Nothing a DBA does makes money

That has been the problem for as long as there have been database administrators, and I don’t think we’ve ever really solved it.

Here’s what I mean. A DBA doing the job properly produces nothing you can see. The environment is up. The queries return. Nobody rings anybody at two in the morning. And because nothing appears to be happening, the people looking after it start to look like a line of cost with nothing attached to it.

In the traditional sense of how businesses think about this, they’re right. Nothing a DBA does makes money. What a DBA does is save money, which is a much harder thing to put in front of a board.

What happens when the maths wins

I worked as a full-time employee at a large organisation years ago, in a database team that ran to fifteen or twenty people across several database management system platforms. Depending on where you are and what you’re paying, a team that size is comfortably north of two million dollars a year in salary packages.

On paper, cutting it looks like the easiest saving available. Nobody in the leadership team can point to revenue the team generated, because there isn’t any. So the team gets cut.

And here’s the part people always miss: the environment doesn’t fall over the next morning. That would almost be better, because at least the cause and effect would be obvious. Instead it carries on. It just carries on with less attention than it needs.

Then performance slips. Then incidents start stacking up. Then it bubbles back up to the same management team, who now want to know why customers are unhappy and why it’s costing so much to put right.

The risk was never removed from the business. It just moved somewhere nobody was looking.

The disconnect is human, not technical

I’ve stopped being frustrated by this, because it’s a completely normal way for people to think. We don’t see it, we don’t hear about it, we don’t understand what these people do all day. So why are we paying for it when all it does is cost money?

It’s the same reasoning that leads someone to stop maintaining a website until the day a decade-old vulnerability gets published and a few hundred thousand sites get compromised in a week. The work was always doing something. You only find out what when it stops.

Where managed services actually fits

I’m not going to pretend this is an argument that everyone should have a bigger DBA team. Most organisations genuinely can’t justify one, and for a lot of them a full-time DBA would be underemployed.

What I’d say instead is that the work still has to happen, and the question is only who does it and what it costs you.

A managed service is one answer. You get the monitoring, the maintenance and the response capability without the salary, the recruitment, or the problem of what happens when your one DBA goes on leave. If you already have a team, it takes the repetitive work off them so they can do the project work that actually gets noticed.

What you’re really buying is that the risk sits with someone whose entire job is watching for it, rather than sitting quietly in a system nobody’s looking at.

That, at least, is something you can point to.

—

Warwick Rudd is a Microsoft Certified Master and Data Platform MVP, and the founder of SQL Masters Consulting. If you’d like to talk about what this looks like for your environment, have a look at our Database Managed Services page or get in touch.