Solution · For SaaS companies
Keep the uptime promise your customers pay for.
A SaaS company sells availability. When the service is down, every customer knows at once, the status page lights up, and the sales team hears about it before the engineers do. The uptime promise is the product — and it needs numbers behind it.
Uptime promises translated into downtime budgets anyone can check.
Customer-facing incidents prioritised above all else.
Outages priced, reviewed, and prevented from repeating.
The problem
Why SaaS outages cost more than downtime
SaaS outages are public by nature. The technical failure is only half the damage — the other half is the trust lost while customers wait for news, and the renewals that quietly do not happen.
Every customer notices at once
An internal outage affects one office; a SaaS outage affects every customer, at the same moment, in public. The support queue fills in minutes, and the team cannot answer everyone at once.
The uptime promise is vague
Marketing promises reliability, sales promises more, and the agreement says something else entirely. When the service fails, the gap between the promises becomes the argument.
Incidents ranked by noise
The customer who shouts loudest gets the fix first, while a silent failure affecting hundreds waits. Without a priority matrix, the queue reflects volume of voice, not volume of impact.
Post-mortems that change nothing
The outage is reviewed, the document is written, and the actions are never assigned. Six months later the same failure returns, and the post-mortem reads like a prediction.
The process
Five steps, and the itserva page for each
Five steps, from the promise to the prevention.
- 01
Turn uptime promises into numbers
Convert each uptime percentage into allowed downtime per week, month and year. The number is the budget the engineers can plan against — and the promise the customers can actually check.
- 02
Price an outage
Work out what an hour of downtime costs in lost revenue, wages and recovery. The number sets the priority of the fix, the size of the investment, and the honesty of the apology.
- 03
Prioritise customer-facing incidents
Score urgency against impact with the customer at the centre: an outage affecting all customers outranks one affecting one, whoever reported it. The matrix is what keeps the queue honest under pressure.
- 04
Compare ITSM platforms
Compare the platforms that would run this work — on price, deployment and fit — with each fact sourced and dated. The platform decision is a multi-year one, and it deserves a comparison, not a demo.
- 05
Review and prevent repeats
After every serious incident, record the timeline, the lessons and the actions — with owners and dates. A review without owners is a document; a review with owners is a fix.
What’s included
Tools, templates and guides for saas companies
Tool
SLA Uptime Calculator
Turn an uptime percentage into allowed downtime per week, month and year.
Comparison
ServiceNow Alternatives
Alternatives to ServiceNow, compared on price, deployment and fit.
Product
SLA Tracker
Track uptime against target for every service, and see the downtime budget that is left.
Measure it
The four numbers worth tracking
Four numbers belong in every SaaS service review — and three of them belong on the status page conversation with customers, too.
Uptime against SLA
Actual availability next to the promised target, in minutes, not percentages. The gap is the number customers understand — and the number the team is judged on.
Downtime cost
What each outage actually cost, in revenue and recovery. It is the number that decides how much the next prevention investment is worth.
MTTR
How fast the team restores service when it fails. For a SaaS product, recovery speed is the product experience during an outage — and customers remember it.
Repeat incidents
Incidents with a cause the team has seen before. A repeat in SaaS is a renewal risk, because the customer has now watched the same failure twice.
Frequently asked questions
What uptime should a SaaS product promise?
What the team can actually deliver, with margin. A promise of 99.9% allows about 43 minutes of downtime a month; promising more than the infrastructure can deliver turns every outage into a breach.
How do we prioritise incidents across customers?
By impact: how many customers are affected, and how badly. The priority matrix turns that into an order the whole company can see, so the loudest customer does not automatically win.
Should we publish a status page?
Yes. Customers can tolerate an outage far better than silence, and a status page answers them all at once. Update it every half hour during an incident, even if the news is only that work continues.
What makes a good post-mortem?
A timeline, a cause found with evidence, and actions with owners and dates. Blame has no place in it; the question is what changes, who does it, and when it is checked.
Put a number behind the uptime promise.
Run the uptime calculator on the target you promise customers, then price an hour of downtime. A free workspace saves both, so the next service review starts from the real numbers.