Companies do not scale smoothly; they break at predictable points. Knowing where lets you build slightly ahead of the failure instead of after it.
- ›What breaks at 15, 30, 50, and 100
- ›What to build at each stage
- ›The over-correction to avoid
- ›Signals you are behind
Around 15 people
What breaks: informal knowledge. Nobody can hold who does what, who is where, and what was decided.
What to build: one place for work, one for people records, and written decisions. Not process, just records. See HRIS for small business.
What not to build: management layers, formal reviews, or a handbook nobody reads.
Around 30 people
What breaks: the founder as the coordination layer. Everything routing through one or two people becomes the constraint on everything else.
What to build: clear ownership per area, a weekly operating rhythm, and delegated decision authority with explicit boundaries. See the weekly business review.
What not to build: an org chart with more layers than the work requires.
Around 50 people
What breaks: consistency. Two teams do the same thing differently, quality varies by who did it, and new joiners get different experiences depending on their manager.
What to build: documented processes for the things that must be consistent (hiring, onboarding, delivery standards), and real management capability. Most companies at this stage have promoted good individual contributors into management without training them, and it shows.
What not to build: process for things that do not need to be consistent. Over-standardization at this stage kills the speed that got you here.
Around 100 people
What breaks: visibility. Leadership no longer knows what is happening without a reporting structure, and the informal signals are gone entirely.
What to build: proper reporting and forecasting, defined career paths, and systems that produce information rather than requiring someone to assemble it. This is where the manual reporting burden becomes untenable and automation stops being optional.
What not to build: an enterprise process framework for a 100-person company.
The over-correction to avoid
The most common scaling mistake is building process for the company you fear becoming rather than the one you are. A 40-person company with 200-person process moves like neither.
Build one stage ahead, not three. The signal to add structure is that its absence is causing repeated, visible pain, not that you can foresee needing it eventually.
Tip: At every stage, ask what a new joiner would need to be effective in week one. The gap between that and reality is your next infrastructure project, and it is almost always more useful than whatever framework you were considering.
Signals you are behind
- ›The same question gets asked repeatedly with different answers.
- ›Information exists but nobody can find it.
- ›Decisions get relitigated because nobody recorded them.
- ›Onboarding depends on who happens to be free.
- ›Leadership finds out about problems from customers.
FAQ
When should we hire an operations person?
Usually 25-40 people, when a founder spends more than a day a week on internal coordination.
How much of this is tooling?
Less than vendors suggest and more than founders expect. Tools make good practices cheap; they do not create them.
What if we grow faster than this?
The same breaks happen, compressed. Fast growth means building slightly ahead rather than reactively, because the reactive window closes.
Brainis gives a growing company one system for work, people, revenue, and finance, on every plan at every stage. See pricing.
Sharing insights on business operations, AI, and modern team management.
Run your company on Brainis
All 11 Operating Systems on every plan, from $29 a month. No per-seat pricing — you pay for AI capacity, not headcount.
See pricing