5 min read

RTO vs. RPO: How Bad Does an Outage Get?

RTO vs. RPO: How Bad Does an Outage Get?

When I sit down with a business owner to review their backups, I ask two questions before we talk about any technology. If your main server died at 10 a.m. today, when do you need it back? And how much of this week’s work could your team redo without real damage?

Most owners have never been asked, so they’ve never considered. But those two answers are so important when putting together a recovery plan. They're called your RTO and RPO, and understanding RTO vs. RPO is the difference between an inconvenient afternoon and a lost month.

This post is part of our ransomware recovery series. It covers what each number means, how they differ, how to set them for your business, and why ransomware makes both harder to hit than most plans assume.

 

Key Takeaways

  • RTO (recovery time objective) is how long a system can be down; RPO (recovery point objective) is how much data, measured in time, you can afford to lose
  • Both are business decisions first, and each critical system should have its own targets
  • In Veeam's 2026 research, 90% of leaders were confident they could recover within their RTOs, yet only 28% of ransomware victims fully recovered their data
  • After ransomware, your real recovery point is your last clean backup, which may be older than your last backup

 

Table of Contents

  1. RTO vs. RPO: Table Comparison
  2. Hand the Calculator to Your Controller
  3. Not Every System Gets the Same Numbers
  4. Ransomware Bends Both Numbers
  5. The Confidence Gap
  6. Pressure-Test Your Numbers This Week

 

RTO vs. RPO: Table Comparison

RTO is the maximum time a system can stay down before the outage seriously hurts the business. RPO is the oldest point in time you can afford to restore data from, which sets how much recent work you're willing to lose. RTO looks forward from the moment of failure while RPO looks backward.

NIST defines the recovery time objective as how long a system can be in recovery before it damages the organization's mission or business processes. It defines the recovery point objective as the point in time to which data must be recovered after an outage.

Here's an example:

  • 2:00 a.m.: Your nightly backup runs.
  • 10:00 a.m.: The accounting server fails.
  • 3:00 p.m.: The server is restored and running again.

In that example, the business was down for five hours. That's your total recovery time. Everything entered between 2:00 a.m. and 10:00 a.m. is gone, because the newest copy was from 2:00 a.m. That eight-hour gap is your recovery point.

If your targets were a four-hour RTO and a one-hour RPO, you missed both. And nobody would have known until the worst possible day.

The easiest way to keep them straight: RTO is about downtime, and RPO is about data loss. You improve RTO with faster restore methods and practice. You improve RPO by backing up more often.

 

RTO (Recovery Time Objective)

RPO (Recovery Point

Objective)

The question it answers

How long can we be down?

How much recent work can we lose?

Direction

Forward from the outage

Backward from the outage

Measured in

Minutes, hours, or days of downtime

Minutes, hours, or days of lost data

What drives it

Lost revenue, idle staff, deadlines, penalties

How fast data changes and how hard it is to re-create

What improves it

Faster restore options, standby systems, a practiced runbook

More frequent backups or continuous replication

What makes it worse

Untested restores, slow internet, nobody knowing who does what

Backups that run once a night, or backups that silently fail

 

Hand the Calculator to Your Controller

RTO and RPO should be set by what downtime and data loss cost your business. Start with idle payroll and lost revenue per hour, add any contract penalties, and compare that to what faster recovery costs. The right target is where the two lines cross.

Here's a simple way to estimate what an hour of downtime costs:

  1. Idle labor: Number of people who can't work × their loaded hourly cost
  2. Lost revenue: Sales, billable hours, or production you can't get back
  3. Penalties and knock-on costs: Late fees, liquidated damages, missed payroll, overtime to catch up

A 20-person office with a $50 loaded hourly cost burns $1,000 an hour in idle wages alone, before a single lost sale. A contractor facing liquidated damages or a manufacturer with a stopped line can multiply that quickly.

Then look at data loss the same way. If your team enters 40 invoices a day, a 24-hour RPO means re-creating up to 40 invoices from email, paper, and memory. Some data is easy to rebuild. Some, like a morning of design changes or a batch of signed documents, isn't.

Tighter targets cost more. Backing up every 15 minutes and keeping a standby server ready costs more than a nightly backup. That's fine, as long as the spend matches the risk. The goal is to buy the recovery you need and stop there.

 

Not Every System Gets the Same Numbers

Every critical system should have its own RTO and RPO. Your accounting and identity systems usually need the tightest targets, while archives and old project folders can wait days. Grouping systems into tiers keeps the plan simple and puts the budget where it matters.

Here's a starting point we use with many small and mid-sized Colorado businesses. Your numbers should come from your own cost math.

 

Tier

Example Systems

Typical RTO

Typical RPO

Tier 1: Run the business

Logins and Microsoft 365 admin, accounting, ERP or job cost, line-of-business apps

A few hours

15 minutes to 1 hour

Tier 2: Do the work

Active project files, shared drives, CRM, email

Same day to next day

A few hours

Tier 3: Reference

Archives, closed projects, old file shares

Several days

24 hours

These tiers line up with the restore order we laid out in our ransomware recovery guide. Systems that control who can log in come back first, then the systems that move money, then active work.

 

Ransomware Bends Both Numbers

Ransomware makes both targets harder to hit. Your RPO shifts back to the last backup taken before the attacker got in, which may be days or weeks old. Your RTO grows because the business has to investigate and clean systems before restoring anything safely.

Most RTO and RPO plans are written with a dead server or a flood in mind. Ransomware is different in two ways.

  • Your newest backup may not be clean. Attackers often sit inside a network for a while before they launch the encryption. If they were in for two weeks, your last two weeks of backups may include their tools. Your real recovery point is the newest backup that predates them.
  • Restoring isn't the first step. Before any data comes back, someone has to figure out how the attacker got in, close that door, and confirm the systems you're restoring to are clean. That time counts against your RTO, and it's usually the longest part.

The research shows how often plans miss. An Omdia study found that 64% of organizations had an outage that ran longer than the downtime they'd planned for.

Backups also have to survive the attack to count at all. That's why we keep pushing immutable backups, which can't be changed or deleted by anyone until their retention period ends, even someone holding your admin password.

 

The Confidence Gap

Most businesses believe they'll hit their recovery targets, and most never test whether they can. That gap between confidence and proof is where recovery plans fail.

Veeam's 2026 Data Trust and Resilience Report surveyed more than 900 IT, security, and risk leaders. Ninety percent said they were confident they could recover within their RTOs, but only 28% of organizations hit by ransomware fully recovered their data. On average, they got back about 72% of what was affected.

The same report found only 69% said their RTOs were fully aligned with their business continuity goals. In other words, about a third of organizations had recovery targets that didn't match what the business actually needed.

In my experience, the gap almost always comes from the same place: the targets were written down once, the backup software showed green checkmarks, and nobody ever timed a real restore. An RTO you've never tested is a guess. We'll cover how to run a proper restore test in an upcoming post in this series.

 

Pressure-Test Your Numbers This Week

Find out whether your RTO and RPO are controlled by asking your IT provider five questions:

  1. How often does each critical system back up? That interval is your real RPO, no matter what the plan says.
  2. When did we last do a full test restore, and how long did it take? That's the closest thing you have to a real RTO.
  3. Where is the clean copy, and could an attacker with our admin password delete it? If the answer is yes, ransomware can wipe out both numbers.
  4. Who restores our systems during non-work hours? A target nobody is on the hook for doesn't mean much.
  5. Do our targets match what our cyber insurer expects? Carriers are asking harder questions at renewal, which we covered in our post on cyber insurance requirements.

With our fractional IT services, we help you set RTO and RPO targets for each system, build data backup and recovery to hit them, and test restores on a schedule, all under one flat monthly rate. If you want to know your real numbers before an outage tells you, let's talk.