When a company has 20 people, IT infrastructure is a $50 router, a two-bay NAS, and a Wi-Fi password on a sticky note. No sysadmin — “Pete from sales knows computers.” Everything works. Then the company grows, and infrastructure starts crumbling — not all at once, but layer by layer, like a building whose foundation was never reinforced.
Over the past 5 years, we have supported the growth of 14 companies from startup to mid-market. The pattern is always the same: there are three thresholds where old infrastructure breaks. 30 people, 80 people, and 150 people. Each threshold demands fundamentally different solutions.
Threshold 1. Thirty people: when the consumer router gives up
Typical situation. The company grew from 12 to 30 employees in a year. Added a second office or expanded the first. Wi-Fi becomes unreliable — a home router handles 10-15 devices, but 40 are now connected (everyone has a laptop plus a phone). Files on the NAS open slowly. The printer is visible to only half the computers. VPN for remote workers runs on the same router and crashes under load.
What you need at this stage:
Proper network equipment. A managed switch, business-class Wi-Fi access points (Ubiquiti, MikroTik), a router with VLAN support. Budget: $800-1,500. It seems expensive compared to a $50 consumer router, but one day when 30 people cannot work because the network is down costs more.
A file server instead of a NAS. A two-bay NAS is a home solution. For 30 users you need at least an entry-level server with 4 drives in RAID-10, ECC memory, and a proper controller. Or a move to cloud storage — depends on how critical the data is and whether it can be stored off-site.
At least one person responsible for IT. Not necessarily a full-time sysadmin. For 30 people, outsourcing is sufficient: a contractor visits once a week, comes urgently when needed. Cost: $300-500/month. “Pete from sales” is no longer enough and, more importantly, should not be — he has a different job.
Threshold 2. Eighty people: when you need architecture
At 80 employees, problems grow non-linearly. No longer one office but two or three. Remote workers appear. Three to five critical systems are running: accounting, CRM, email, file storage, possibly ERP or specialized software. Each system sits on a separate server (or on the same server — which is worse).
Typical problems at this threshold:
Every employee has a separate password for everything. Email password, accounting system password, CRM password, VPN password. Someone leaves — someone forgets to disable their access to one system. Six months later you discover a former manager still has access to the client database. At this point you need Active Directory or equivalent — centralized account management. One login for everything. Employee leaves — one button revokes access to all systems.
Backups run, but nobody verifies them. We audited a 90-person company. Daily backup of their accounting database was configured and had been running reliably for 2 years. Testing revealed: it copied the database but never verified integrity. The last restorable backup was 8 months old — the rest were corrupted. Eight months of work could have simply vanished.
No documentation. The sysadmin (now full-time — at 80 people you cannot go without) configured the network, servers, VPN, email. Everything is in their head. They go on vacation — if something breaks, nobody knows where to look. Extended illness — catastrophe. This is not the sysadmin’s fault. It is a management failure: nobody required documentation.
What you need at this stage: server virtualization (minimum two physical hosts so VMs migrate when one fails), centralized user management, infrastructure-wide monitoring, documented network diagrams, regular backup restoration tests.
Threshold 3. One hundred fifty people: when IT becomes strategy
Beyond 150 employees, IT stops being “support” and becomes part of business strategy. At this level, the company likely spends 3-7% of revenue on IT. A CTO appears (or should).
Problems at this threshold are qualitatively different. Not “Wi-Fi is down” but “how do we secure data across three offices and 20 remote workers.” Not “the server is slow” but “how do we ensure 99.9% uptime for critical systems — that is 8 hours 45 minutes of downtime per year.”
At this stage you need: a failover server cluster, DLP for data leak prevention, automated workstation provisioning (a new hire gets a configured laptop in an hour, not three days), a proper ticketing system, role-based access control.
Infrastructure budget: $50,000-150,000 upfront + $2,000-5,000/month for support and licenses. Sounds like a lot. But losses from a single serious security incident or a day-long outage at this scale run into hundreds of thousands.
The biggest mistake growing companies make
Postponing infrastructure upgrades until everything breaks. It is not greed — it is inertia. A founder remembers that with 20 people everything worked fine, and subconsciously expects it will still work at 80. It will not.
Building infrastructure proactively costs 2-3 times less than rebuilding in an emergency. Planned upgrades are predictable. Emergency upgrades mean “call an engineer on Saturday, everything is down” with double rates and quick-fix solutions that have to be redone later.
We design IT infrastructure with room for growth. Hardware selection, network setup, monitoring, ongoing support. If your company is growing and you feel the infrastructure is not keeping up — get in touch. We will audit your setup and show what needs to change now and what can wait.