WordPress maintenance proposal template
The proposals that close are the ones that show the client something they did not know about their own site. This structure leads with the audit, not with your service list.
Free and ungated. Built to be sent after a site audit, not before.
Service-list proposals lose to price
The typical maintenance proposal opens with a description of the agency, follows with a list of included services, and closes with a monthly price. It reads like every competing proposal, because it is structured like every competing proposal.
When two documents look the same, the only visible difference is the number at the bottom. That is how maintenance becomes a price competition, which is a competition nobody wins.
The alternative is to lead with something only you can provide: specific findings about their specific site. A proposal that opens with 'your backups have never been tested and three plugins have published vulnerabilities' is not comparable to a generic service list, and it does not get shopped on price in the same way.
What the audit should surface
- Plugins and themes with published vulnerabilities, with dates
- WordPress core and PHP versions against current supported releases
- Backup presence, frequency, storage location and whether one has ever been restored
- SSL certificate status and expiry
- Page speed and Core Web Vitals on the commercially important pages
- Administrator accounts, including ones belonging to people who have left
- Uptime over the past period if you can measure it
Six sections, in this order
The order matters more than the content. Risk first, plan second, price last.
What we found
The audit, stated factually and without alarm. Specific findings with dates and versions. No recommendations yet, and no fear-selling. The findings should be uncomfortable enough on their own that exaggeration would weaken them.
What that means for you
Translate each finding into a business consequence in plain language. An unpatched vulnerability is not a technical fact, it is a risk of the site being defaced during their busiest week. A never-tested backup is not an oversight, it is an unknown recovery position.
What we would do about it
The remediation work, separated clearly into one-off fixes and ongoing maintenance. Keeping these separate is important: the client should understand they are buying a repair and a prevention plan, not one bundled number.
The care plan
Your recommended tier and precisely what it includes: hours, response targets, backup frequency, reporting cadence, and the exclusions. Reference the care plan template rather than paraphrasing it loosely.
What it costs
One-off remediation and monthly plan, stated separately and plainly. No ranges, no 'from', no options requiring the client to design their own package. Give a recommendation and a price.
What happens next
One specific action with a date. 'Reply to confirm and we start Monday' closes considerably better than 'let us know your thoughts', which is an invitation to do nothing.
What makes proposals close
Audit before proposing
A proposal written without looking at the site is a service list. Twenty minutes of inspection changes the entire document.
Consequence, not jargon
Every technical finding needs a sentence explaining what it means for their business. Assume the reader is smart and non-technical.
Do not fear-sell
Overstating risk destroys credibility and invites a second opinion. Accurate findings are alarming enough without embellishment.
Separate one-off from ongoing
Bundling remediation into the monthly fee hides the value of the repair and inflates the recurring number.
Recommend one tier
Presenting three options and asking the client to choose transfers your job to them. Recommend one and mention the others exist.
Close with a date
A specific next step with a date converts far better than an open-ended invitation to respond.
Pricing the remediation
The one-off remediation is where agencies most often undercharge, usually because it feels like a barrier to the recurring revenue they actually want. Discounting it to win the plan sets an anchor that makes every future project quote harder.
Price the remediation at its real cost and treat it as a separate decision. If a client will not pay to fix known problems on their own site, they are unlikely to be a good ongoing client, and finding that out before you sign a twelve-month plan is a favour to yourself.
The exception worth making is a small goodwill fix during the audit itself. Patching one critical vulnerability before you send the proposal demonstrates competence better than any paragraph describing it, and it costs you ten minutes.
Proposal mistakes that cost the deal
- Opening with agency background instead of their site
- A generic service list with no audit findings
- Bundling one-off repair into the monthly fee
- Three options and no recommendation
- Closing with 'let us know your thoughts'
Questions about maintenance proposals
A short one, yes. Twenty to thirty minutes of inspection costs you little and transforms the proposal from a service list into a document about their site. A deep audit taking several hours should be paid, and can often be credited against the first invoice.
Enough to be specific and credible, not so much that it becomes the deliverable. Five to eight concrete findings with dates and versions is persuasive. Thirty findings is a report they will use to brief a cheaper agency.
Recommend one and mention the others exist. Presenting three and asking the client to choose transfers the decision to someone with less information than you, which usually results in the cheapest option or no decision at all.
Do them, do them well, and put a reminder in your calendar for three months. Clients who decline a plan often accept one after the second unexpected problem, and having fixed the first one properly is the strongest possible argument.
Generate audits and proposals from real data
Run a portfolio scan, turn the findings into a branded pre-sales audit, and send a proposal built from what the scan actually found.
Free during early access. No credit card required.