Why do small capital projects matter if each one is minor?

Because the aggregate is often large and nobody owns it. Each item is approved proportionately and the total was never a decision. In multi-site organizations the sub-threshold pool can be a meaningful share of annual capital and is frequently the only part of it that has never been added up.

How do you stop projects being split to avoid approval thresholds?

Set the authority band by total project value rather than by individual invoice or phase, and review the cluster of requests sitting immediately below each threshold. A concentration just under a line is rarely coincidence.

Should small CapEx projects go through the full approval process?

Generally no. Routing routine capital upward costs more in cycle time than the decision is worth, which is why thresholds exist. The better answer is light control of a different kind: visibility of the total, an annual envelope per site, a minimal intake record and sampled review.

How do you calculate sub-threshold capital spend?

Split twelve months of capital spend at your lowest approval threshold, total what cleared locally, express it as a percentage of total capital, count the items as well as the value, break it down by site to find outliers, and examine the items sitting just under the line.

What is the difference between small CapEx and operating expenditure?

Small capital creates or extends the life of an asset, so it is capitalized and depreciates, and it draws on the capital envelope. Operating spend is consumed in the period. The size of the item does not change the treatment; the capitalization threshold in your accounting policy does.

Small CapEx Projects: Why Sub-Threshold Capital Is the Least Governed Money You Spend

Capital governance tends to concentrate where the numbers are largest. The Investment Committee sees the major projects. Finance models them, operations defends them, internal audit tests them afterwards. That's a reasonable allocation of attention, and it leaves a gap.

Below the lowest approval threshold sits a pool of capital that clears locally, gets recorded, and is examined by almost nobody. Individually each item is too small to be worth a committee's time, which is exactly why the threshold exists. Collectively the pool is often large enough that leadership would want to know about it, and in most organizations nobody has ever added it up.

This guide covers what counts as a small capital project, why the money below the line escapes scrutiny, how to size your own sub-threshold pool from records you already hold, and how to govern it without recreating the delay the threshold was designed to remove.

What counts as a small CapEx project?

Small is relative to your approval structure rather than to any absolute figure. A project is small if it clears at the lowest level of your delegation of authority without escalating.

That definition matters because it moves with the organization. Raise your bottom threshold to speed up routine decisions and you've just enlarged the pool of capital that receives the least scrutiny. The decision is usually made for good reasons and the consequence usually goes unexamined.

Typical contents of the pool look unremarkable one at a time. Equipment replacements at site level. Minor plant modifications. IT hardware refreshes. Small building works. Tooling, instrumentation, vehicles below the fleet threshold. Individually routine, and that's the point.

Two things separate this from operating spend and both matter. It's capital, so it lands on the balance sheet, depreciates and affects the asset base for years. And it competes for the same capital envelope as everything else, whether or not anyone is comparing them.

Four reasons small projects escape governance

The threshold is doing exactly what it should Routing routine capital to head office costs more in cycle time than the decision is worth. The threshold exists to prevent that, and it works. The unintended effect is that everything below it also leaves the reporting that follows the approval chain.

Nobody owns the aggregate Each site owns its own small projects. Nobody owns the sum of all of them. In organizations with a lot of sites, the number can be substantial and structurally invisible, because no single person's report contains it.

The business case bar is low or absent Above the threshold, a project arrives with costs, benefits, timing and risk. Below it, approval is often a conversation and a purchase order. That's proportionate for a pump replacement and it means there's rarely a stated benefit to review against afterwards.

Post-investment review has a threshold too Most organizations review projects above a set value. Small projects fall below it by definition, so nothing in the process ever asks whether the money achieved anything. The estimating and delivery habits inside that pool never get corrected, because nothing measures them.

How to calculate your sub-threshold aggregate

Most organizations have never produced this number. Five steps, and the data already exists.

Step 1. Split twelve months of capital spend at your lowest approval threshold Everything that cleared locally goes in one pile.

Step 2. Total it, and express it as a percentage of total capital spend The percentage is the number that gets attention, because it's comparable across sites and across years.

Step 3. Count the items, not just the value Volume tells you something different from value. A pool made of four hundred small items has a different governance problem from one made of forty medium ones.

Step 4. Break it down by site or business unit Look for the outlier. One site consistently spending twice the sub-threshold average of comparable sites is either managing an older asset base, which is useful to know, or splitting projects, which is more useful to know.

Step 5. Look at the items sitting just under the line A cluster of requests immediately below a threshold rarely happens by chance. It usually means the threshold is shaping the requests rather than the requests being assessed against the threshold.

Take those five outputs to whoever owns capital planning. In most organizations this is the first time anyone has seen the pool as a single number, and it tends to change the conversation about where governance effort should go.

Three risks that live below the line

Risk	How it shows up	What to do about it
Aggregation	Many small approved items add to a commitment nobody sized. Each decision was proportionate; the total was never a decision at all	Report the sub-threshold pool as a line in the capital pack, with its own annual envelope, and treat a breach of that envelope as a governance event
Splitting	A project is broken into pieces that each clear locally. Sometimes deliberate, more often the natural result of phasing work that could have been scoped as one project	Set the band by total project value rather than by individual invoice or phase, and review the cluster of items sitting just under each threshold
Deferred maintenance in disguise	Sustaining work that should be a planned capital programme arrives instead as a stream of small reactive replacements. Cheaper each time and more expensive in total	Track the ratio of reactive to planned within the pool. A rising ratio is an early signal that a larger asset decision is being avoided

The third is the one that costs most and shows up latest. A pump replaced four times in three years never appears as a capital problem, because no single replacement was one.

See the whole portfolio

Governing capital starts with being able to see all of it, including the part below the threshold. A short walkthrough shows the portfolio view at site, division and committee level.

Request a demo

How to govern small capital without slowing it down

The instinct once this pool becomes visible is to lower the threshold and route more of it upward. That reverses the improvement the threshold delivered, adds queue time to hundreds of decisions and teaches sites that the process is an obstacle to work around.

Control at the bottom of the structure should be light and different in kind, not the same control applied more widely.

Visibility rather than approval The centre needs to see the pool. It doesn't need to approve the items. A monthly total by site, with item counts, does more than an approval step and costs nobody a day of waiting.

An envelope rather than an item review Give each site an annual sub-threshold allocation. Within it they proceed. When they approach it, the conversation happens once rather than four hundred times.

A minimum record, not a business case Three fields are enough for most small capital: what it is, why now, and whether it's planned or reactive. That last field is what makes the reactive-to-planned ratio measurable, and it costs about ten seconds to capture.

Sampling instead of full review Post-investment review at this level should be a sample, not a census. Ten items a year, chosen at random, produce enough signal to keep estimating honest without creating work that exceeds the value of what's being reviewed.

One rule with teeth Total project value sets the band, not the invoice. Everything else here is light. This one isn't, because it's the control that stops splitting.

What good looks like

The pool has a number and an owner. Someone can state what percentage of capital cleared below the threshold last year, and it appears in the capital pack rather than being reconstructible on request.

Sites operate inside an envelope they know about. Local decisions stay local and stop being unbounded.

Planned and reactive are distinguishable. One field at intake, and the ratio becomes a leading indicator of asset condition rather than a surprise in three years.

Thresholds are reviewed annually against their own data. Item counts per band, and the cluster sitting just below each line.

Small projects appear in the portfolio view. Not individually, but as a block competing for the same capital as everything else, which is what they've been doing all along.

Frequently asked questions

What is a small CapEx project?: One that clears at the lowest level of your delegation of authority without escalating. The definition is relative to your approval structure rather than to a fixed dollar figure, which means raising a threshold enlarges the pool of capital receiving the least scrutiny.
Why do small capital projects matter if each one is minor?: Because the aggregate is often large and nobody owns it. Each item is approved proportionately and the total was never a decision. In multi-site organizations the sub-threshold pool can be a meaningful share of annual capital and is frequently the only part of it that has never been added up.
How do you stop projects being split to avoid approval thresholds?: Set the authority band by total project value rather than by individual invoice or phase, and review the cluster of requests sitting immediately below each threshold. A concentration just under a line is rarely coincidence.
Should small CapEx projects go through the full approval process?: Generally no. Routing routine capital upward costs more in cycle time than the decision is worth, which is why thresholds exist. The better answer is light control of a different kind: visibility of the total, an annual envelope per site, a minimal intake record and sampled review.
How do you calculate sub-threshold capital spend?: Split twelve months of capital spend at your lowest approval threshold, total what cleared locally, express it as a percentage of total capital, count the items as well as the value, break it down by site to find outliers, and examine the items sitting just under the line.
What is the difference between small CapEx and operating expenditure?: Small capital creates or extends the life of an asset, so it's capitalized and depreciates, and it draws on the capital envelope. Operating spend is consumed in the period. The size of the item doesn't change the treatment; the capitalization threshold in your accounting policy does.

Where to go from here

Sub-threshold capital is the one pool where the governance question isn't how to control it more tightly. It's how to see it at all, and then apply the lightest control that makes it visible without slowing down the local decisions the threshold was designed to protect.

If you can't currently state what percentage of your capital cleared below the lowest approval threshold last year, that calculation is a good afternoon's work. It's usually the first time anyone has looked, and the number is usually larger than expected.

See how CapEx360 gives visibility of capital at every level, from local decisions through to the Investment Committee.

Request a demo