PRACTICE 01
Security.
Authentication fixed, secrets out of the source code, access narrowed to those who need it, dependencies patched on a schedule instead of after a scare. The shared-password era ends in the first sprint.
The tool your business runs on shouldn't depend on one person.
Production Hardening takes the software your company built, or had built, or inherited, to production-grade and keeps it there. Moonello becomes the engineering department behind your internal tools: security, monitoring, backups, and a number to call that always answers
Stabilization sprint firstFlat monthly retainerYour tools, our pagerAudit fees credit in
What happens when it breaks on a Tuesday?
For most companies running self-built software, the honest answer is a name, not a plan. One person gets interrupted, everyone else waits, and the business holds its breath until it's back.
It was never actually finished.
Internal tools go live the day they start working, not the day they're ready — no monitoring, no backups tested, no error handling for the weird order that shows up in month eight. At Moonello we'd tell you that's normal: "working" is a milestone, production grade is a discipline, and almost nobody budgets for the second one.
Every fix makes it more fragile.
Patches on patches, each one urgent, none of them documented. We've inherited systems where the fear of touching anything had quietly frozen the whole roadmap. The business wanted ten improvements and the honest answer was that nobody dared. That's not a code problem; that's the absence of a safety net.
The builder became the hostage.
The person who built it can't take real vacations, can't be promoted cleanly, and can't leave without the business inheriting a crisis. Our clients usually frame hardening as protecting the company, but the builder is often the most relieved person in the room. Same finding, other side of the table.
What does production hardening actually cover?
Hardening at Moonello is the difference between software that works and software that's operated. Six practices, running continuously. The same ones we run on our own products.
PRACTICE 02
Monitoring and alerting.
The system tells us it's struggling before your team feels it. When something fails at 9 a.m., or 2 a.m., the question changes from "how would we find out?" to "we already know, and here's the fix window."
PRACTICE 03
Backups and recovery.
Not "we think it backs up"; tested restores, documented recovery time, and a rehearsed answer to the worst Tuesday. Most systems we inherit have backups; almost none have ever restored one.
PRACTICE 04
Error handling and stability.
The weird order, the double-click, the file that's almost the right format, all handled gracefully instead of crashing the afternoon. Fragility usually lives in the edges, and the edges are where your people work.
PRACTICE 05
Documentation and continuity.
How it works, why it works that way, and what to do when it doesn't. All written down, kept current, owned by the company instead of a memory. This is how the builder gets their vacations back.
PRACTICE 06
Improvement hours.
Every retainer includes a fixed block of engineering time for the roadmap. The features that were frozen by fear start shipping again. Hardening isn't embalming the system; it's making it safe to change.
What having us feels like
A retainer is a relationship, so here's its shape and the rhythm our clients live with month to month.
It starts with a stabilization sprint.
Two to four weeks against the highest-risk items. From your audit roadmap if you have one, from our own assessment if you don't. The bleeding stops first: security holes, missing backups, the alert that should have existed all along. You'll notice the difference as silence.
Then the watch begins.
Monitoring runs around the clock, and a response commitment in writing says what happens when something trips — who's notified, how fast we act, what you'll hear from us. Your team goes back to their jobs; the pager is ours.
Every month, visible work.
Maintenance, patches, and your improvement hours shipped against a shared list you control, plus a monthly report in plain language: what happened, what we did, what's next. Our clients forward it to leadership unedited, which is exactly how we write it.
And the system gets easier to leave us with.
Documentation grows every month, decisions get recorded, and nothing we build locks you in. The measure of good hardening is that you could take it in-house and the reason clients stay is that they no longer want to.
From first call to steady state in about a month
Hardening at Moonello starts fast because it starts from evidence. A roadmap if you've been audited, a focused assessment if you haven't. Either way, the first month has a shape, and after it the rhythm takes over.
Baseline
If you arrive with an Internal Build Audit roadmap, we start from it directly. That's the fastest path, and it's why audit fees are credited. If not, we run a focused baseline of our own: enough assessment to rank the risks, not a full audit's depth. Either way, week one ends with a ranked sprint plan you've approved.
Stabilization sprint
The highest-risk items get fixed in priority order: the authentication hole, the untested backup, the missing alerts, the error that takes down the afternoon. This is the only phase that feels like a project with visible changes, quick wins, the bleeding stopped.
The watch turns on
Monitoring deployed, alert routes configured, the response commitment signed and who gets told, how fast we move, what you hear from us. The day this switches on is the day your builder's phone stops being the monitoring system.
The monthly rhythm
Maintenance and patches on schedule, improvement hours shipped against your list, and the monthly report in plain language. Cancel anytime with written notice. Retainers at Moonello are held by the work, not the contract.
One month in, the system is watched, backed up, documented, and improving. All while staying that way stops being anyone's side job.
What does production hardening cost?
Hardening at Moonello starts at $10,000 for the stabilization sprint and from $2,500 a month for the retainer.
Stabilization sprint — from $10,000
Two to four weeks
Two to four weeks against the ranked risks: security, backups, monitoring, the fixes that can't wait for a rhythm. Scoped from your audit roadmap or our baseline, priced flat before work begins. If you stop after the sprint, you stop safer than you started. Everything we build stays yours.
The retainer — from $2,500 a month
Ongoing · thirty days' notice, no lock-in
The ongoing engineering department: around-the-clock monitoring, a written response commitment, maintenance and patching, improvement hours against your roadmap, and the monthly report. Scales with what we're watching; a single tool sits near the floor; multi-system platforms with deeper response commitments run $8,000 to $12,000.
Audit fees credit toward the sprint; the sprint's baseline work carries into the retainer. Each step pays for the next one; the ladder is the discount.
For scale: a single mid-level engineer runs $130,000 a year before recruiting, ramp, and the risk of them leaving. And internal tools rarely need forty hours a week. They need the right hours, every week, from people who've seen this exact failure before. That's what a retainer is: the fraction of an engineering department your tools actually require.
Systems stay up when someone owns them
Case study: a decade of production, one operations platform
- The situation
SGS, a national telecom inspection company, runs its entire operation: nationwide crew scheduling, dispatch, job tracking, field reporting. All on a custom platform we built with them from the ground up.
- The stuck point
When the platform is down, hundreds of field jobs stall. There is no "we'll catch up tomorrow" in inspection work; reliability isn't a feature of the system, it's the business itself.
- The work
Moonello has operated the platform continuously since launch. Monitoring, maintenance, security, and a steady cadence of improvements shipped while the system stayed in production. The same rhythm this page describes is the one SGS has lived for years.
- The result
The company grew roughly tenfold on the platform, with 98% job report accuracy and dramatically faster project turnaround. Growth the system absorbed instead of obstructed. The builder-hostage problem never formed, because the engineering department was never one person.