Okay, after reading all the answers and comments here, I would like to add a few additional points to my answer.
First off, as freelancers we normally trade our time for money. So it seems natural to base the price of a monthly service off of how much time it takes us to do it. And that might be fine if you're just getting started, but like Joe (OP) said, how should you charge for something that took you time to setup, but doesn't take any time to continue running, like a cron job?
I think the answer (as hinted at in the comments of some of the other answers), is to look at the value you're providing. Or another way of looking at it: what does the client miss out on if they don't buy your service?
- Their site could go down for 24 hours? A week? Forever?
- Total data loss/have to start over from scratch
- Missing out on sales and leads
- Taking time out of their busy schedules to futz around with WordPress
There are plenty more, that's just scratching the surface.
What's not on the list? "Joe must be at his keyboard at least x hours per month"
They could care less about the mechanism of how the benefits are delivered, all that matters is whether they get the benefits.
Okay so what should you charge? That depends on two factors:
- Your absolute minimum should be based on how much time you think it will take you per month.
- Beyond that, think about how much value your clients are getting from their website in the first place, and what kind of price they would tolerate. For example, a local quilting club website probably wouldn't pay much, while a t-shirt printing company that primarily sells product through their website would pay quite a bit to make sure their business remained open and profitable.
What I charge and offer in my maintenance service:
$150/month for the following:
- Smart updates...the whole backup, test, launch (or restore) routine
- Malware monitoring, protection, and cleanup
- Uptime monitoring
- Daily offsite backups
I chose $150/month because I figured maybe once per year each client would need some extra attention to resolve a plugin conflict, crash, or something similar. So the months where everything went smoothly would help subsidize the one problem per year, and they don't get stuck with a $1000 repair bill.
The last 3 components of my service are mostly automated, but help to round it out and make it look more enticing.