Support and Maintenance · For tech businesses
Systems that stay boring at 2am.
A production system's finest quality is uneventfulness - and uneventfulness is manufactured. We keep platforms healthy as a standing discipline: monitored by people who know the codebase, patched and upgraded on schedule, and improved every month instead of merely kept alive.
Watched. Patched. Improving.
What it is
Operations as a discipline, not a scramble.
Every system decays by default: dependencies age into vulnerabilities, traffic grows past assumptions, and the small oddities nobody investigates become the outage everybody remembers. Teams feel it as a tax - the roadmap keeps losing sprints to firefighting.
Standing care flips the economics. Monitoring catches drift before users do, upgrades land on schedule while they're still small, and an engineering allocation each month makes the system better - faster, cheaper to run, easier to change - instead of just still-alive.
Humans behind the alerts
Monitoring wired to engineers who know your codebase - so an alert becomes a fix, not a forwarded email.
Decay handled on schedule
Dependencies, frameworks and infrastructure upgraded as routine - while upgrades are small, cheap and reversible.
Improvement, budgeted
Standing monthly capacity for performance, cost and reliability wins - care that compounds instead of care that preserves.
Who it's for
For systems that matter more than their staffing.
This is for businesses running production systems without a team dedicated to their health - the platform an agency built and left, the system whose builder now has one foot in new features and one in the pager, the acquisition nobody fully understands. If it matters when it breaks, it deserves better than incidental attention.
- Dependency upgrades are postponed until they're scary, then postponed again.
- Incidents are discovered by customers, then diagnosed by archaeology.
- The roadmap loses a sprint a quarter to unplanned production work.
- One person holds the system in their head, and they'd like a holiday.
How it works
Baseline, stabilise, watch, improve.
Baseline
Learn the system honestly
Architecture, dependencies, infrastructure and observability audited - the risk register and the baseline every improvement is measured from.
Stabilise
Pay the overdue debt
Outstanding patches, fragile deploys and missing monitoring fixed first - carefully, so the ground is solid before the routine starts.
Watch
SLAs with teeth
Monitoring, alerting and agreed response times by severity - ownership and the clock settled before the 2am that needs them.
Improve
The monthly better
Standing engineering spent where it compounds - performance, cost, reliability - prioritised with you and reported in numbers.
What you get
A production system that's someone's job.
An SLA that means it
Response and resolution commitments by severity, honoured by engineers who already know the system.
Observability that answers
Uptime, performance and error monitoring with dashboards and alerts wired to people - not to a channel nobody reads.
Scheduled currency
Dependencies, security patches and infrastructure kept current on a cadence, with backups tested rather than presumed.
A monthly engineering report
What happened, what improved, what it cost and what's next - the system's health in numbers a board can read.
The case for standing care
Downtime never invoices in advance.
The cost of care is a line item; the cost of its absence arrives as an incident - revenue down, trust dented, a quarter's roadmap consumed by the post-mortem. Standing maintenance costs a fraction of one serious outage, and unlike an outage, it pays back monthly: a system that gets faster, safer and cheaper to change while it runs.
Source: ITIC 2024 Hourly Cost of Downtime survey
Proof
Systems under our care.
The platforms in our case studies run under this discipline - built, then kept boring.
Broadreach Healthcare
Frontline healthcare teams, off paper
A digital tool for the Front-line Service Executives delivering care in underserved settings, plus the workflow system that keeps those teams scheduled, supported and accountable.
In their words
“Touchfoundry have been fantastic collaborators - they 'get' every concept we describe, and take it further with great energy and creativity. Even though we are in different hemispheres there is never a sense of distance - they are responsive and organised - and they do great work.”
Martin Jones
Director at Artinsite obo NHS
Common questions
Asked before almost every engagement.
Routinely - onboarding starts with an honest audit and a stabilisation pass, so we know the risk register before the pager is ours. Systems built by departed teams and quiet agencies are most of what arrives.
Production running on luck?
Let's put it under care.
Tell us what you run and what worries you at 2am. We'll audit it, hand you an honest risk register, and quote the standing care that would make the pager quiet.