There is nothing magic about ten million dollars. It is a useful marker because of what tends to be true by the time a business reaches it: more people, more transactions, more locations or product lines, and usually the same operating setup that was designed when the business was a quarter of the size.
What changes is not the volume itself. It is that the informal mechanisms holding everything together stop scaling. Those mechanisms were never written down, because they were never designed. They accumulated.
The three things that quietly stop working
One person knowing everything. In a smaller business, someone always holds the full picture: what that customer is really owed, why that inventory count is off, which invoice is disputed. That person is a genuine asset until the business gets big enough that no single head can hold it all. The failure is gradual. Answers get slower, then approximate, then wrong, and nobody can point to when it changed.
Exception handling by memory. Every business has exceptions: the customer with special pricing, the vendor paid on different terms, the product that ships differently. At small scale these live in people's heads and it works. At larger scale the exceptions multiply faster than revenue, because each new customer, product and location brings its own. Eventually there are more exceptions than anyone can reliably recall, and the errors that result look random.
The spreadsheet layer. Almost every business at this stage has one: a set of spreadsheets sitting between the accounting system and reality, holding what the system cannot. Commission calculations, inventory adjustments, revenue schedules, the real customer list. It is invisible in any org chart and load-bearing in practice.
Why it shows up as a people problem first
This is the part that causes businesses to solve the wrong problem, sometimes for years.
When systems stop keeping up, the symptoms look exactly like understaffing. The close is late, so finance needs another person. Reports take too long, so hire an analyst. Orders have errors, so add a coordinator to check them. Each hire genuinely helps, because more hands do absorb more manual work.
But the manual work is growing with the business, so the relief is temporary. Businesses can go several years and several hires down this path before noticing that finance headcount has doubled while the close has not gotten meaningfully shorter. The tell is that ratio: if support functions are growing faster than revenue and output has not improved proportionally, you are buying capacity to feed a process rather than fixing the process.
What is actually different at this size
Four specific things tend to be true above roughly ten million that were not true below it, and each one breaks a different assumption in an entry-level setup.
More than one of something. More than one location, entity, warehouse, currency or legal structure. Every system built around a single instance of these requires manual work to bridge the gap, every period, forever.
People who need access but not full access. At five people, everyone seeing everything is fine. At thirty, it is a genuine risk and usually a compliance problem. Entry-level systems have coarse permissions, so the practical choice becomes giving people too much access or keeping them out and having someone else run their reports.
Approvals that need to be enforced rather than remembered. Purchase approvals by email work at low volume because everyone knows the rules. At higher volume they fail silently, and you find out during an audit or after a payment that should not have gone out.
Outside parties who want to see the numbers. Lenders, insurers, acquirers, auditors. Each brings a documentation standard that internal reporting never had to meet.
What this does not mean
Crossing a revenue threshold is not by itself a reason to replace anything. Plenty of businesses operate well past this point on modest systems because their operations are genuinely simple: one location, few SKUs, straightforward revenue, a small team.
The question is not your revenue. It is whether the manual layer holding your operation together is growing faster than the business, and whether the people maintaining it could be replaced if they left. Revenue is just the point at which that question usually becomes impossible to keep deferring.
What growth looks like when systems are the constraint
These show up in businesses whose operations have outgrown their setup, regardless of the exact revenue figure.
Finance headcount growing faster than revenue
You have added people to accounting and the close has not gotten shorter. That is the clearest available signal that the extra capacity is being consumed by manual process rather than producing more output. Two more people producing the same close in the same time means the process, not the team, is the limit.
Nobody can answer a question without a spreadsheet
Simple questions such as margin on a product line, or which customers are past due, require someone to build something. In a system holding the right data, those are lookups. When they are projects, the data exists but is not connected in a way anyone can query.
Key knowledge lives with specific individuals
There is a person whose absence would genuinely stop something. They know the pricing exceptions, the reconciliation, the way the report is built. This is normal at small scale and a real operating risk at larger scale, and it is a symptom of process living in heads rather than in systems.
Errors that feel random but are not
Wrong prices on invoices, shipments to old addresses, duplicate payments. Each looks like carelessness. The pattern underneath is usually that the correct information exists somewhere the person doing the work could not easily see, or exists in two places that disagree.
Questions we hear most often
Is $10M actually a meaningful threshold?
Not on its own. It is a marker for what tends to be true by then: multiple locations or product lines, a team large enough that not everyone can see everything, and outside parties starting to ask for documentation. A simple single-location business can run well past it on modest systems, and a complex one can hit the same wall at four million. Complexity matters more than revenue.
How do we tell a systems problem from a staffing problem?
Look at what happened after your last hire in the affected function. If the close got shorter and stayed shorter, it was capacity. If it improved briefly and drifted back, the manual work is growing with the business and you bought time rather than a fix. Tracking whether support headcount is growing faster than revenue answers this over a few quarters.
We are growing fast. Should we wait until things settle down?
Growth is what makes the manual layer expensive, so waiting for it to settle usually means waiting for the problem to get worse. The more practical consideration is that migrating during rapid growth is harder than migrating during a steady period, so if you can see a calmer quarter ahead, that is the window, rather than waiting indefinitely for one.
What is the cost of waiting another year or two?
Mostly hidden: staff hours on work a system should do, decisions made on stale numbers, and errors that reach customers. There is also a compounding data cost, because every year of workarounds adds inconsistent records, so a business that eventually migrates is moving messier data than it would have moved earlier. Waiting does not avoid the cost of change, it moves it later and adds to it.
Does this mean we need a full ERP?
Not necessarily. The right answer depends on where the complexity actually sits. A business whose pressure is entirely financial reporting and consolidation needs something different from one whose pressure is inventory and production. Diagnosing which of those you are is the useful first step, and it is a cheaper exercise than evaluating software.
Related problems
- Outgrown QuickBooks: what breaks and what comes nextThe full guide to where QuickBooks stops fitting and what the options are.
- Running your business on spreadsheetsWhere spreadsheets break as a business grows and what it costs when they do.
- Departments disagree on the numbersWhy each team ends up with its own version of the truth.
- 7 signs your systems are quietly holding you backThe small daily frictions everyone has learned to live with.
Get a clear picture of where your gaps actually are
The free ERP Readiness Scorecard covers reporting, inventory, purchasing, multi-entity complexity, and more. Get a score based on your specific situation in under 5 minutes.
No sales pitch. No vendor push. Just a clear picture of where your systems stand.