decisions, how to compare proposals that are not naturally comparable, why the shape of the whole set matters as much as each project, and how to tell whether the business is genuinely choosing or simply approving. It is general information for owners and managers who decide where money and effort go.
Three conclusions that should not be collapsed
A proposal needs to clear three separate hurdles, and they are often merged into one:
- Worthwhile: its benefits exceed its relevant costs.
- Affordable: the business can fund it, including the cash flow along the way.
- Prioritised: it deserves scarce resources ahead of the other things those resources could do.
A business case usually proves the first, sometimes the second, and rarely the third. A proposal should not move from “worthwhile” to “approved” without someone explicitly comparing it with the alternatives. The what the NPV spreadsheet hides article covers testing the numbers inside a single case.
Five kinds of opportunity cost
Opportunity cost is the value of the best alternative use of a resource. Money is only one of the resources involved:
- Capital: the funds that could go to another investment, debt reduction or a cash buffer.
- Capability: the skilled people who could be working on something else. Wages already being paid do not make their time free.
- Time: a project started now may delay another that would have delivered value sooner.
- Attention: the owner’s and managers’ time and focus, the scarcest resource in most small businesses.
- Resilience: committing reserves to one project reduces the business’s ability to absorb a shock.
Two cautions. Resources the business already owns, such as a spare shed, an existing team or a machine, still have an opportunity cost if they could be used elsewhere. And money already spent is a sunk cost: it should not influence the decision about what to do next, even though it often does.
Compare with real alternatives
A proposal compared only with “do nothing” will usually look good, especially when the do-nothing option is described as a frozen version of today rather than what will realistically happen. For any significant proposal, ask for at least three alternatives:
- Do minimum: what if we meet only the essential need?
- A different solution: is there another way to achieve the same outcome?
- Different timing or scale: could we stage it, start smaller or do it later?
Record why rejected alternatives were rejected. That makes later reviews more honest and shows whether the preferred option was ever seriously challenged.
Compare like with like
Proposals often arrive in incompatible terms. A compliance upgrade has no return but cannot be avoided. A resilience project prevents losses whose likelihood is uncertain. A growth project promises revenue nobody can size precisely. Forcing them all into a single ranked list usually produces a false order, and often ends up favouring whatever is cheapest.
A workable alternative is to sort proposals into a few classes and compare within each:
| Class | How to compare | Example |
|---|---|---|
| Must do | Size it sensibly; do not rank it | Safety, legal or licence-to-operate work |
| Growth | Expected value and how sure we are | New products, markets or channels |
| Efficiency and resilience | Cost or risk reduced per dollar | Equipment, systems, backup suppliers |
| Capability | What it enables later | Skills, systems other projects need |
The owner then decides how much of the available money and people each class receives, ideally before the proposals arrive. That is a genuine strategic choice, made once a year. Decided afterwards, the split tends to be argued backwards from whichever proposals are most persuasive.
The shape of the whole set
Individual rigour does not add up to a sound set of commitments. A set of proposals can each pass cleanly and together be undeliverable or dangerously concentrated. Look at four features of the whole:
- Balance across horizons. How much sustains the current business, how much improves it and how much builds something new? Under margin pressure, businesses drift towards the first two, which makes sense each quarter and weakens them over a decade.
- Concentration. Do the biggest expected benefits depend on the same supplier, the same customer group, the same technology or the same person? If the largest shared dependency failed, what share of the expected benefit would go with it? If it is more than about a third, the set has a concentration problem regardless of how well each project is managed.
- Load on scarce people. Against the few people every project needs, is the set over-committed, and when? The how much change a business can carry article covers this in detail.
- Timing of value. If every project returns value in year three, the business has no early evidence that its assumptions hold. Some short-cycle work is valuable because it produces information.
Highest return is not always best
A proposal with the highest expected return can still be the wrong choice:
- Downside matters, not just the average. If a plausible bad outcome would threaten cash, safety, service or reputation beyond what the business can absorb, the expected return is not the deciding figure. Willingness to take risk is not the same as capacity to survive it.
- Correlated risks add up. Ten projects can be one bet if they all rely on the same assumption.
- Existing performance is at risk too. A new project can disrupt the current business by pulling key people away, changing systems during a busy season or confusing customers. Count that cost.
- Staging reduces exposure. Smaller first steps, trials and option-like commitments let the business learn before committing fully. Separate learning money from scaling money.
Choosing, not just approving
A simple test shows whether a business is genuinely choosing: what did we decline in the last twelve months, and why? If the list is empty, or contains only proposals that were going to fail anyway, the business is validating rather than selecting.
Two conditions make real choice possible. There must be more credible proposals than the business can fund, which can feel wasteful but is the price of choice. And proposals must be comparable, which is what the classes above provide. Choosing well at the start is not the end, either; projects that stop deserving their resources should be stopped. The stopping projects well article covers that.
Revisit the choice during the year
A set of choices made once a year can be out of date by the middle of it. A large customer leaves, a supplier fails, a cost rises or a trial shows a stronger result than expected. Set a short quarterly review of the whole set, not just the progress of each project, and agree in advance which events trigger an earlier look: a key assumption failing, a sharp change in cash, or a scarce person becoming unavailable. Reprioritising without moving money or people changes nothing, so each review should end with a decision about where resources go next.
A worked example
This is an illustration. A family-owned hardware and building supplies business has about $400,000 available for investment this year, and its operations manager is the key person for most changes. It receives five proposals:
- Racking repairs and a forklift upgrade, about $60,000, after an inspection found items needing repair.
- An online ordering and click-and-collect system, about $120,000.
- A second trade counter in a nearby town, about $350,000 including a five-year lease and fit-out.
- Solar panels and a battery, about $150,000, with an estimated payback of six years.
- A replacement inventory system, about $90,000.
Every proposal has a positive case on its own. Together they cost $770,000.
The owner sorts them into classes. The racking repairs are a must do: sized, not ranked. Online ordering and the second trade counter are growth. Solar and the inventory system are efficiency and resilience. Before comparing, the owner decides that after the must-do work, at least a quarter of the remaining money should go to efficiency and resilience, and that a meaningful reserve must be kept for the spring peak, when stock levels and cash needs rise.
The owner then looks at the shape. The online system cannot work well without accurate stock data, so it depends on the inventory system. Both growth proposals depend on the same trade customers and the same operations manager. The second trade counter has the highest expected return but also the largest downside: a five-year lease the business could not easily exit if the town’s demand proved weaker than hoped.
The decision:
- Approved: racking and forklift ($60,000), the inventory system ($90,000) and the online ordering system ($120,000), sequenced so inventory comes first.
- Learning money: $15,000 for a three-month trial trade desk in the nearby town, operating two days a week from a builder’s yard, to test demand before signing a lease.
- Deferred with a date: solar, to be reconsidered in twelve months when cash allows.
- Reserve: the remaining $115,000, held for the spring peak and any surprises.
In total, $285,000 is committed and $115,000 is held. The inventory system, at $90,000, meets the owner’s rule that at least a quarter of the $340,000 left after the must-do work, about $85,000, goes to efficiency and resilience. The owner records what was declined or deferred and why, so the decision can be reviewed honestly next year.
How this applies to a small Australian business
- Ask what each proposal displaces, not just whether it pays back.
- Count people and attention as costs, not just money.
- Require real alternatives, including doing less or later.
- Sort proposals into classes and compare within them.
- Decide the split between classes before proposals arrive.
- Check concentration on shared customers, suppliers and people.
- Keep a reserve and separate learning money from scaling money.
- Record what you declined, and why.
Signals worth watching
- Every proposal approved because each pays back.
- The same few people named in every project.
- Business cases compared only with doing nothing.
- No proposals declined in the past year.
- Growth plans that all depend on one customer group.
- New projects disrupting the busy season.
Common mistakes
- Treating a positive return as approval.
- Ignoring the cost of internal people’s time.
- Ranking incomparable proposals on one list.
- Letting the loudest champion decide the mix.
- Choosing the highest expected return without checking the downside.
- Letting sunk costs drive the next decision.
Frequently asked questions
Is this too formal for a small business? The thinking takes an afternoon a year. Writing down the alternatives and what was declined takes less.
How big should the reserve be? Enough to cover the business’s seasonal cash swings and a plausible surprise. An accountant can help estimate it from past cash flow.
What if a must-do item uses most of the budget? Then that is the year’s reality. Make sure it is sized sensibly, and look for staged or cheaper ways to meet the requirement.
How do we value capability projects with no direct return? By what they enable: which future projects become possible or cheaper because of them.
Should we always diversify? Not necessarily. A focused strategy naturally concentrates risk. The point is to know where concentration sits and to manage it, not to avoid it entirely.
Questions to ask
- What are we not doing because we fund this?
- Which people and how much attention does it consume?
- What are the real alternatives, including doing less or later?
- How should our money and effort be split across must-do, growth, efficiency and capability?
- What shared dependency could undermine several projects at once?
- What did we decline last year, and was it the right call?
Bringing it together
Most proposals look good on their own; the hard part is choosing between them. Treat every investment as a claim on scarce money, people and attention, and ask what it displaces. Compare proposals with real alternatives, sort them into classes that can be compared fairly, and decide the overall split before the persuasive cases arrive. Look at the shape of the whole set: balance, concentration, load on key people and the timing of value. Keep a reserve, stage uncertain bets, and keep a record of what you said no to. A business that never declines a good project is not choosing.
Source: KEVOS notes, drawing on teaching material on capital budgeting, project selection, portfolio balance and risk-informed selection, the Australian Government’s 2006 Handbook of Cost-Benefit Analysis, and P. W. G. Morris and A. Jamieson (2005) on portfolio management practice. Examples and figures in this article are illustrations. This article is general information.
