A maintenance plan keeps the software current: updates, backups, uptime checks. A fractional retainer buys ownership of the outcome, including the failures that never produce an error message. They are different products, not the same product at two prices.
The most useful question I get asked is some version of “I already pay someone every month, so why is anything still broken?”
It is a fair question, and the answer is almost always that the thing being paid for and the thing being expected are two different products.
A maintenance plan keeps the software current: updates applied, backups taken, uptime watched, typically €35 to €230 a month. A retainer buys ownership of the outcome, which means one person whose job is that the system works, and who changes it when the business changes. They are not the same product at two prices, and almost every complaint I hear about maintenance plans comes from paying for the first while expecting the second.
Both are legitimate. They answer different questions. Here is how to tell which question you are actually asking.
The difference shows up on a bad week, not a good one
The difference shows up on a bad week, not a good one. On a good week both products look identical, because nothing needed doing and both invoices were the same as last month.
On a bad week you find out which one you bought. A plan answers “were the updates applied?” A retainer answers “why did enquiries stop arriving on Thursday, and it is fixed now.” The first is a checklist. The second is an outcome, and only one of them protects revenue.
Buying the wrong one is expensive in a quiet way. You are not overpaying, you are underinsured: paying real money every month for a service that was never designed to catch the thing that eventually costs you.
The two products, side by side
| What you are comparing | Maintenance plan | Fractional retainer |
|---|---|---|
| What it protects | The software | The outcome |
| Typical price | €35 to €230 a month, more for ecommerce | €1,200 upward, €2,500 and above for development time |
| Who does the work | Usually automated, reviewed in bulk | A named senior engineer |
| Scope of “a problem” | The site is down or a plugin failed | Anything between the customer and the outcome |
| Changes and improvements | Quoted separately, or not offered | Included time, worked in priority order |
| Who notices a silent failure | Nobody, until you do | Monitoring, then the engineer |
| What you get monthly | A report that updates ran | A report of what changed and what is next |
| Right when | The site is simple and stable | The systems matter commercially |
What a maintenance plan actually covers
I want to be fair to it, because a good maintenance plan is a real product and worth its price. It applies core, theme and plugin updates. It takes backups and, if it is any good, occasionally tests that one actually restores, which is rarer than it should be. It watches uptime and tells you when the site stops answering. It applies security patches, usually within a defined window.
That work is genuinely necessary. Skipping it is how sites end up compromised, and the market rate reflects a mostly automated service reviewed across many sites at once, which is why it can be sold at €90 a month and still make sense for the provider.
What it buys you is the absence of one specific category of disaster: the site that fell over because nobody updated anything for two years.
Where the gap opens
The gap is everything that breaks without breaking.
A plan watches whether the site responds. It does not watch whether the contact form still delivers.
Everything green, nothing arriving
A forms plugin started silently dropping submissions after an update. Every page loaded perfectly. Uptime was a clean 100%. The dashboard was entirely green, and the enquiries had simply stopped arriving. Nothing in a maintenance plan is built to catch that, because by every measure such a plan tracks, the site was healthy.
I keep meeting the same shape in different clothes. Scheduled jobs that never once ran, because WordPress fires its cron from page loads and a quiet site never triggers one (the whole story is here). A caching layer cheerfully reporting a successful purge while serving a stale page, because the routine it called had stopped existing. An image in the newest, smallest format making phones slower than the older format it replaced, because decoding it cost more than downloading it saved (measured over ten runs here). That last one took me the better part of a week to pin down, and I was looking for it.
None of those produce an error message. None appear on an uptime dashboard. All of them cost money for as long as they run, and every one was on a site somebody was paying to have maintained.
A maintenance plan tells you the site is up. It cannot tell you the site is working. Those are different questions, and only one of them is about your revenue.
What a retainer covers that a plan does not
Three things, and the third is the one people underestimate.
Someone owns the outcome. Not the tickets, the outcome. When enquiries stop, the question is not whether it falls inside a scope document. It gets fixed, and then it gets fixed properly so it cannot recur.
Change is included. Most of what a growing business needs is not repair, it is change: a new integration, a landing page, a performance push, a report that finally answers the question the business keeps asking. Under a plan each of those is a quote and a wait. Under a retainer they are a queue, worked in priority order.
The person already knows the system. This is the part that never appears on a comparison table and matters more than either of the others. When I have run a platform for a year, I do not need a discovery phase to change it. I know which part is fragile, which decision was deliberate, and which clever thing from two years ago everything now quietly depends on. Work that costs an outsider a week of orientation costs me an afternoon. You are not buying hours, you are buying the permanent absence of ramp-up.
What the market actually charges
Useful numbers, because the gap between these bands is where most of the confusion lives.
- Care plans, 2026: roughly €35 to €90 a month for a basic plan, €90 to €230 for a small-business care plan, and €230 to €650 where ecommerce or higher assurance is involved.
- Growth retainers sit separately, starting around €800 and rising.
- Agency retainers: roughly €1,500 to €4,000 a month for a single discipline, and €4,000 to €19,500 for full service. Both have risen more than 30% since 2023.
- Retainers below about €2,300 are widely observed to fund execution only, meaning delivery of agreed work rather than anyone thinking about your systems between tasks.
Notice the shape. There is a large empty space between a €230 care plan and a €4,000 agency retainer, and most established businesses sit squarely in it: too complex for a checklist, not big enough to fund an agency’s overhead. That space is what fractional engineering exists to fill.
Prices vary across Europe, and these are bands rather than quotes. The useful part is not the exact figure, it is the size of the gap.
When a maintenance plan is the right answer
Often, and I would rather say so plainly than pretend the answer is always the more expensive one.
A maintenance plan is the correct purchase when the site is genuinely simple, the content rarely changes, nothing commercially important depends on an integration, and a day of downtime would be irritating rather than expensive. A brochure site for an established local business is exactly this. I have talked people out of a retainer for precisely that reason: paying retainer money for a site like that is waste, and anyone telling you otherwise is selling.
It stops being the right answer at a specific point: when something on the system carries revenue, and nobody’s job is that it keeps working.
Ask who would notice
Ask one question, and answer it honestly.
If the contact form silently stopped delivering tomorrow, how long until someone noticed, and whose job would it be to notice?
If the answer is “we would spot it within a day, and it would be obvious”, a maintenance plan is fine. Keep your money.
If the answer is “honestly, we would find out when someone rang to ask why nobody had called them back”, you do not have a maintenance problem. You have an ownership problem, and no amount of update-applying fixes it.
That is the whole distinction. Everything else is pricing.
Before changing anything, ask your current provider for the last three months of reports. If they only show updates applied and uptime percentages, you now know exactly what you are buying, and exactly what you are not.