Charge for website maintenance with a recurring monthly retainer tied to a defined scope, not an hourly rate billed after the fact. Price it on the value and risk you absorb — uptime, security patching, backups, and a guaranteed response time — then set the number high enough that a quiet month is profitable and a bad month is survivable. Maintenance is the most predictable revenue a freelance developer can build, and the easiest to underprice. The trap is simple: you quote a small number because the work "isn't much," then absorb every emergency, content request, and plugin conflict that arrives inside it.
Table of Contents
- What "maintenance" actually includes — and what it doesn't
- Which pricing model fits which client
- Building the number from the bottom up
- Writing the agreement so the scope holds
- Common ways freelancers lose money on retainers
- How to sell maintenance without a hard pitch
- Frequently Asked Questions
What "maintenance" actually includes — and what it doesn't
Before you can price anything, write down what you are selling. A maintenance plan is usually a bundle of recurring technical obligations: core, theme, and plugin or module updates; security monitoring; backups with a tested restore; uptime checks; broken-link and form checks; performance review; and a monthly report. On WordPress that means core and plugin updates on a staging copy first. On Drupal it means tracking security advisories and applying patches, which can mean real code work rather than a one-click update.
What maintenance does not include is new work. New pages, redesigns, new functionality, migrations, SEO campaigns, and ad management are projects. If you fold "small content edits" into the plan, cap them — a fixed number of hours or requests per month, with anything beyond that billed at your project rate. The line between the two is the single most important sentence in your agreement. Clients do not distinguish between "update the plugins" and "add a booking system" unless you do it for them in writing.
Which pricing model fits which client
Four models dominate, and they are not interchangeable. A reasonable default: offer tiers for small business sites, and a custom retainer for anything with custom code, e-commerce, or multiple environments. Those sites carry more risk, and risk is what you are charging for.
- **Monthly retainer, fixed scope.** The default for most freelancers. Predictable for both sides, and it rewards you for keeping a site stable rather than for hours spent fixing it.
- **Hourly, billed as used.** Honest but fragile. It punishes efficiency, invites invoice disputes, and gives you no income in a quiet month even though you were on call the whole time.
- **Prepaid hour blocks.** A client buys ten or twenty hours and draws them down. Good for clients with lumpy, unpredictable needs who refuse a subscription. Set an expiry so blocks don't haunt you two years later.
- **Tiered plans.** Three named tiers — something like Essential, Growth, and Priority — differing by update frequency, included support hours, response time, and whether performance or SEO reporting is included. Tiers let the client choose instead of negotiating your number down.
Building the number from the bottom up
Do not pick a price by guessing what sounds acceptable. Build it. Start with your target annual income plus business costs — software, hosting resale, insurance, accounting, pension — and divide by the billable hours you can realistically sell. Most full-time freelancers sell well under half their working hours; the rest goes to admin, sales, and unpaid learning. That gives an effective hourly cost floor. Then estimate the plan's monthly hours honestly, and multiply.
For a typical small brochure site, routine work might be an hour or two a month of actual hands-on time. But you must add three things on top: the cost of tools you supply (backup service, uptime monitor, security scanner, premium plugin licences), a buffer for the month something breaks, and a premium for the response-time promise itself. If you guarantee a four-hour response, you have sold availability, and availability has a price even when nothing happens. Finally, sanity-check against the client's downside. A site that takes orders all day loses real money per hour of downtime. A retainer that is a small fraction of one day's lost revenue is easy to justify and easy to keep.
Writing the agreement so the scope holds
The agreement does the work your goodwill otherwise does for free. Put these in it: Bill in advance and automate collection. Card or direct-debit subscriptions reduce late payment dramatically compared with manual invoices, and they make the relationship feel like a service rather than a favour.
- **Exactly what is included**, itemised, and a short list of what is explicitly excluded.
- **Response and resolution targets** — these are different. You can promise to respond in four business hours without promising to fix everything in four.
- **Support hours and channels.** If you don't name them, the channel becomes whatever the client prefers at 11pm.
- **Overage terms.** The rate for out-of-scope work and whether you need written approval before starting it.
- **Term, billing date, and notice period.** Monthly automatic billing in advance, with a 30-day cancellation notice, is common and avoids chasing payments for work already done.
Common ways freelancers lose money on retainers
Scope creep is the obvious one, but it usually arrives disguised as a small favour. The fix is not refusing — it is logging the request, telling the client it falls outside the plan, and quoting it. Do this the first time, every time, with the first client. The precedent is the product. Underpricing emergencies is the second. If a hacked site means a weekend of cleanup, restore, and hardening, that is incident recovery, not maintenance. Decide in advance whether your plan covers recovery or only prevention and backups, and say which.
"We take tested backups and will restore from them" is a fair promise. "We will fix any hack for free" is an unlimited liability. Third is supplying services at cost. If you pass through hosting, licences, or a CDN, build a management margin into the price, or you are doing procurement and support for nothing. Fourth is never raising prices. Review each retainer annually, give notice in writing, and tie the increase to something real — a site that has grown, more plugins, more traffic, a tighter response promise. Clients who have had three years of stable uptime rarely leave over a modest rise.
How to sell maintenance without a hard pitch
The best moment to sell a plan is at the end of a build, when the client can still see how much work the site took. Include the first month in the project proposal and present ongoing care as the normal continuation, not an upsell. A site handed over with no maintenance plan is a site that will degrade, and saying so plainly is accurate rather than pushy. For existing sites you did not build, lead with an audit.
A paid audit of updates, backups, security, and performance gives the client a concrete list, gives you an accurate picture of what you'd be taking on, and prices the retainer on evidence instead of optimism. It also protects you from inheriting a neglected site at a healthy-site price. Then make the value visible every month. A short report — updates applied, uptime, backups verified, issues caught, anything that needs a decision — is what stops a client asking what they are paying for. The month where nothing went wrong is the month you have to document, because that is the month the retainer bought.
Frequently Asked Questions
Should I offer a maintenance plan to a client who hosts elsewhere?
Yes, but inspect the host first. If you cannot create a staging copy, take backups, or get timely support, either require a migration or exclude anything you cannot control from the plan.
Do I need to charge extra for e-commerce sites?
Generally yes. Checkout, payment gateways, and tax or shipping plugins break in ways that cost the client money immediately, update testing is far more involved, and downtime is more urgent.
What if a client only wants maintenance when something breaks?
Sell prepaid hour blocks at a higher rate than your retainer equivalent, with no response-time guarantee. On-demand work is worth more per hour precisely because it interrupts planned work.
How do I handle an emergency outside support hours?
Define an out-of-hours rate in the agreement and require explicit approval before you start. Pricing after-hours work is far easier than refusing it mid-crisis.




