IT Documentation for Small Business: What to Write Down Before Your Tech Person Leaves
Every small business has one. The person who knows where things are. Which registrar holds the domain. Which admin account can reset the point-of-sale terminal. Why the office router has that one strange port-forwarding rule nobody wants to touch. Sometimes it’s an internal employee, sometimes a contractor, sometimes the owner’s nephew from six years ago. And when that person leaves — or gets busy, or stops answering — the business discovers that a meaningful chunk of its operations lived in someone’s head rather than anywhere you can actually look it up.
That’s the problem IT documentation for small business solves. Not compliance theater, not a 200-page manual nobody reads. Just a written, findable, current record of how your technology is set up and who controls it. We’ve seen companies lose a week of productivity — and real revenue — because nobody could prove ownership of a domain name during an outage. The fix costs a few afternoons. The absence of it costs far more, and always at the worst possible moment.
Why Undocumented Systems Are a Business Risk, Not an I.T. Problem
It’s tempting to file documentation under “nice to have.” It isn’t. Undocumented systems create concrete, recurring business exposure:
- Key-person dependency. One resignation, illness, or falling-out and your ability to operate degrades immediately.
- Slower incident recovery. During an outage, the first hour is usually spent figuring out what exists and who can log in — time you don’t have.
- Vendor lock-in by accident. If your IT provider is the only one who knows how things are wired, switching providers becomes a project instead of a decision.
- Duplicate spend. Nobody knows what’s already licensed, so it gets bought again.
- Security gaps. Accounts belonging to former staff stay live because no one has a list to check against.
- Failed audits and insurance claims. Cyber insurance applications and client security questionnaires increasingly ask questions you can only answer from documentation.
If losing one person would stop you from logging into something you depend on, you don’t have an I.T. system. You have an arrangement.
The goal isn’t to write everything down. It’s to make sure that a competent replacement — a new hire, a new provider, or you at 11 p.m. during an outage — can find out what exists, get access to it, and understand why it’s configured the way it is.
The Minimum Viable Documentation Set
Start here. If you have nothing today, these six records will cover the majority of your practical risk. Everything else is refinement.
1. The Asset and Account Inventory
A single list of everything your business depends on that has a login or a bill attached. For each entry, capture the vendor, what it’s used for, the billing owner, the renewal or expiration date, the account or customer number, and who inside your company owns the relationship. Include the boring infrastructure — domain registrar, DNS host, web host, email platform, phone system, backup service, payment processor, accounting software, CRM, and any industry-specific tool your operation runs on.
Most owners are surprised by this exercise twice: once at how many services they’re paying for, and once at how many are billed to a personal credit card or a departed employee’s email address.
2. The Access and Ownership Map
Credentials themselves belong in a password manager, never in a spreadsheet or a shared doc. But the map of who has access to what belongs in your documentation: which accounts exist, which are admin-level, which have multi-factor authentication enabled, and which are shared versus individual. This is also where you record the recovery path — the backup email, the recovery phone number, and the person who controls them. Recovery contacts are the single most commonly overlooked item we encounter, and they’re the ones that lock businesses out permanently. If you want to go deeper on the access side, our guide to identity and access management for small business covers how to structure permissions properly.
3. The Network and Infrastructure Diagram
It does not need to be beautiful. A photo of a whiteboard beats nothing. Capture what connects to what: internet provider and account number, modem and router models, where the equipment physically lives, the guest and internal Wi-Fi setup, any static IPs, VPN configuration, and any on-premise servers or NAS devices still in the building. Note the things that would confuse a stranger — the port forward, the odd VLAN, the switch in the closet that isn’t labeled.
4. The “How Do I…” Runbooks
Short, step-by-step procedures for the tasks that happen often enough to matter but rarely enough that nobody remembers them. Typical examples: add a new employee, remove a departing employee, reset the payment terminal, publish a page to the website, restore a file from backup, forward the main phone line after hours, or update a DNS record. Each runbook should be short enough to fit on one screen and specific enough that someone who has never done it can follow along.
5. The Vendor and Support Contact Sheet
Who do you call, and what do they need from you? Support phone numbers, portal URLs, account numbers, contract terms, response-time commitments, and the name of an actual human where you have one. Add your escalation order: first call, second call, who has authority to approve emergency spend.
6. The Recovery Basics
Where backups live, what they cover, how often they run, when they were last verified, and the actual steps to restore. A backup nobody has tested is a hypothesis, not a safeguard. This overlaps with your broader continuity planning — see our disaster recovery plan for small business for how to turn these notes into a plan you can execute under pressure.
What Good Documentation Actually Looks Like
The most common failure isn’t writing too little. It’s writing the wrong thing. Here’s the difference in practice:
| Weak documentation | Useful documentation |
|---|---|
| “Website is hosted with our web guy” | Host name, account number, billing owner, renewal date, support URL, who has admin access |
| “We back things up” | What’s backed up, where, how often, retention period, last successful restore test and date |
| A 40-page PDF written two years ago | A short living page per system, with a “last reviewed” date on each |
| Passwords in a shared spreadsheet | Credentials in a password manager; documentation records who holds access, not the secrets |
| “Ask Dave” | A runbook Dave wrote, that someone else has successfully followed at least once |
Three principles keep documentation useful over time. First, date everything — a page with no review date can’t be trusted, and trust is the whole point. Second, write for a competent stranger, not for yourself; the test is whether a new hire could follow it without asking a follow-up question. Third, keep secrets out of documents and in a password manager where access can be granted and revoked cleanly.
Where to Keep It
Documentation that lives in one person’s Google Drive folder has simply moved the single point of failure. A few practical rules:
- Store it somewhere the company owns. A company-controlled workspace, not a personal account, and not a contractor’s system.
- Make it searchable. A shared wiki, a documentation tool, or a well-organized shared drive all work. Fancy tooling is optional; findability is not.
- Keep an offline copy of the critical parts. If your identity provider is down, your documentation about how to fix your identity provider needs to be reachable another way. A printed or encrypted-offline copy of the recovery essentials is worth the small effort.
- Control who can edit. Everyone should be able to read what’s relevant to them. A smaller group should be able to change it.
Documentation you can’t reach during an outage is documentation you don’t have.
How to Build It Without Stopping Everything Else
Nobody has a free week. Do it in slices, in this order:
- Week one: the inventory. Pull the last three months of card and bank statements and list every recurring technology charge. This single step surfaces most of what you own — including things you’d forgotten you were paying for.
- Week two: ownership and recovery. For each item on that list, confirm the account is in a company-controlled email, the billing contact is current, and multi-factor authentication is on. Fix what isn’t.
- Week three: the top five runbooks. Ask your team which tasks generate the most “how do I…” questions and write those first.
- Week four: infrastructure and vendors. Diagram the network, photograph the equipment closet, and build the contact sheet.
- Ongoing: capture as you go. The cheapest documentation is written the moment something changes, by the person changing it.
Tie it to events you already have. Every new hire and every departure is a natural checkpoint — our IT onboarding and offboarding checklist pairs directly with the access map above. Every vendor renewal is a prompt to confirm the inventory entry is still accurate.
Keeping It Alive
Documentation decays. Systems change, vendors get replaced, people move on, and a record that was accurate in March quietly becomes misleading by October — which is worse than having none, because people act on it. Two habits prevent this. Put a recurring quarterly review on someone’s calendar with a named owner, not a vague team responsibility. And make updating the documentation part of the definition of “done” for any change: if you migrate the email platform, the migration isn’t finished until the record reflects it.
Once a year, run a simple test. Hand your documentation to someone who wasn’t involved in writing it and ask them to complete one routine task using only what’s written. Whatever they get stuck on is your next update. It takes twenty minutes and it’s the only honest measure of whether IT documentation for small business is doing its job.
The Payoff
Good documentation makes your business more valuable and less fragile at the same time. Onboarding gets faster. Outages get shorter. Vendor negotiations get easier because you know exactly what you have and what it costs. Security questionnaires and insurance applications stop being fire drills. And if you ever sell the business or bring on a partner, a clean record of your systems is one of the clearest signals that the operation is professionally run.
Most importantly, it removes the quiet anxiety that comes from depending on one person’s memory. That’s not a technology upgrade — it’s an operational one.
If you’d like help building this out, or you’re not sure what you already have, we do this work for clients across Orange County and beyond as part of our I.T. services and communications practice. Reach out through frozencrow.com for a free, no-obligation consultation and we’ll walk through your setup with you. Our team, your goals.







