SmartBear + Sapiens Decision Making Sure Every Decision is the Right Decision

Model for Excellence

Back in the early 2000s, Larry Goldberg, the founder of Sapiens Decision, worked with Barbara van Holle to create the “decision model” – a way to abstract business logic to a point where it had meaningful impact. They built technology around this model. The key to Sapiens Decision is that it enables business analysts to manage the logic in their technology. “That’s the key driver of our business,” says Goldberg.

The Decision Suite

Gil Segal, Director of Implementation and Training, explains how this works in practice. Within a suite – called the Decision Suite – there are two primary modules. A user, typically a business analyst, uses Decision Manager to model their business logic, validate it to see it’s structured properly, and test it to make sure it does what it’s supposed to do.

Once that’s done, it’s deployed to another module called Decision Execution. That’s a hot deployment – once something is deployed to Decision Execution, it’s available instantly. How does Sapiens Decision ensure that the deployed models execute at scale? That’s where the ReadyAPI toolset comes in.

Automating With ReadyAPI

Segal explains, “We use ReadyAPI to help us with actually being able to test, to see that the performance we deliver is as expected. As we go from version to version to ensure that we’re hitting those goalposts, we’re also using the tool set to identify potential improvements. So we use the tool set to identify where, in our solution, we have bottlenecks, and where we can make improvements. Then we measure those improvements to ensure we’re actually delivering.”

Many times, when people think about adding automation, they want to reduce the time and costs that come with manual. A better measurement for Sapiens Decision was the alignment-to-performance thinking throughout the process, and being able to measure, at different points in their process, what they’re doing so they can be certain.

Scale Is King

The cloud has had a huge impact on the ability to scale, particularly in financial services. When customers migrate to the cloud and realize the scalability that’s now available to them, it’s critical for their solutions to scale right alongside them. Once you’re at scale, every single microsecond becomes a factor. ReadyAPI allows Sapiens Decision to meet those scalability needs.

Cloud Migration and Performance

Cloud migration makes every moment at every point in the cycle important. This becomes even more significant at scale, for clients who run literally billions of decisions a day – possibly even trillions. This puts pressure on Sapiens Decision, which, according to Goldberg, “Needs to be far more precise in our measurements, far more exact in our performance capabilities.”

ReadyAPI allows Sapiens Decision to get a handle on where the performance bottlenecks are and where the improvements can be made, which allows them to communicate this to the client. Goldberg says, “I’ve been extremely happy because we’ve been able to expose this information to our clients to a degree we never dreamed of being able to do before.”

ReadyAPI and Complex Systems

Anyone in financial services or any regulated environment is familiar with a high level of complexity. ReadyAPI delivers great results here.

There are several dimensions involved. One dimension is the complexity of the logic itself. Another is the volume of the data that needs to be shipped into the web services and out. ReadyAPI allows Sapiens Decision to measure the impact of that data, and what kind of performance they can expect based on the data size.

But it’s also about the hardware setup – the architectural typology. Segal points out, “If you’re dealing with billions of decisions, how many servers do you need? What kind of power, what hardware parameters do you need to specify for those servers? How many CPUs, how much memory?” ReadyAPI allows Sapiens Decision to measure all of those dimensions, and come up with metrics to help their customers be predictive.

Goldberg agrees that ReadyAPI has become an essential part of their development environment. “Many clients run at different levels of performance at different points in the day. Planning that and planning the cloud architecture is very important. Now that we have ReadyAPI, we’re able to do all these exacting tests to come up with these metrics before we deliver the product, as we issue new releases and new technology.”

How SoapUI Helped, and How ReadyAPI Improved

Before Sapiens Decision used ReadyAPI, they used the open source version of SoapUI, also part of the SmartBear portfolio. SoapUI helped them get to where they wanted to go. Segal explains, “We were looking for an automated way to run significant loads at scale, and measure those, change something, run them, run the exact same load again, measure it again. And that’s something you just can’t do with a manual process. So we had to automate it, and SoapUI of course didn’t offer that.”

The learning curve for moving from SoapUI to ReadyAPI was low, allowing Sapiens Decision
to be up and running within days. In fact, the implementation of ReadyAPI took less time than purchasing it!

Final Impressions

Goldberg says, “My excitement is actually having visibility into what’s possible and what our options are in terms of what’s possible, what the price is of adding specific functionality around messaging and additional metrics that come with each decision, how to think about performance versus functionality – all that stuff. I feel so much more confident when a client asks about a volume of decisions and what it’s going to take. And our pricing is now informed by that as well.”

Confidence is provided by building ReadyAPI into the product acceptance testing cycle, to ensure high quality of what they deliver. “That gives us a lot more confidence about knowing that when we say something, it’s going to be real. We used to guesstimate these things. It’s not good to guesstimate when so much is at stake.”

Segal backs him up: “We were pretty good at guessing. Now we can come in scientifically and say, here’s what you’re going to need.”