Helper CTO Series 01 | How Much Does a System Cost to Maintain Every Month: Three Pricing Structures Explained

"The system's done — so how much do we pay every month from here?" That's the single number the owner wants. But two websites can both be "just a website" — one with backups, documentation, and a deployment process that can be repeated, the other with none of that — and the monthly effort required differs by several times over. Quote a single number and you have nothing to compare it against. It's not that you're unwilling to pay — you don't know what the money is actually for, or what the other side is responsible for once you've paid it. Here are three things that break that monthly fee down for you: three pricing structures, what a monthly fee includes and doesn't include, and four places to check before you sign.

Why "how much per month" is hard to answer directly

Maintenance fees have a structure behind them, just like development quotes do. Why the same system build gets wildly different quotes from different vendors covers development; the monthly fee after launch comes down to three variables instead:

• The state of the system: whether the code is readable, whether backups and documentation exist. The worse the condition, the more time it takes just to keep it from breaking.
• The scope of responsibility: just "someone fixes it when it's down," or "someone proactively updates it every month and handles small changes along the way."
• How long you can wait: needing a response the same day versus next week calls for a completely different staffing setup.

A vendor who quotes a monthly fee without ever looking at the system is usually planning to figure it out after they've been paid, or quoting a number so conservative you won't be able to afford it. Neither works out in your favor.

Three pricing structures: pay-per-incident, flat monthly fee, monthly fee plus development capacity

Maintenance pricing on the market generally falls into three types. Check who each one suits before you look at the numbers.

• The risk with pay-per-incident is that nobody's watching in between, so problems are usually spotted by your customers first — and every call means re-explaining the system from scratch.
• A flat monthly fee is the most reasonable starting point for most small and mid-sized businesses: someone is watching consistently, and you don't have to negotiate price before something gets fixed.
• Monthly fee plus development capacity isn't about spending more every month — it's about not having to re-quote and re-queue every single time something needs changing.

What a monthly fee usually includes, and usually doesn't

The word "maintenance" covers wildly different scopes. Go through the following items one by one — check off what's included, and ask exactly what's extra for anything that isn't. Usually included:

• Someone watching the server, handling anomalies before you notice them.
• Debugging when the system throws an error: finding the cause, fixing it, and confirming it doesn't happen again.
• Small adjustments: changing some text, adding a field, reordering a process.
• Regular security updates: vulnerabilities keep being found in the underlying packages a system depends on, and someone needs to patch them on a schedule.

Usually not included, or needs to be negotiated separately:

• New feature modules, such as a report that didn't exist before, or connecting a new payment gateway.
• A full site redesign or a change in the underlying technical architecture.
• Cleanup on first handover, getting the system to a state where it can run properly — quoted separately based on condition.
• The server's own rental cost, which is a separate line item from labor — don't compare the two as if they were one.

If a plan's entire description is just the words "system maintenance" plus a price, that plan itself is the risk.

Four places to check in the contract first

Once price is settled, what matters isn't the total amount — it's four places that often get skipped over:

1. Response time: how quickly a server issue gets handled, and how quickly a bug you report gets an assessment back — and what counts as a response. "We've received it" doesn't count; "here's the next step" does.
2. Update frequency: security updates shouldn't happen only when there's spare time. If the contract doesn't state a frequency, there's no real commitment.
3. Exit terms: can you stop, how much notice is required, and who owns the code and data once you do. A lock-in contract usually protects the vendor.
4. Document ownership: whether the deployment process, account lists, and change logs get handed over to you when it ends. Without this clause, switching vendors means starting an excavation from scratch.

These four items determine whether the other party is obligated to show up exactly when you need them most. Why one-off outsourcing so often fails comes down in large part to none of this being discussed at signing.

How Nerdtechnic charges for this

• Basic maintenance from NT$6,000/month: watching the server, handling issues before they escalate, debugging errors, small adjustments, regular security updates.
• Billed monthly, and you can stop anytime — no lock-in contract quietly folding first-time cleanup costs into your monthly rate.
• Commitments written into the contract: a response within 1 business day when the server or a service goes down, an assessment within 2 business days when you report a bug, security updates at least once a quarter.
• When the engagement ends, the documentation and records are yours — switching to someone else doesn't turn into a black box.

The word "from" means exactly this: we don't quote without looking at the system first — only a number based on what we've actually seen means anything to you.

What happens every month after a system is built is our core business. Nerdtechnic takes over and maintains your system as Helper CTO — like an in-house tech lead, just not on your payroll. The first step is a 60-minute system health check: we only need to see the code, no credentials to your production environment required, and within 3–5 business days you'll get a one-page report you can actually understand. If you want to know what your system should reasonably cost each month, and what responsibility that money buys, start here: Helper CTO: System Maintenance Plan.