If you work in software, you probably take a few things for granted. Your laptop is managed. Secrets live in a vault. Backups get tested. Someone offboards people when they leave.

Now picture a business where none of that is automatic, the “office” is a trailer that disappears in eight months, and the laptops spend their lives in truck beds. That’s a lot of small construction companies, and the gap between how developers handle risk and how they handle it is surprisingly fixable with ideas we already know.

I’ve been reading about this lately, and here’s how I’d translate common engineering habits for a contractor.

  1. Treat laptops like cattle, not pets.

In ops, the idea is that servers should be replaceable. If one dies, you reprovision it. The same mindset works for field laptops.

Build every laptop from a standard image or provisioning profile.
Enforce full-disc encryption by policy.
Enable remote lock and wipe.
Patch on a schedule, not “whenever someone restarts”

The goal is that losing a laptop is boring. A technician reprovisions another one, and the user is working again the same day, because nothing important lived only on that machine.

  1. Assume the network is hostile.

Zero-trust thinking fits job sites perfectly. The “office network” might be a cellular router shared with subcontractors, inspectors, and an owner’s rep. You can’t trust it.

So make identity and devices the perimeter:

Put guests and subs on a separate network, away from estimating and accounting systems.
Prefer cloud-first apps so crews only need a browser and a connection.
Test the signal before the project starts, and keep a backup connection.

  1. Enforce MFA and stop sharing credentials.

The most common breach path isn’t clever. It’s a stolen or reused password. Two controls close most of that door:

MFA on email, accounting, cloud storage, and project tools
A team password manager (with shared vaults, audit logs, and admin recovery) that works on phones, since that’s where field staff log in

Also kill shared crew logins. One shared username and password means no accountability and no clean way to revoke access.

Email comes first. It controls password resets for almost everything else.

  1. Make offboarding a checklist, not a memory test.

Seasonal hires and subcontractors mean constant turnover, and in many small companies former employees keep access for months. Here’s what a simple offboarding checklist looks like if you write it like a config:

yaml
offboarding:

  • disable_email_account
  • remove_shared_folder_access
  • revoke_password_manager_vault_access
  • collect_company_devices
  • rotate_any_shared_credentials
  • log_date_and_completed_by

It’s unglamorous, but ten minutes of process close a surprisingly large gap.

  1. The backup that was never restored isn’t a backup.

Developers know this one, but it’s worth repeating, because it’s the most commonly skipped control in small businesses.

Follow 3-2-1: three copies, two storage types, and one off-site and isolated from the main network (ransomware hunts for connected backups).
Don’t assume SaaS is backed up. Many cloud tools protect their platform, not your accidental deletions.
Run restore tests at least quarterly, using real data like a recent bid folder.
Write down your recovery targets: how long can you be without email, accounting, and plan sets?

A job that completes tells you nothing about whether the data is usable. Only a restore does.

  1. The threat model is different: bid data and payment fraud

Here’s where construction diverges from the usual app-security story. A bid file contains labour rates, supplier pricing, and margin strategy, which is valuable to competitors and criminals. And large payments plus long email chains between parties who don’t always know each other make fake invoices and “update our bank details” scams very effective.

The cheapest effective control isn’t a tool. It’s a rule:

Verify any change to payment details by phone, using a number already on file, never the one in the email.

Technical controls like impersonation filtering and link scanning help, but this human-layer control is the one worth writing into policy.

A 30-day starting plan

If you’re helping a friend, family member, or client in a field-heavy business, sequence by risk:

Week Focus Action

  1. Inventory list every device, router, and cloud account, and who can access what
  2. Close big gaps Enable MFA, deploy a password manager, and verify encryption.
  3. Backups: Verify scope and destination, then run a restore test.
  4. Standardise: Create a standard device profile, a site connectivity checklist, and an incident sheet. Want the non-developer version?

I found a plain-English guide that covers all of this for contractors, including job site internet options, rugged laptop choices, how to compare break-fix vs. in-house vs. managed support, and seven questions to ask before hiring a provider: IT support for construction company job sites, laptops, and bid data.

It’s from ZIA Networks, an IT services company in New Mexico that supports construction firms.