How to get analytic results across the enterprise

David Pope
David Pope|December 7, 2013

Over the years, in all my roles within SAS, I've been asked by customers, friends, family and all sorts of people, "What does SAS do?" Bottom-line: SAS delivers results across the organization, because that is what it was designed to do.

Specifically, SAS was designed to deliver analytic results, which ultimately provide two main benefits to any organization:

  1. To be more efficient (or grow revenue).
  2. To reduce waste (or cut costs).

This helps explain why SAS has been so successful for 37+ years and why SAS is a great place to work.  Who doesn’t like helping others succeed? Just like SAS Visual Analytics provides great insight through visuals, I've crated the following picture to show how SAS delivers results across the organization. Read More »

Post a Comment

Family, friends, love and … Selena Gomez?? What we're thankful for, according to a Twitter analysis

Trent Smith
Trent Smith|December 6, 2013

On Thanksgiving, I took a few moments out of my day to use the real-time Twitter search capabilities of SAS Social Media Analytics, and the search term #thankful. I didn’t do this because it’s almost year-end review time and I can say I worked on a holiday. Not entirely, anyway. I did it to get a quick analysis of what we, as a society, are thankful for this year.

I let it pull in live tweets for several minutes until it reached more than 6,000 (or until I sensed my wife’s evil eye, boring into my head because I was ignoring the family on Thanksgiving).  After removing re-tweets, we’re left with more than 4,100 tweets that included #thankful. The brilliant Tom Sabo from the SAS Federal business unit took the data and helped me create “The Top 10 Things We Were Thankful For in 2013…Based on a Quick and Dirty SAS SMA Analysis Done While In a Turkey Coma.” Read More »

Post a Comment

What do you get from the new SAS Visual Analytics Community?

Jennifer Marchi
Jennifer Marchi|December 6, 2013

Looking for an information exchange for visual analytics? I'm happy to announce the arrival of the new SAS Visual Analytics support community!

Have a question? Ask it. Have an idea to share? Post it. Join and you’ll have access to lots of great information and resources.

The SAS Visual Analytics Community is a central hub for anyone looking for information, ideas or best practices applied in the working SAS Visual Analytics environment, including:

  • Visualization and analytic techniques - check.
  • Report/dashboard designing and publishing - check.
  • Data preparation and system administration for SAS Visual Analytics – check, check.

Read More »

Post a Comment

Coupons: Six ways you get segmented

Jeff Alford
Jeff Alford|December 4, 2013

I’m not a coupon junkie. In fact, I feel like a coupon victim. Hardly a day goes by that I don’t get an avalanche of offers from:

  • Email.
  • Snail mail (in catalogs, flyers, coupon books, etc.).
  • In-store flyers.
  • Mobile apps.
  • Store checkout.

The vast majority get tossed in my garbage (uh, I mean recycled) or deleted. Some of them are so off target that I know the vendor is using a shotgun approach – blast everyone. Others, the ones that are spookily clairvoyant, get saved. Depending on my budget (or lack of willpower, truth be told) some get used for purchases that I’d actually been considering.

How did they know what I was thinking about buying? The answer for successful retailers is the science of coupon optimization.

Read More »

Post a Comment

How is a customer demonstration like Thanksgiving dinner?

David Pope
David Pope|November 27, 2013

Jennifer Whaley, one of our energy analytic utility experts, and I were talking about customer demos recently and discussing the work that goes into a proof-of-concept done with customer data.  If you've never been directly involved in preparing for a customer demo, you probably don't realize how much time and effort that goes into this type of project.

I'll explain the Thanksgiving connection at the end. Until then, you might think of this as a how-to guide for working with a vendor on developing a demo:

  1. Set a limited project scope with the customer and agree on data sources that can be used to solve the customer problem. Defining success for the project is critical, so that everyone knows what success will look like.
  2. Data has to be gathered by the customer. Even before that step, though, a none disclosure agreement needs to be in place between SAS and the customer, which means legal departments from both organizations need to be involved.
  3. Believe it or not, step one is the "easy part." When we receive the customer data, typically it is in a CSV file or spreadsheet, or sometimes in a database format.  This in itself can be challenging, especially in the "big data" world we live in now. Very rarely is data received in a nice, neat SAS data set that is ready to analyze. (And remember, moving big data from one place to another can take a lot of time.)
  4. This data has be understood, and it helps to have a customer subject matter expert available to help discuss and understand the data.
  5. The data needs to be imported, cleaned, transformed, etc ... all of that 80 percent data prep work takes place before the customer issues can be solved.  (IMPORTANT: your scope must be limited use cases before even starting such a project because success criteria should be agreed upon because without it no one will know what success looks like.)
  6. Now we analyze the data using "magic" to find great insights (hopefully) and prepare reports to deliver the results at an official presentation.
  7. If all goes well, the presentation shows amazing results and there may be a live demonstration to showcase the process and ease of use.  Done well, the actual live demonstration part that shows results may involve 15 minutes or so, not counting time for the questions and answers.

How is this like thanksgiving dinner?   Well, just like thanksgiving, everyone who eats the dinner does not necessarily see all the hard work organizing the time, place, people and preparing all the food. But done well, the "eating" part is all done in 15 minutes or so.  Everyone loves it and then goes off to watch football (except those who clean up).

That's how a custom demonstration is like thanksgiving dinner. Yes, it's a lot of work, but it's almost always worth it.  Is anyone else hungry now?

I hope everyone has a nice Thanksgiving.

Post a Comment

SAS loves stats: Annie Dudley Zangi

Alison Bolen
Alison Bolen|November 26, 2013
Annie Dudley Zangi in her garden

Annie Dudley Zangi in her garden

Annie Dudley Zangi was applying statistical methods to her science projects in primary school before she even knew about statistics. It just came naturally to her, and so did the experiments she conducted at her father's strawberry farm before she learned about split-plot designs. Read this full profile to learn how Dudley Zangi's interest in science and experimentation helped guide her into a career in statistics. And be sure to check out the rest of the SAS loves stats series as we focus on the International Year of Statistics.

What do you do at SAS?
I’m a research statistician developer. I write the code to make the Control Chart Builder platform in JMP work.

What’s your educational background?
Though both my undergrad and master’s degrees are in statistics, it’s the courses I took for a minor in computer science that helped me bridge the statistics knowledge into programming. My first job out of school was working in the Animal Genetics department as a statistician at my alma mater, Virginia Tech.

How do you use statistics in your job?
Part of my job involves researching both new statistical methods and requested published methods. However, focusing on how the engineer or scientist might need to solve a problem is the most interesting and challenging aspect of my job. I prefer to start implementing a new feature from the users’ perspective by designing a straightforward interface with their needs in mind, then figuring out how to solve it in the output statistically and graphically.

Read More »

Post a Comment

Chasing Elephants: How to fit "big math" into the data center

Alison Bolen
Alison Bolen|November 25, 2013

Paul Kent's LinkedIn profileIf you look up Paul Kent on LinkedIn, you’ll find a two-word description for his job at SAS that says, simply: Chasing elephants.

Are you picturing a Disney cartoon? Or an over-sized circus act? What if I told you that Kent’s official title is SAS VP of Big Data? Now, do you know what elephants we’re talking about?

We’re talking about Hadoop.

“My job has been to make big data analytics fit into the data center,” says Kent. He tells a great story about the evolution of high-performance analytics at SAS, and he tells it without the aid of PowerPoint slides or intricate data center drawings. Instead, he uses a dry erase board and a dozen magnets.

Read More »

Post a Comment

Retailers: “The more experiments you make the better”*

Diana McHenry
Diana McHenry|November 22, 2013

The No. 1 retail trend for 2014 will be innovation powered by advanced analytics. Why? Because retailers with the ability to surface and validate new ideas from customer data will be able to test new ideas faster, launch them faster in stores, and roll out profitable programs faster.

Transformation today
Retail is in the midst of significant transformation. This transformation is about thinking differently. Old operations and systems do not apply. Out-of-the box thinking has:

  • Created transformations at established companies, like Belk and Macy’s.
  • Transformed the retail landscape with players like Amazon and Google.
  • Created new companies like Gilt Groupe.

This rapid change has been spurred by the power shift to consumers and enabled by the speed of new computing platforms. According to a recent survey from KPMG, retail executives cited social media (71 percent) and mobile and online shopping (52 percent) as the most relevant technology trends for their industry. In the same study, executives say data analytics is being used for customer insight (72 percent), brand and product management (67 percent), and pricing decisions (56 percent).

Read More »

Post a Comment

3 ways to use SAS with Hadoop (video)

Alison Bolen
Alison Bolen|November 21, 2013

How does Hadoop fit into your IT architecture? Does it replace your data warehouse, or complement your EDW? How can you access data in Hadoop? And what is the best way to take advantage of the parallel processing capabilities of Hadoop for advanced analytics?

These and other questions are answered in the following webcast, which features Brian Garrett, Principal Solutions Architect at SAS and John Kreisa, VP of Startegic Marketing at Hortonworks.

If you want to skip around, the first third is a thorough introduction about big data and Hadoop from Kreisa. In the second section of the video (starting at 17:30), Garret describes three ways that SAS works with Hadoop. And the last section (starting around 37:00) is a detailed Q&A session with live viewers.

The Modern Data Architecture for Big Data Analytics with SAS and Hortonworks

You can learn more about the SAS partnership with Hortonworks or read up on the Hadoop big data trend. 

Post a Comment

SAS loves stats: David Olaleye

Kim Darnofall
Kim Darnofall|November 19, 2013
David Olaleye

David Olaleye, SAS

As a child growing up in Nigeria, David Olaleye would follow his dad, an auto mechanic, as he drove to work every morning. So he always thought he would end up being an engineer in the automobile industry.

But when he went to college at the University of Ife (now Obafemi Awolowo University) in Nigeria, he could not secure a spot in the mechanical engineering degree program, so he studied his second choice: statistics.

“I always loved working with numbers,” said Olaleye, who was also fascinated about the notions of chance and probability, even as he played cards with friends.

In college, he began to see statistics as a way of explaining things and made the connection between what he was learning and the games he played as a child.

“These mathematical expressions explained what I knew intuitively when I tried to guess what cards my opponents had,” Olaleye said. “We make decisions every day of our lives -- tough choices. We need data and statistical analysis to make an informed choice.”

After getting his bachelor’s degree in statistics, Olaleye worked in the Office of Statistics in Nigeria for three months, then got his master’s in social statistics so he could apply the theories he learned as an undergraduate. He also taught introductory statistics to first-year college students.

He got a training scholarship to work on his PhD degree on demography and population studies at the University of Pennsylvania in Philadelphia. “Statistical study of human population encompasses three areas: birth, migration and death,” he said.

While writing his dissertation, he worked as an assistant systems manager at the University of Pennsylvania Medical Center, where he was introduced to clinical trials.

“I love the idea of working on things that actually impact human lives.” David Olaleye

“I worked with physicians who were writing protocols and looking for research support and grants to do clinical trials for new drugs,” he said. “My goal was to help them with data collection and data analyses, and do SAS programming.”

The hospital environment is where his real interest in statistics began. “I love the idea of working on things that actually impact human lives,” Olaleye said.

Read More »

Post a Comment