Analytics Pioneer SAS Emphasizes Perennial Strengths, Embraces Technological Openness

Analytics Pioneer SAS Emphasizes Perennial Strengths, Embraces Technological Openness

(Bryan Harris, EVP, CTO, SAS. Source: SAS.)

SAS (Cary, N.C.) was a pioneer in data analytics for business, but in recent years, its visibility in the market has been affected by a proliferation of new solution providers working with rapidly shifting sources of data and new processing capabilities. Formerly among the noisiest players in the market, SAS seemed to be undergoing some repositioning in order to reemerge in strength. Recently IIR had the opportunity to meet Bryan Harris, SAS’s Executive Vice President and CTO. What he had to say tended to confirm our impressions. The company needed to recalibrate its strategy to leverage its acquired strengths in a new market environment. The result, as we learn from Harris, is a re-assertion of the company’s perennial emphasis on the analytics lifecycle, combined with a new openness to interoperability with other providers’ technology within an ecosystem paradigm. Its strengths are concentrated in its offering of SAS Viya, which serves as a unifying platform for a carrier’s analytics initiatives. The virtue of Viya, in Harris’s words, is that it breaks down data walls and silos, and uses data supported by appropriate governance, managed with a view toward regulatory compliance. The value proposition is not only to reduce complexity and total cost, but also to enable enterprises to leap-frog from a state characterized by a hodge-podge of point solutions to one of an enterprise approach that permits them to be a truly data-driven business.

Insurance Innovation Reporter: Let’s start with an industry-level view. What are the greatest changes happening affecting how insurers use data and analytics?

Bryan Harris, Executive Vice President & Chief Technology Officer, SAS: There are two major technological shifts happening: the adoption of cloud and artificial intelligence—AI.

The significance of cloud is in the agility it affords businesses. Being programmable in the cloud gives organizations the ability to access and dispose of external resources, integrate with services and create an agile platform capable of adapting to shifting business imperatives.

The second great shift is the adoption of AI. The complexity and sheer volume of data handled by companies is expanding rapidly. The only way an organization can make sense of this data explosion is to become proficient in analytics, machine learning and AI. These technologies enable organizations to take advantage of this massive data expansion by scaling human observation and decision-making.

These dual shifts are happening across all industries. Everybody is trying to figure out how to use them together and determine where they need to be on the maturity curve to compete effectively. If you can get more of your technology into the cloud faster and develop more insights with AI and machine learning in the cloud, you’ll be able to create truly differentiated services—and stay ahead of the competition.

IIR: So, insurers—among other organizations—need to cope not just with change, but with the pace of change. How does that play out?

BH: You could look, for example, at what these trends mean for the way risks are underwritten. Insurers are asking themselves, “Can I get more data sources to help inform the way I underwrite risks? How quickly can I do that and how fast are my competitors doing that?” Insurers are also considering how these new technologies can help them create new services and improve customer experiences. In the claims arena, for example, the use of photo analysis technology to authenticate damages has risen from 49 percent in 2018 to about 81 percent in 2021. Of course, these advances open the door to fraud-fighting applications, as well. Accomplishing these tasks requires significant data manipulation and advanced analytics.

IIR: How are insurers of different sizes positioned to react to the increasing pace of change? What is most important to their progress in this regard?

BH: Historically, it has been more difficult for larger incumbents to make the shift. Smaller insurers can be more agile, and startups aren’t burdened by legacy technology and culture – not that startups don’t have their own limitations. Whatever the organization, insurers need the ability to integrate new technology quickly, while recognizing what needs to remain consistent in their value proposition to policyholders. There’s a balancing act between innovation and conservation, but effectively responding to change is the greater challenge.

That brings us back to cloud, the precise benefit of which is making it easier to pull in different data sources, technologies and services to make insurers more agile in their response to changing market conditions.

IIR: Let’s talk about SAS’ own adaptation. SAS was a pioneer in data analytics, an important vendor in insurance as well as horizontally across industries for over 40 years. How has the company changed to ensure its continued relevance?

BH: The short version is that we had to disrupt ourselves. First and foremost, we had to make sure our SAS Viya platform was not only cloud-native, but cloud-portable. Similarly, we saw a balance of innovation and conservation. One thing SAS has done longer than anyone else is focus on and develop the analytic lifecycle—understanding what it means to take raw data and improve its quality for operational purposes, readying it for meaningful analytic modeling.

One crucial change in our approach is to move away from a proprietary approach to technology. We had that luxury for a long time. Rethinking how our technology fits into a larger insurance ecosystem has fueled innovation that makes our platform and solutions more open and flexible.

IIR: Why has that become important in your market space?

BH: Today, enterprises create road maps to plan how they’re going to modernize their analytics capabilities. Typically, those road maps involve an analytics ecosystem rather than a partnership with a single vendor. In a market crowded with vendors, we need to build our software such that we don’t assume we own the whole thing. We recognized that we needed to onramp different technologies and languages. We’re no longer just about SAS—now you can bring your own language. Within SAS today, you can write analytics in Python, with a full syntax highlighting. You can then take that model and drop it into SAS Studio or our Intelligent Decisioning and Model Manager.

IIR: How would you summarize SAS’ market approach today?

BH: If you think about the demands of the market, it’s the resiliency and agility of the cloud and AI that’s helping insurers make sense of the disruption that’s occurring. What we’re pitching is SAS Viya as the most productive analytics platform in the market to help customers adapt in the face of exponentially growing datasets.

That’s really the driver for everything. When I think about our investments moving forward, it’s around how to enable low-code/no-code application building for our customers. As they use SAS, and create value from SAS, they can monetize that by creating new services, outcomes and datasets.

IIR: How are you going to market with the pitch on SAS-as-platform? What are major characteristics of the platform that serve the goal of accelerating an enterprise approach to data analytics?

BH: SAS Viya is our cloud-native, AI, analytic and data management platform. It’s fully containerized, running on Kubernetes, and supported by continuous integration and continuous delivery. Last year, we released SAS Viya on key cloud providers, including Microsoft Azure, Amazon Web Services, Google Cloud and Red Hat OpenShift. And we’ll release SAS Viya on Open Source Kubernetes for on-premise deployments later this year.

Capabilities required for the complete analytics lifecycle. Source: SAS. (Click to enlarge.)

SAS Viya makes it possible for users of all skills, and at every level of the organization—from data scientists, to business analysts, to executives—to collaborate and scale while operationalizing discovery and decision-making. This is even more important now during this time of uncertainty. Customers across industries are finding it easier with SAS Viya to build models and get them into production to impact the business faster than our competitors. We’ve been intensely focused on the analytics lifecycle because without this perspective, you can’t achieve the pace of model creation and deployment, or the governance needed, to effectively operationalize models on an ongoing basis.

IIR: This implies that SAS Viya addresses historical challenges that insurers—and others—have faced, such as the tendency to deploy analytics as point solutions, with little or no enterprise perspective.

BH: Yes. Organizations have been prone to hoard data within departments, making it harder to connect the dots across them. In insurance, this has also been perpetuated by using InsurTech offerings as point solutions—which is not to say InsurTechs don’t contribute value. The challenge is to break down those walls and silos, and use data supported by appropriate governance, managed with a view toward regulatory compliance. Insurers must have a vision for data as a foundation of their entire workflow. They need to demand that of their vendors, while embracing the task of unlocking the value of their own data. Without an enterprise approach, the complexity of the environment will multiply and drive up total cost of ownership.

Data needs to reside in a manner that you can protect it while making it more accessible across the organization. SAS Viya provides a way to connect the dots between actuaries, underwriters, claims adjusters, fraud investigators and others. In the absence of a platform such as SAS Viya, data continues to be siloed, and carriers struggle to channel it into workflows that dramatically impact the business. With a platform like SAS Viya, insurers can pursue a strategy based on rational, enterprise-wide use of data and analytics. They start to become truly data-driven companies.

IIR: What does such an approach mean for how insurers work with their vendor partners, SAS or otherwise?

BH: Insurers must take the position of owning their data first. If a vendor wants to put proprietary data on top of that, we recommend against it. Optimally, insurers should keep themselves in a position where they can swap in and out vendors that can unlock value in their data. That fits with SAS’ shift to work in open formats as much as possible. The starting point is for insurers to have a strong data ecosystem and then explore tools that fit into an enterprise road map and a culture of adding the value of data and analytics to the entire business.

IIR: So, then SAS Viya was designed as a platform to enable this approach by providing a unifying platform for the most important aspects of analytics?

BH: Yes, and it’s an expression of our perennial focus on the entire analytics lifecycle. SAS Viya is a platform designed to drive data and analytics innovation and navigate disruption. It’s a way to unite and govern enterprise data and drive new and better decisions, whether it is for underwriting, mitigating enterprise risk, managing claims, detecting fraud, or delivering a better customer experience.

It also serves as an enterprise platform for maintaining the compliant use of data and analytics, which is more important as the accelerating use of AI raises concerns about data privacy and bias. SAS Viya has full traceability regarding how models are created, all the way back to the raw data and the models used to analyze its accuracy. We built that governance into everything companies can do across the platform, as part of our commitment to responsible innovation.

We know insurers must strive on multiple fronts to keep up with how technology is changing the way business is done. SAS Viya is designed to accelerate transformation in data and analytics.

Northwestern Mutual’s Christian Mitchell: Agile Response to Customer Needs