How Do We Spend Money? Finding the Spending That Buys Nothing

The problem: Your biggest wasted spending is not a vendor line, it is salary going into work that only exists because information is scattered.

The solution: Find the repetitive work first, then organize the data underneath it so the work stops being necessary at all.

When a business decides to look at spending, it usually looks in the same places. Vendor contracts. Software subscriptions. Insurance. Fuel. The things that arrive as invoices, because invoices are visible and easy to compare.

That review is worth doing and it rarely finds much. A few thousand in duplicate subscriptions, a renegotiated rate, a service nobody uses anymore. Useful, and small.

The larger waste is somewhere else, and it never arrives as an invoice.

Payroll is your biggest line and your least examined one

In a business between $5M and $15M, payroll is almost always the largest cost. It is also the one nobody audits at the task level, because doing so feels like questioning people rather than questioning work.

But the question is not whether your people are working hard. They are. The question is whether the work itself needs to exist.

Some of it does not. Somewhere in your business, a person is copying numbers from one system into another because the two do not talk. Somebody is chasing a signature for the third time. Somebody rebuilds the same report from the same five sources every Monday morning. Somebody re-enters an order that arrived by email into the system that will produce the invoice.

None of that work is visible as spending. It is spread across salaries you have already committed to, and it looks exactly like normal operations, which is precisely why it never gets questioned.

Why this work exists at all

It is worth being clear about the cause, because it determines the fix.

This work exists because information is scattered. When the quote lives in one place and the job lives in another and the invoice lives in a third, somebody has to be the bridge. That person is the integration. They are doing by hand what a connection between two systems would do automatically, and they are doing it because at some point it was faster to have a person do it than to fix the underlying problem.

That trade was probably correct when the business was smaller. At $10M it is expensive, and it grows with you. Every new customer, job, and employee adds volume to work that scales linearly with headcount rather than being absorbed by a system.

A look at a $12M distributor

Take a distributor doing about $12M a year, with roughly forty people, running on accounting software, a warehouse system, and a great many spreadsheets.

A cost review of the usual kind would turn up something like $14k of annual savings across subscriptions and a renegotiated freight rate. Real money, and it would take a month of somebody's attention.

Now look at where the hours go, which nobody has done. You would expect to find four people in the office, on salaries averaging around $70k, each losing roughly a day a week to moving information between the order system, the warehouse system, and accounting. Orders that arrive by email get typed in twice. Stock counts get reconciled against two sources that disagree. The monthly customer statements are assembled by hand.

A day a week each is about 20 percent of four salaries. Call it $56k a year of payroll buying nothing at all. Nobody has ever seen that number, because it does not appear anywhere. It is not a line item. It is four people being slightly less productive than they could be, every week, forever.

The full picture would be worse than that. Two more people in the warehouse would be spending significant time on counts that are only necessary because the system cannot be trusted. Added up, a business like this would be carrying roughly $180k a year of payroll doing work that exists only because its information is disorganized. That is more than twelve times what the vendor review found, and it has never been examined.

None of this means laying anybody off. The office manager who had been doing statements by hand would move onto collections, where you would expect her to recover more in the first quarter than her share of the automation cost. That is usually how this goes: the value is not in removing people, it is in getting people back onto work that produces something.

How to find yours

1. Follow the information, not the org chart. Ask where a customer order goes from arrival to payment, and count how many times a human retypes something that already exists somewhere else.
2. Count in days per week, then convert to salary. "Two days a week" is abstract. Forty percent of a $65k salary is $26k, and that lands differently.
3. Look for work that only exists because two systems disagree. Reconciliation, double entry, and verification counts are the clearest signals.
4. Fix the information before automating the task. Automating a broken handoff makes a faster broken handoff. Connect the underlying data first and some of the work disappears rather than being sped up.
5. Decide where the freed time goes before you free it. The savings only become real if that time moves onto work that produces revenue or reduces risk.

The takeaway

Knowing how you spend money means looking past the invoices. The spending that buys nothing is mostly payroll, it is invisible because it is spread across salaries you have already committed to, and it exists because the information your business runs on was never organized. Fix that and the work stops being necessary, which is a better outcome than doing it faster.