I’m thinking through the topic of ROI for infrastructure software again (obviously for our PaaS stuff). I get asked about this every year or two. Each time I look at it, I get more confused. This ask always come from sales, presumably because the buyers are asking it. But, I don’t think the ask is about the basic ROI calculations: the amount we paid for this tech is less than the amount of money we make using it.

First, that is a very difficult thing to figure out with infrastructure software. What is the ROI of Linux (I guess I mean RHEL), Java, or a laptop?

In each of those cases, the ROI could be infinite: if you didn’t have them, your business would not function.

In other cases true ROI is zero: there is no direct cause between having a laptop and increasing margins on USB cables by 0.5% because your supply chain negotiator could type on a laptop, maybe even accessing the intranet for contract templates with a VPN while they were waiting for their flight to board. But, they weren’t going to do it with pen and paper, so you needed that laptop.

My Java example has a twist. If you’re running your business with apps you create, the ROI of a programming language is infinite, per above: if you have no programming language, you have no app, you have no business.

The question is more (1) why Java? and (2) how much, if any should I pay for Java? But this is not an ROI ask, it’s (1) a competitive feature analysis (“we could do this in two lines of Rust!” the developers say), and, (2) am I getting “a good buy on it.”

Then there’s the question of where the money (the “return”) comes from.

Most of the ROI analysis I see (lots of Forrester TEIs) get the return based on “productivity,” you need less staff.

This is also weird. How do you get value out of needing less staff? You either (1) fire them (“layoff,” if you prefer), or, (2) repurpose them to do more work with their free time: now we can ship three features a month instead of two, meaning we can improve existing ways and add new ways of…doing whatever our business does.

Firing people is straight forward, if inherently anti-chill - see labor’s AI freakout!

The second depends on the business (1) actually doing something additional/new, and, (2) those business ideas being good and profitable

Anyhow: when someone (sell or buy side) is asking for ROI, what are they asking for?

(My theory is that they’re more asking: can this technology do the job I need, and am I paying the cheapest price possible for it? I mean, that’s classic competitive advantage in effect. But it also means that ROI conversations are actually about feature matrixes, PoC’ing, confidence that a technology will solve your problems, and then horse-trading on price. There’s no “return” as such, there’s just “it can do the job and I got a good deal.")