Free 20-point conversion & SEO audit for export stores — Free website audit — claim yours

hello@woo.one Mon–Fri · 9:00–18:00 (UTC+8)

What a Website Maintenance Contract Should Actually Cover

The gaps between what clients assume and what contracts say: staging, hours, response times, abandoned plugins — and the four scenarios any plan should answer.

Maintenance contracts are usually sold as “updates and backups” and bought without reading what is excluded. The failures that matter — an exploited plugin, a restore nobody tested, a checkout that broke on a Friday — tend to sit in the gaps between what the client assumed and what the contract says.

This is what a contract should specify, and where the common gaps are.

What must be in scope

Core, theme and plugin updates, applied through staging where anything non-trivial is involved. Not “notify the client that updates are available”.

Backups, with the frequency, retention and offsite requirement stated as numbers. Plus restore testing, because an untested backup is a hypothesis — see the recovery notes.

Security monitoring, including file integrity and login monitoring, with a defined response when something triggers.

Uptime and error monitoring, with someone actually alerted. Monitoring that emails an address nobody reads is decoration.

A defined response time for site-down issues, distinct from the response time for everything else. “Within 24 hours” is a fine SLA for a content change and a useless one for a checkout outage.

A monthly report: what was updated, what broke and was fixed, backup status, uptime, notable security events. This is what distinguishes a service from a retainer.

Where the gaps hide

“Updates included” without staging. An update applied directly to production is not maintenance; it is a change with no rollback. Ask what happens when an update breaks the checkout.

Hours of development excluded. Most contracts cover maintenance but not the fix, and the fix is where the hours are. A clearly stated monthly allowance, with an hourly rate beyond it, is fairer than an ambiguous boundary.

Third-party services excluded. Hosting, CDN, payment gateway, email delivery. When email stops delivering, the maintenance provider is often the first person blamed and the last with access.

Content and new features excluded but undefined. “A small text change” is different from “a new landing page”. Define the threshold — for example, up to 30 minutes of work per request.

Abandoned plugins. If a plugin stops being maintained, replacing it is a project, not maintenance. Say so in advance, rather than discovering it during a security incident.

Response times, written as commitments

Severity Example Target
Critical Site down, checkout failing, data loss Acknowledge in 1 hour, work started immediately
High Major feature broken, security alert Acknowledge same business day
Normal Content change, minor display issue 2–3 business days

“Acknowledge” and “resolve” are different commitments and should be priced differently. Promising resolution times for unknown causes is how providers end up either losing money or breaking promises.

Pricing models and their incentives

Flat monthly fee — predictable for the client, and the provider is incentivised to minimise work. Works when the scope is genuinely bounded and the site is stable.

Retainer with an hours pool — flexible, and the client sees what is consumed. Risk: hours get used on requests rather than on preventative work.

Hourly — no commitment, no prevention. Sites on hourly maintenance get fixed after they break, which is the expensive way to run a store.

Percentage of build cost — common in agencies, roughly 15–20% annually. Reasonable for complex sites, opaque for simple ones.

The incentive problem worth naming: preventative work is invisible when it succeeds. A provider who spends the month hardening and testing looks like they did nothing, while one who fixed three visible bugs looks productive. Contracts with a stated preventative component — audits, restore tests, update rehearsal — protect both sides from that.

What the client should be able to ask

A good contract makes these answerable:

  • When was the last successful restore test, and how long did it take?
  • Which plugins are no longer maintained, and what is the plan?
  • What is the current uptime, and what were the incidents last quarter?
  • Who has admin access, and when was it last reviewed?
  • If the site is compromised on a Saturday, who is called and how?

If a provider cannot answer the first two on request, the contract is describing activity rather than outcomes.

What is not maintenance

Worth stating explicitly, in both directions:

  • New features, redesigns and migrations are projects
  • Content writing and SEO campaigns are separate services
  • Fixing something a third party broke is chargeable work, however unfair that feels
  • Supporting a customer’s own bad decision — installing an unvetted plugin, editing theme files — is not covered, and should be said kindly but clearly

The last item causes the most friction. A client who adds a plugin over the weekend and breaks the site on Monday expects it fixed as part of the fee. A contract that addresses this in advance preserves the relationship; one that does not turns it into an argument about goodwill.

The honest version

Maintenance is not glamorous and it does not demo well. What it buys is the absence of certain bad days: no untested restore, no abandoned plugin running on a checkout, no security incident discovered by a customer.

If you are evaluating what to include, the test is simple: ask what happens in the four scenarios that actually occur — an update breaks checkout, the site is compromised, the host loses data, a plugin is abandoned. If the answer for any of them is “we would figure it out”, the contract has a gap. That is the framing we use for maintenance plans.

S

Infrastructure and security engineer at woo.one, based in Foshan, China. Works on WordPress, WooCommerce and OpenCart systems, self-hosted mail, and the operational side of running stores for exporters.

More about the team

No pressure · No sales script

Ready for a store that sells while you sleep?

Send us your current site or product list. You will receive a fixed-scope proposal, a 20-point audit and a realistic timeline within two business days.

Reply within 1 business day NDA on request 100% code ownership
Chat with us
WhatsApp Get a quote