Getty Images/iStockphoto

SAS analytics and BI platform evolves toward cloud

Though the analytics vendor has invested heavily in AI over the past few years, the most significant change to its platform has been its embrace of the cloud.

An embrace of the cloud and evolving openness to integration with other vendors is driving SAS as it continues expand its analytics and business intelligence platform.

With more and more organizations moving their analytics to the cloud, in 2020, SAS fully redesigned Viya, its primary analytics and BI platform, to make it cloud-native while still offering SAS 9.4 to serve its on-premises customers.

It's a move that's in line with some of SAS' competitors -- MicroStrategy, Qlik and ThoughtSpot are among analytics and BI vendors that have similarly adopted a cloud-first mindset and made cloud-native versions of their platforms their primary offerings -- and resulted in immediate growth for SAS.

According to the vendor, which was founded in 1976 as part of North Carolina State University and is now based in Cary, N.C., its global cloud revenue jumped 19% in 2021, the first full year after it made Viya cloud-native.

Invest in AI, but dash to cloud

Three years ago, SAS pledged to invest $1 billion in adding augmented intelligence capabilities to its analytics tools, Viya being one of them but not the only one. Shortly after that, the vendor added more AI capabilities, including automated data management, automated machine learning, natural language processing and computer vision.

But while not as flashy as the addition of a host of AI capabilities, SAS' embrace of the cloud is its most significant move in recent years, according to Jay Upchurch, the vendor's executive vice president and CIO.

A sample SAS dashboard
A sample SAS dashboard shows how an organization is using the vendor's BI tools to analyze scientific data.

"The biggest step function in our technology has been the rewrite of Viya in a cloud-native way," he said. "The reason the company decided to pivot away from our traditional SAS 9 base was due to the cloud, and while we were pivoting, technologies that were running in the cloud changed around us. We had to recognize the world changing and adopt."

It was 2019 when SAS had to decide whether to continue prioritizing its enterprise platform and on-premises customers or pivot toward a cloud-native architecture for analytics and BI. And during the time SAS was deciding to make Viya its primary platform and redevelop it to make it cloud-native, the concept of containerization emerged as a significant trend.

SAS, as a result, rebuilt Viya with containers to run on Kubernetes clusters.

"The company really reached an inflection point where we had to decide whether to continue on the path we were on or pause and rewrite," Upchurch said. "We made a bet that the operability of our software and its performance on the cloud would be the number one feature our customers would want as opposed to the next function or algorithm."

That embrace of not only the cloud but also Kubernetes has been key for SAS, according to David Menninger, an analyst at Ventana Research.

We made a bet that the operability of our software and its performance on the cloud would be the number one feature our customers would want as opposed to the next function or algorithm.
Jay UpchurchExecutive vice president and chief information officer, SAS

SAS' target audience has historically been data scientists steeped in knowledge of coding languages and statistics, rather than business users. SAS' tools, therefore, can be more complex than those of vendors whose platforms were designed from the start to serve the needs of self-service users.

Containerization makes Viya easier to navigate, according to Menninger. A pricing structure that includes free trials and training is also significant, he noted.

"Two things that are probably most useful to customers are the adoption of Viya 4 based on Kubernetes throughout the product line, and a more simplified named user pricing model including free viewers or readers," Menninger said. "Both of these address one of SAS' biggest challenges -- the breadth and corresponding complexity of their product line."

Meanwhile, because of its embrace of the cloud, SAS now delivers new analytics and BI functionality much faster than it did in the past, and the cloud makes it faster and easier for users to update to the latest version than an on-premises version.

Rather than make major updates every quarter, six months or year, SAS now rolls out new analytics features on a monthly basis. And users can get updates with the click of a mouse.

Embracing openness

Beyond reprogramming Viya to make it cloud-native and increasing its pace of cloud-enabled innovation, SAS has also allowed users to choose public clouds in addition to its own cloud.

Menninger noted that SAS began retooling its analytics capabilities for the cloud long before 2020 when it made Viya cloud-native, but the vendor wasn't aggressive in its acceptance of public clouds. SAS manages its own cloud, and its initial cloud-native capabilities were built to work in conjunction with SAS' private cloud rather than the major public clouds, AWS, Google and Microsoft Azure.

In 2020, SAS unveiled a partnership with Microsoft that made the tech giant the preferred cloud provider for SAS and made SAS' analytics capabilities available on Azure, though SAS still works with other public clouds and SAS customers are free to use the cloud of their choice.

In 2021, an integration enhanced the connectivity between SAS and Azure.

"SAS has been re-architecting its products for many years now, but [at first] they weren't aggressive about offering their products as public cloud, managed services offerings," Menninger said. "From an architecture perspective, they were in decent shape, and now they are investing heavily and making significant progress as evidenced by their partnership with Microsoft around Azure."

That embrace of public clouds beyond the SAS environment is symbolic of a larger trend at SAS.

Just as SAS previously built capabilities for its own cloud, it used its own coding language to develop products, and users had to write code in that language to develop reports, dashboards, predictive models and other analytics assets.

Now, however, SAS enables customers to use R and Python, and even other analytics platforms such as Microsoft Power BI for visualizations in concert with SAS, according to Upchurch.

"For a long time, SAS was viewed as a walled garden," he said. "You had to embrace SAS as the only language for modeling, you had to use only SAS tooling, you had to use SAS visualizations, so if you wanted to use SAS, you had to go all in on SAS. But we're very aware of the world around us, and people want to build with the tools of their choice."

The analytics and BI vendors that are succeeding, therefore, are those that have become flexible and enabled customers to integrate tools from various vendors to build their analytics pipelines, Upchurch continued.

Beyond the dominant public clouds, SAS also integrates with cloud data platforms including SingleStore and Snowflake, Upchurch noted.

"We continue to open up our platform," he said. "We've opened up our systems and said, 'Bring what you want and we're happy to integrate with it to give you the performance you're expecting.' We're continuing to empower our customers' choice."

On the roadmap

In July 2021, SAS was rumored to be up for sale.

Two days after a sale to Broadcom was reported, however, talks fell apart. Just a few weeks later, James Goodnight, co-founder and CEO of SAS, revealed the vendor is working toward an initial public stock offering and hopes to go public by 2024.

If SAS does go public, it would join fellow analytics and BI vendors Qlik -- which has already filed IPO documents with the Securities and Exchange Commission -- and Pyramid Analytics and ThoughtSpot, which are all currently waiting out unfavorable market conditions.

Meanwhile, SAS continues to invest in product development.

Adding more AI and embedded analytics capabilities are on the vendor's roadmap, as are industry-specific applications that enable users to reduce the time it takes to go from data exploration to insight and action, according to Upchurch.

One particular area of investment will be related to digital twins -- a virtual version of a product or workflow -- so users can build simulations and do scenario planning.

"We'll be continuing to help customers evolve their analytics and AI through new techniques," Upchurch said.

In addition, SAS is working with customers to help them migrate to the cloud.

And while AI, embedded analytics, domain-specific analytics and cloud migration are areas where other analytics and BI vendors are similarly focused, by adding capabilities that may not be the most innovative but are in line with ongoing trends, SAS is not simply remaining relevant but serving the needs of its customers, according to Menninger.

SAS is the largest independent analytics and BI vendor, with more than $3 billion in revenue, he noted.

"It's hard to argue with success," Menninger said. "While they may not be the most innovative or fastest-moving company, they are one of the most capable and reliable analytics vendors, which is extremely valuable to many of the largest companies in the world."

Dig Deeper on Business intelligence technology

Data Management
Content Management