By size

The band where the system genuinely stops being adequate

Between $25M and $100M, most companies cross three thresholds at once — entities multiply, the close stops fitting in the first week, and the volume of routine finance work starts requiring hires. This is the size where a real ERP decision becomes unavoidable and where it is still possible to make it without a nine-month project.

company size served →implementation effort →$1M$25M$100M$500M$5B+mid-market capability, low implementation burdenQuickBooks OnlineXeroQuickBooks EnterpriseOdooSage 100 / 300AcumaticaSage IntacctDynamics 365 BCNetSuiteSAP Business OneSAP S/4HANAerp.ioour read of the market · criteria and method at /software/methodology
The most common size we serveLive in weeks, not quartersHonest about the upmarket cases

The situation

What changes between $25M and $100M.

Entities stop being incidental

Two or three legal entities become normal — an acquisition, a subsidiary, a holding company — and consolidation moves from an annual nuisance to a monthly dependency.

The close stops fitting

What took five days at $15M takes eleven at $60M with the same team, because volume grew and the process did not change. It degrades a day a quarter without anyone noticing.

Volume forces hiring

Somewhere around 300 bills a month the manual work justifies another person, and that hire gets classified as a growth cost rather than a systems cost.

Someone starts asking for segment reporting

A board, a lender, or a new CFO wants department and location P&L, and nobody can produce it without a spreadsheet nobody trusts.

Revenue gets complicated

Multi-element contracts, ramps, and modifications arrive with scale, and ASC 606 stops being a formality handled in a workbook.

External scrutiny arrives

A first audit, a lender covenant, a growth round, or an acquisition approach. Systems that were adequate internally get examined by someone paid to be sceptical.

Why this size is where the market fails you

Below $25M, QuickBooks or Xero with good process is genuinely adequate and most vendors selling upmarket are overselling. Above $150M, the mature systems earn their implementation cost — the breadth, the statutory depth, and the audit precedent are worth the nine months.

In between is the band where neither answer fits well. The accounting system has stopped being enough and the ERP options were designed for companies twice your size, with implementation timelines and partner dependencies calibrated accordingly.

The mid-market is not a smaller version of the enterprise. It is the size where the available answers are both wrong in opposite directions.

What we would actually do at this size

Almost always in this order, and it takes weeks rather than quarters because it does not start with replacing the ledger.

  • Reporting first. Department, location, and entity P&L on top of whatever you run. It is the thing being asked for and it is usually recoverable from data you already have.
  • AP second. The largest block of manual work and the one that would otherwise justify a hire in the next twelve months.
  • Close third. Daily reconciliation and a structured checklist, which is where the duration number moves.
  • Ledger last, if at all. Only where multi-entity elimination, ASC 606, or fixed assets are the binding constraint — and only after a shadow ledger has tied for three closed months.

When you should look upmarket instead

Three cases, and they are not marginal at this size. If you manufacture or hold real inventory, NetSuite or Acumatica is a better answer and we do not build that functionality. If you file statutory returns in several countries, that is two decades of accumulated work we have not done. And if a sponsor or acquirer expects a recognisable system, that is a rational procurement preference rather than a technical argument.

In each of those we would rather implement the right system for you — we do that as a service — than sell you ours and have the constraint surface in month four.

Where to start

A realistic first quarter at this size.

Weeks 1–3

Reporting on your current ledger

Read-only connection and the dimensional P&L somebody has been asking for. It also establishes whether the dimensions are recoverable or need fixing at entry.

Weeks 3–8

Accounts payable

Drafted and reviewed at first. This is the workflow that would otherwise justify a hire, and the one where the corpus builds fastest.

Weeks 8–12

Close and reconciliation

Daily reconciliation and a structured checklist with owners. Most customers see close duration fall by a third by the second cycle.

Month 4+

Decide about the ledger

With a shadow ledger having reconciled for a quarter, that becomes an evidence-based decision rather than a leap — and plenty of companies decide not to.

Questions

What people ask.

Is $25–100M really different from $10–25M?
Materially, and the difference is usually entities and external scrutiny rather than volume alone. A $15M single-entity business with no audit is a different problem from a $60M three-entity business with a lender.
Do we need to replace our accounting system?
Not necessarily, and roughly a third of companies at this size that assess with us conclude they should not yet. Reporting and AP automation resolve the pain for a large share.
How long does a real implementation take at this size?
Eight to twelve weeks for a fixed-scope implementation on a traditional system, or weeks for our layer approach because it does not replace the ledger first. Anyone quoting under eight weeks for a full ERP implementation has not finished scoping.
What if we are growing fast?
Weight time-to-value heavily. A nine-month implementation at $40M finishes when you are $60M with different requirements, which is a common way projects get judged as failures despite being delivered as specified.
Should we hire a CFO first?
Frequently yes, and we will say so. A systems decision made without someone who will own the numbers tends to optimise for the wrong things.

Find out what to fix first.

Three questions about entities, close, and volume, and a written answer including when to change nothing.