Kevin Surace explains how GENIE pricing works: you buy bundles covering 1 to 10,000 applications, annual audits verify what you actually use, and extra fees apply if you run tests on cloud GPUs. The model trades off simplicity against per-seat pricing.
So let's talk a little bit about the journey here into the world of GENIE and AI. I'm sitting here, I've got massive software QA effort underway across Thousands of applications, iPhone, 1800 advance. Say, Kevin, let's sort this mess out. How do I buy it? How does the licensing model work? What do I do? Am I buying? Is this super expensive or is this. I can get going relatively economically? What do I do first?
So, so, so the technology is priced in app bundles. The first bundle is 1 to 10, so you can use it on up to 10 applications and then after that it's like 50, and then it's 102, 50 and 500. 1,000 goes out to 10,000 applications. Right. You know, 10,000 applications is many, many millions of dollars. That's. That's not a surprise because there's so. And all the other things, you know, in the test. In the SQA world, people aren't used to pricing per application, they're more used to pricing. But there's really no test seats here. You know, we don't care how you know, you're going to have one or two or three people be the robot overlords, but you don't have 10,000 people on the system, right? So you have to really price it by application. And then there's an audit at the end of the year that makes sure that it was used on X number of applications, not more, et cetera. So, by the way, this is very common outside of software qa.
So I've got total freedom if I just go, I have got 10 apps and I want to generate a million test cases against those 10 apps. Just exaggerating for effect. I have at it, right? You've got genie, you go for gold. Doesn't matter if you've got a huge app, a little app, something in between. That's right.
There's some usage fees if it's in our cloud. Because if you say, I'm going to use those GPU servers a lot, there's a certain cost per hour of that and we charge that, right? So most people are going to generate their test scripts and probably not going to come back and regenerate them for months, because they've got them, they're done. These are regression tests, right? Regression tests don't change, you don't want them to change. So therefore they're not changing for months and sometimes years. So you're not tapping genie again to do that. They've been generated, they're good. If you need to tap it again a week, a month later. Fine. But there'll be some usage charges there. It's not a lot, but it can add up if it's in your cloud. Not our problem. You've got your own costs of those servers and you're paying for them one way or the other. Right, so there's some usage, but that's not a surprise. Again, outside of sqa, kind of. This is not a surprise. This is how the world charges for SaaS systems. It's how the world.
Absolutely. Yeah.
In the cyber security realm. Right, so. So there's no CIO who's going to be shocked that that's how they pay for this per app and some additional usage, especially for gpu. Nobody's shocked at that.
