Your software is live, customers or staff depend on it, and the team that built it has moved on. Maybe it was an agency, a freelancer or an internal team that no longer exists. Software maintenance services are how you keep that system secure, working and up to date without rebuilding it. This guide explains what maintenance covers, what a good agreement includes, how to budget for it, and how to hand maintenance over to a new partner without losing control.
Five things to take away
- Software that is not maintained does not stay the same. It slowly becomes less secure and harder to change.
- Maintenance is more than fixing bugs. It also means keeping up with platforms, preventing failures and making small improvements.
- A good agreement defines scope, monitoring, backups, update cadence and priorities agreed with you.
- A backup only counts once someone has restored it.
- If you are changing vendor, start with an independent review of the code before anything is changed.
Running software nobody is looking after?
Tell us what you have in production and we will suggest how to organise its maintenance.
Why software maintenance services matter after launch
Launch day is when the software starts to age. The code does not change on its own, but everything it depends on does.
- Dependencies age. Libraries, frameworks and runtimes release new versions and stop fixing the old ones.
- The world around it moves. Browsers, operating systems and third-party APIs change. An integration can break because another company changed something on its side.
- Security fixes keep coming. New vulnerabilities are found in widely used code, and someone has to apply the patches.
- Data grows. Queries that were fast with a few thousand records can crawl with millions.
In regulated sectors such as healthcare or finance, the stakes are higher, because downtime or a data leak affects patients or customers directly. The work is the same, but it has to be done with more care and better records.
The four types of maintenance, in plain words
- Corrective: fixing bugs that users or monitoring have found.
- Adaptive: keeping up with platforms and runtimes. When a PHP version reaches end of life, for example, the code has to move to a version that still gets security fixes. Our guide to PHP version support lists the dates.
- Preventive: the work done before something breaks, such as updates, refactoring and monitoring that warns early.
- Perfective: small improvements that make the system better to use, like a faster page or a clearer report.
Many contracts only pay for the first type. The other three are what stop the same problems coming back.
What a software maintenance agreement should include
When the partner is in another country, a clear agreement matters even more. Make sure it covers:
- Scope. Which applications, environments and integrations are included, and what is not.
- Monitoring and alerts. What is watched, and who is told when something fails.
- Backups. How often they run, where they are stored, and when a restore was last tested.
- Update cadence. A regular rhythm for dependency, runtime and server updates, and how changes are tested before production.
- Priorities. What counts as critical and what can wait, defined together so both sides know what happens when something goes wrong.
- Reporting and contact. What you receive each month, who your named contacts are, and how you reach them.
Budgeting: a fixed monthly frame or ad hoc
You can buy maintenance ad hoc, paying for hours when something breaks. It looks cheap until it is not. Bills are unpredictable, and preventive work rarely happens because nobody has asked for it.
The alternative is a fixed monthly frame with an agreed number of hours. It gives you a predictable budget and room for updates and small improvements every month. Keep a shared backlog so the hours go where they matter most, and review the frame each quarter. If the hours are always used up by new features, that work may belong in a separate project.
Warning signs that maintenance has slipped
- Dependencies have not been updated in a long time, and nobody can say which versions are running.
- Only one person knows how to deploy, or nobody does any more.
- Backups exist, but a restore has never been tested.
- The same incidents come back every few weeks.
- The system runs on a runtime or database version that no longer receives security fixes.
Two or three of these together usually mean the system is carrying more risk than anyone has noticed.
Taking over from a previous vendor
Taking over a system someone else built is common, and it is manageable if the handover is done properly. Start with a code audit, so you know what you have before anyone changes it. Then collect:
- Access to every repository, with full history.
- Credentials moved into a password vault you control, not left in emails or spreadsheets.
- Hosting, domains, DNS and third-party accounts registered to your company, not to the old vendor.
- Documentation of setup, deployment and integrations, however incomplete.
Our code audit services can do that review independently, whoever ends up maintaining the system.
How we run software maintenance services
Our software maintenance service begins with a short review of the system and an agreement on scope and priorities. After that we work in a steady monthly rhythm: updates, monitoring and a shared backlog, with developers from our offices in Denmark, London, Islamabad and Cairo as the technology requires.
Every month you get a short report. Ask for the same from any partner:
- Updates applied.
- Incidents, and how they were resolved.
- Risks we see coming, such as a runtime approaching end of life.
- What is planned for next month.
If the work grows beyond maintenance, a dedicated development team can take on larger changes while ongoing maintenance continues as before.
Want a steady hand on your production software?
See what our software maintenance covers and how we get started.
Frequently asked questions
It depends on the size of the system, the number of integrations and how old the technology is. We review the system first and then propose a monthly frame that fits.
Yes. We start with a review of the code and operations so we understand the system before changing it.
Support helps users with questions and problems. Maintenance is work on the system itself: fixes, updates, monitoring and improvements.
Yes. Working today does not mean secure tomorrow. Dependencies, runtimes and the platforms around the system keep changing, and unapplied security fixes add up quietly.
What we would do
List every system you run and who maintains it today. Check which runtimes and dependencies are in use, and when a backup was last restored. If the original vendor is gone, secure your access and credentials first. Then agree a fixed monthly frame with clear priorities and a short monthly report. That turns maintenance into routine work instead of emergency calls.
Ready to put maintenance on a steady footing?
Book a short call and we will look at your system and agree a scope.


