Business Data Backup and Recovery: Why One Copy Is Never Enough

Backup & Recovery

Business data backup and recovery is not a storage bill. It is the answer to one question: if you needed your files back tomorrow, could you actually get them?

Picture a normal Tuesday where somebody needs a file from four years ago, and it is not where it should be, so you go looking in the backup you have been paying for every month and the login does not work. You call the company to ask why, and nobody picks up, because the company does not exist anymore. Your data is fine. It is sitting on a working drive in a working building, and you cannot reach a single byte of it.

That is close to what happened this year to a public television station in St. Louis, at a scale most businesses will never deal with. Roughly 50 terabytes of footage, going back about seventy years, was stored with a cloud vendor that stopped answering the phone in February and then went out of business. The drives were sitting in a colocation facility, and the facility took the position that the hardware belonged to its customer and not to the station, so opening it up was not something they could do. The station spent months in court arranging a supervised recovery, and as of this writing both sides are still filing progress updates with a judge.

50TBof archive footage locked behind a closed vendor
70 yrsof programming at stake, including local history
6 mofrom lost access to a court-approved recovery plan

Here is the part worth sitting with. Nobody at that station did anything careless. They paid their invoices for years, they renewed on time, and they had no way to see the trouble coming, because the failure happened one level above them inside a vendor relationship they were never shown.

Plain English

If you have one copy of your data, you do not have a backup. You have a single point of failure with a monthly invoice attached to it.

01

The failure was not a fire or a hacker. It was a business closing its doors.

When people imagine losing everything, they imagine something dramatic, like water coming through the ceiling above the server or a ransom note on every screen in the office on a Monday morning. This one was much quieter than that, and that is exactly why it is worth paying attention to.

The closest everyday comparison is a storage unit. You rent a unit from a company, you pay every month, and everything inside it is yours. Then the company that rented you the unit goes under, and it turns out they were leasing the building from somebody else, and that somebody else has no record of you, no contract with you, and no legal reason to cut the lock for you. Your furniture is right there behind a door you cannot open, and being the rightful owner does not get you inside any faster.

Swap the furniture for your client records, your job files, your accounting history, and your email archive, and you have the situation the station walked into. Here is how quietly it unfolded.

Feb The vendor stops responding. Renewal time comes around and nobody at the storage company picks up the phone.
Mar Access is cut with no warning. The contract allowed a window to pull data out. That window never opened.
Apr Lawyers get involved. The station files suit and finally confirms which building its data is physically sitting in.
Jun A court agrees the station owns the data. Ownership was never the problem. Reaching it was.
Aug A supervised recovery is approved. A third party pulls the data off the hardware, back storage fees get paid, and updates are still due to the judge.

Smaller versions of this happen constantly and never make the news. A bookkeeping platform sunsets with ninety days notice, a vendor gets acquired and the old system goes dark, or the one person who set up the backup portal under their personal email leaves the company. None of that is a disaster on paper. All of it turns into somebody spending a week rebuilding records that should have taken twenty minutes to restore.

02

One backup is not a backup plan

The rule that has held up for decades is called 3-2-1, and it is deliberately simple, because a plan you cannot remember is a plan nobody follows.

Three Copies of your data The live data you work in, plus two real backups
Two Kinds of storage Local and cloud, because they fail in different ways
One Copy kept offsite Somewhere a pipe, a break-in, or bad wiring cannot reach

The station in this story had plenty of storage. What it did not have was independence, because every copy it counted on ran through the same vendor relationship, so when that one relationship broke, access to all of it broke at the same time. That is the piece worth borrowing no matter the size of your business, and it costs nothing to think about before you need it.

03

Physical media is fading, and the risk moved instead of disappearing

We will admit a soft spot here, because there is something genuinely reassuring about media you can hold in your hand, whether that is a labeled drive on a shelf or a disc in a sleeve. You can see it, you can carry it out of the building, and nobody can switch it off from a keyboard three states away.

That era is mostly over for business use, and for practical reasons rather than fashionable ones. Tape needs handling and testing, optical media degrades sitting still, and any plan that depends on a person physically swapping a drive every Friday quietly stops happening around week six in a busy office. Cloud backup is faster to restore from, easier to verify, and much harder for everyone to forget about, which is why it won.

What convenience actually did

It did not remove the risk. It moved it.

When your backup lived on a shelf, the things that could hurt you were fire, theft, and human forgetfulness.

When your backup lives in a service, the things that can hurt you are account access, vendor stability, and whatever the contract says about getting your data out.

Both lists are real, and only one of them ever ends up in a news story.

So the honest version is not that cloud is good and physical is bad. It is that you should be able to say out loud where every copy of your data lives, whose name is on the account it lives in, and how fast you could pull it out if you had to leave tomorrow. That thinking is baked into how we set up cloud services for Metro Detroit clients, and it is the same reason we will never call one synced folder a backup.

04

What business data backup and recovery looks like when it actually works

Backup is the copy you make. Recovery is the part that saves your week. Most businesses buy the first one and assume the second comes along with it, and that assumption is where the pain usually lives, because technology does not create a good outcome on its own. Good businesses use it well, and using it well here means somebody owns these five things.

The backups run on a schedule and a real person checks them

Backup jobs fail quietly and politely. A drive fills up, a credential expires, a server gets renamed during a project and nobody updates the job, and the copy stops happening while the dashboard still looks calm. Somebody has to read those alerts and act on them, which is the entire difference between managed backup and a piece of software nobody has logged into since it was installed.

Restores get tested instead of assumed

A backup you have never restored from is a theory, and theories do not hold up on the worst day of your year. Testing means actually pulling back a file, then a folder, then ideally a whole system, so you know the data is usable and the process works when people are stressed and the phones are ringing.

You know your recovery time before you need it

There are only two questions underneath all of this. How much data can you afford to lose, counted in hours, and how long can you be down before it genuinely hurts the business. Those two answers decide how often your backups run, where the copies live, and what gets restored first, and asking them is not overthinking it. We used to believe the fastest answer was the best service. Asking a few more questions up front is what actually keeps a bad week from turning into a bad month.

Access does not hang on one person or one vendor

Your business should own the accounts, with more than one administrator, so you can reach your own data without needing a specific provider's cooperation or a former employee's password. This is the exact hinge the story above turns on, and getting it right up front costs you a ten-minute conversation.

Backups are kept separate from your live network

Ransomware goes looking for backups first, because attackers know that a company with clean copies has options and a company without them does not. Copies that are isolated or cannot be altered after they are written are what let a business restore instead of negotiate, and if that side of things is new to you, our approach to managed security covers how the two pieces fit together.

Two-minute 3-2-1 self-check

Check the boxes you can honestly say yes to today. Nothing gets saved or sent anywhere.

0 of 6 checked

Start checking boxes and you will see pretty quickly where your setup stands.

05

Four backup mistakes we run into the most

01

Treating file sync like a backup

OneDrive, Dropbox, and Google Drive are built to mirror your files, which is useful right up until the thing being mirrored is a deletion or a folder somebody just encrypted, because then the copy matches the damage within seconds.

What to do instead: run a real versioned backup alongside the sync, with a retention window long enough to reach back past the day things went wrong.

02

Backing up the server and forgetting where the data actually went

Your records have spread out over the last several years into Microsoft 365 mailboxes, SharePoint, QuickBooks Online, your CRM, and whatever industry software your team lives in, and a lot of those platforms protect their own infrastructure rather than protecting your specific data from an honest mistake made inside your company.

What to do instead: write out every system holding records you would need after a bad day, then confirm each one is genuinely covered instead of assumed.

03

Never testing a restore

This is the big one, and it is the reason most failed backups had been failing for months before anybody noticed, because nothing ever asked the system to hand something back.

What to do instead: restore something real on a schedule and write down how long it took, because that number is your actual recovery time whether you like it or not.

04

Not knowing whose name is on the account

If the backup portal is registered to a former employee's email address, or it sits inside a vendor's account with your data tucked in a folder, you are carrying a dependency you cannot see and cannot fix in the moment you need it. That is the trap the station in this story fell into.

What to do instead: move every backup account under a company-owned identity with at least two administrators who know they are administrators.

Almost nobody calls us because a backup just failed. They call because they found out it had been failing for months and nobody was watching it, and the restore test is the step everyone skips, even though it is the only one that proves the rest of it worked.

Simply Technology service team, Grosse Pointe Woods
06

Backup and recovery questions Michigan business owners actually ask

Three copies is the working standard, meaning your live data, a local backup you can restore from quickly, and an offsite or cloud copy that survives whatever happens at your office. For most businesses in the 15 to 100 user range that is very achievable without a big budget or a complicated setup, and it is usually cheaper than people expect once they see what it replaces.

Cloud backup is a strong piece of a plan and a weak plan by itself, because pulling a large amount of data back down over an internet connection takes real time and a single provider is still a single dependency. Pairing a local copy with a cloud copy is what gives you speed and separation at the same time.

Spot-check a file restore monthly and run a fuller recovery test at least twice a year, then test again any time you add a server, change a main application, or move something into the cloud. Those changes are exactly when coverage gaps show up, and they are the easiest gaps to catch early.

Managed backup is the ongoing service that keeps the copies running, the alerts watched, and the restores tested so the system stays trustworthy. Backup and recovery is the wider plan around it, covering how fast you need to be working again, what comes back first, and who does what while it is happening. You want both, and you want them written down before the day you need them.

What to take away from this

  • Losing your data does not require a disaster, because a vendor quietly closing its doors is enough to do it.
  • Three copies, two kinds of storage, one of them offsite. That structure is old and it is still the answer.
  • Cloud is the better tool for almost every business, but it moved your risk over to account access and vendor stability rather than erasing it.
  • A backup nobody has restored from is a guess, so test it and write down how long it took.
  • Own your accounts, and never let one vendor or one employee stand between your business and your own files.

The station in this story will most likely get its archive back, and it will have spent months and a pile of legal fees to end up exactly where it thought it already was. That is the whole lesson in one sentence. You do not find out whether your backups work on a calm afternoon, you find out on the worst day of the year, and by then the answer is already decided.

If you want a second set of eyes on how your current setup would actually hold up, that is a short conversation and not a sales pitch. We handle backup and recovery planning for businesses across Wayne, Oakland, Macomb, and Washtenaw counties, usually as part of fully managed IT so the whole picture stays in one place. Worth a read too: the difference between network monitoring and network management, which comes down to the same idea that somebody has to be watching.

Let's take a look at your backups

We will walk through where your copies live, whether a restore has ever been tested, and what a real recovery would look like for your business. Plain English, no runaround.

Schedule a quick backup review One vendor, all your tech. We'll Take IT From Here.

Reporting on the Nine PBS archive dispute via Ars Technica.