Every June, Florida businesses check their shutters and their generators. Almost none check their backups. Then a storm takes the power out for four days, a server does not come back up, and a business that survived the wind loses its records to a dead drive. After 16 years running infrastructure through Florida storm seasons, including keeping a Doral freight warehouse online through grid drops, here is the checklist we actually use.
This is not theoretical for us. In October 2024, Hurricane Milton spawned a record 45 confirmed tornadoes across Florida in a single day, including the EF3s that devastated the Port St. Lucie area while squall lines put Melbourne under repeated tornado emergencies. I personally helped three businesses in that region get back on their feet afterward: recovering what could be recovered, rebuilding what could not, and seeing up close which preparations had mattered and which had been theater. That experience shaped every item below.
Print this, or forward it to whoever handles your technology. The items are ordered by when they matter: before the season, when a watch is issued, and after the storm passes.
Before the season (do these now, they take a week)
1. Test a restore, not a backup. Every business says they have backups. Almost nobody has restored from one this year. Pick one critical system and actually restore a file from backup today; if it fails or nobody knows how, you found the problem while it is still free to fix. Our security baseline covers what a tested backup setup looks like.
2. Know where your data physically lives. If the answer is "a server in the office," that server is in the storm path with you. Critical data needs a copy that is out of the building, whether that is an encrypted cloud backup or a drive that leaves with you. If everything is already cloud based, verify your team can actually work from anywhere: laptops, logins, and MFA that does not depend on an office phone.
3. Protect the power path. Battery units (UPS) on every server, network rack, and phone system, with batteries tested, not assumed. Batteries degrade in two to three years and the flicker that precedes an outage is what corrupts equipment. We rebuilt exactly this for a Doral warehouse whose gear dropped with every storm season flicker; it has ridden through every season since.
4. Write the two phone lists. One: every employee, personal cell numbers included. Two: every vendor that matters (internet provider, IT, insurance, landlord, payroll). Print both. The moment you need them is the moment nothing electronic works.
When a watch is issued (72 to 48 hours out)
5. Run a fresh backup of everything and verify it finished. Not scheduled for tonight: now, watched to completion.
6. Shut down properly before the storm, not during it. A server that gets hard killed by an outage risks corruption; a server shut down cleanly the evening before boots back like nothing happened. Decide in advance who does this and when.
7. Physically elevate and unplug. Equipment off the floor (storm water finds floors), unplugged from the wall (surges arrive when the grid comes back), and away from windows. Ten minutes of moving hardware prevents most of the damage we see after storms.
8. Set communications expectations. Tell customers how to reach you and when you expect to reopen, on your Google profile, your voicemail, and your website. A one line update beats silence, and it is the difference between a customer who waits and one who calls your competitor.
After the storm
9. Power up in order. Network equipment first, wait for the internet to stabilize, then servers, then workstations. Bringing everything up at once on a shaky grid is how surge damage happens after the storm already spared you.
10. Verify before you trust. Check that backups resumed, that security cameras recorded, that phones ring, and that card payments process, before the first customer walks in. The failures that cost the most are the quiet ones nobody notices for a week.
If something did not survive: stop, do not reformat or replace anything until the data situation is clear. Drives that seem dead are often recoverable, and the worst losses we see come from well meaning cleanup done in a hurry.
The honest pitch, and the free version
Everything above is doable without hiring anyone: that is why we published it. If you would rather have it done for you, storm readiness is built into our managed IT plans: pre season checks of every client backup, UPS batteries, and recovery plans, because we live here too.
And if you just want a second opinion on whether your business would survive a four day outage, a free assessment answers exactly that, no purchase required. Hurricane season runs through November 30. The checklist works best in August, not the night a cone appears on the map.
Want your storm readiness checked by an engineer instead of a checklist? The free assessment covers backups, power, and recovery, and gives you a straight answer.