Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.
May 2, 2019
4 Min Read
Image: Jakub Jirsak - stock.adobe.com
When it comes to data analytics in the enterprise, SAS has pretty much always been among the top players. Its customers are in the big league, too: Bank of America, Honda, and Nestle, for instance. SAS claims that it works with 94% of Fortune 100 companies in some capacity. Gartner classifies SAS as a "Leader" in data science and machine learning platforms.
Machine learning and artificial intelligence technologies are among the hottest identified by top companies for investment over the next several years. Gartner's 2018 CIO survey showed that only 4% of CIOs have implemented AI, but that 46% have developed plans to implement it. It's an environment that would seem like it should be driving a gold rush for SAS. Yet it is not without its challenges.
As the demand has grown for data analytics and machine learning, so has the competition. There are big data platform vendors such as Cloudera and Databricks that have grown over the last several years, data visualization specialists such as Tableau, and stalwarts in the space such as IBM. There are also a number of popular open source tools such as R and Python that have been embraced by a new generation of developers, engineers, and data scientists.
SAS recently acknowledged the demand for artificial intelligence technologies with a pledge to invest $1 billion in machine learning, natural language processing, computer vision, and more over the next three years. The company elaborated on some of those investments this week during its SAS Global Forum user event in Dallas, spelling out product developments it has planned and its strategy going forward.
The investment and announcements marked a stake in the AI ground that SAS needed to make.
"Attaching a 'B,' as in billions to its AI investment plan signals an effort to raise the company's profile and remind people that SAS is a large company that has been focused on machine learning and the analytics space generally for a long time," Doug Henschen, principal analyst and vice president at Constellation Research, told InformationWeek. "Open source frameworks like TensorFlow, scikit learn, XGBoost, mxnet, and Pytorch are getting all the attention these days, and when you hear about innovators and Internet giants, you're often hearing about their use of these frameworks and public cloud capabilities. It's a risk for SAS to be left out of the AI conversation, so it's clearly taking steps to retain some mindshare."
SAS may grabbing that conversational attention at the right time. IT leadership knows it needs to invest in AI, but has found actual operational AI to be a bit of a challenge. While many may have planned to build their own systems from scratch internally, getting up to speed quickly may require a pivot to investing in an established system.
"Too many companies are caught in AI science-project mode and don't have the know-how to make the leap to a machine learning model that meaningfully affects business," said Oliver Schabenberger, COO and CTO, in a statement released during SAS Global Forum. SAS is hoping to help them get there.
Henschen noted that among SAS's strengths is the "comprehensive, end-to-end nature of its platform, spanning from data ingest, data preparation and model development and, importantly, through to model deployment, monitoring, and ongoing management of models in production at scale. Lots of hot companies in the ML and AI arena only cover bits and pieces of that total life cycle," he said.
Gartner has noted that SAS customers continue to ask for more support for open source tools, libraries, and frameworks. The next generation of data analysts and data scientists are swinging the pendulum, and named Python as their preferred language in quantitative executive recruitment firm BurtchWorks' 2018 survey of data science language preferences.
Plus, SAS needs to improve support for Docker and other containers.
But the enterprise case for SAS potentially remains a strong one, according to Henschen.
"Companies have to weigh the benefits of the breadth, familiarity, comprehensive nature and performance of the SAS platform against the cost and performance of open source and cloud capabilities and what they would have to do to bring together all the tools needed to support similarly scaled deployment and ongoing management," he said.
For more on AI and machine learning check out these recent InformationWeek articles.
Also, review the Emerging Technologies track at the upcoming Interop19 conference.
About the Author(s)
Jessica Davis is a Senior Editor at InformationWeek. She covers enterprise IT leadership, careers, artificial intelligence, data and analytics, and enterprise software. She has spent a career covering the intersection of business and technology. Follow her on twitter: @jessicadavis.
You May Also Like