SAS, Don’t Lose Your Way

SAS Institute has been around for a long time. Founded in 1976, SAS (originally an acronym for Statistical Analysis System) became and remains to this day the dominant statistical software vendor. Today, as business intelligence (BI) vendors that know little about statistics are promoting themselves as analytics companies, SAS has taken a wrong turn in its effort to defend its status. This is especially true in regard to statistical graphics. When BI companies show their ignorance of analytics by promoting flashy graphics that look cool but are analytically impoverished, SAS is in a great position to remind the world that statistical graphics are about statistics: meanings derived from quantitative data using proven mathematical methods, which are only valuable to the degree that they are accurate and enlightening. No vendor is in a better position to promote statistical integrity than SAS. So why is SAS taking the low road—one traveled by many BI vendors—to tout its wares?

SAS is now building products that try to compete with the likes of SAP’s Xcelsius, which obscures data behind 3-D effects of light and shadow. I recently spent a day with the bright and thoughtful folks at SAS who developed the visual exploratory data analysis tool named SAS Visual Analytics, and for the life of me I couldn’t fathom why they would undermine their product with flashy nonsense that threatens the integrity and reputation that SAS has worked so hard to build. My concern grew even greater when I was recently told that SAS has on two occasions featured David McCandless of Information Is Beautiful fame as a keynote speaker at major European events.


Photo credit: Tricia Aanderud

Perhaps more than any other proponent of information graphics today, McCandless has lured fledgling practitioners of data visualization to the dark side of sloppy analysis and eye-popping displays that rob information of its clarity, accessibility, accuracy, and meaning.

Effective data visualization is informed by science: statistics and those fields that strive to understand human perception and cognition. There’s much that we can learn from other disciplines as well, including the graphic arts, but only by focusing clearly on the goal: finding, understanding, and communicating the truth that resides in data about things that matter to preserve and improve them. SAS is undermining its own work by promoting impoverished graphics and those who advocate their use. Have the reins of the company been handed to sales and marketing executives who don’t understand or value statistics? I implore my friends at SAS: “Remember who you are—if not for your own sake, for the sake of your customers.”

Take care,

16 Comments on “SAS, Don’t Lose Your Way”


By Bill Droogendyk. September 12th, 2012 at 9:20 am

Most disturbing to this long time SAS (exclusive) user.

By Andrew. September 13th, 2012 at 3:30 pm

What an unfortunate turn of events. I know that SAS is split into many different modules, so maybe it’s just the viz people who’ve gone off the deep end; hopefully the statistics side stays sane. Still, it’s sad to see the company itself backing this kind of garbage.

I guess SAS is now an acronym for “Sparkly And Shiny”.

By Stephen Few. September 13th, 2012 at 3:43 pm

Andrew,

The group at SAS that I’ve worked with almost exclusively — the JMP product group — remains steadfast in its commitment to effective visualization. This group has always functioned fairly independently from the rest of SAS. Unfortunately, because JMP is a part of SAS, these bad practices and choices reflect poorly on JMP, even though they aren’t participating in them directly. I’m hoping that the JMP team can speak up as a clear voice of sanity within the organization before more harm is done.

By Jordan Goldmeier. September 13th, 2012 at 5:47 pm

For whatever reason, we (as a society) bestow a privileged status onto whatever presents itself as information visualization, data, analytics and any variation thereof. By privileged, I mean we treat processed data as if it were first principles. When we should be more critical and dig deeper, we accept as fact what is presented by non-authorities like McCandless, vendors, and the news media. The ubiquity of data around us has encouraged the false perception that we make better decisions today because we can see more of the “big picture.” The reality, of course, is that virtually every platform through which we receive information is dominated by a few key players, most of whom don’t take their responsibility very seriously as your experience surely demonstrates.

By Andrew. September 14th, 2012 at 11:40 am

@Stephen: “This group has always functioned fairly independently from the rest of SAS.”

It sounds like that’s something that needs to change.

“I’m hoping that the JMP team can speak up as a clear voice of sanity within the organization before more harm is done.”

I’m hoping that the rest of the organization will listen to them.

By Jeff Pittges. September 14th, 2012 at 1:23 pm

You have lost sight of the golden rule in business — the customer is always right. SAS has a responsibility to their shareholders to sell their products. Successful companies like SAS are driven by the market. Every feature added to the product is based on customer demand. Once a customer buys a product from SAS, then SAS and others may educate the customer on how to use the product.

By Stephen Few. September 14th, 2012 at 5:42 pm

Jeff,

I assume that your tongue is planted firmly in your cheek. Surely you don’t believe that vendors are obligated to build features into their products that undermine product effectiveness and to advocate practices that lead their customers astray. While it is true that many vendors build products and provide services that are primarily designed to sell rather than to provide great value, this isn’t a way to build customer loyalty or a lasting company, and to thereby deliver great value to shareholders. Rather, this is the road to eventual demise.

By Andrew. September 17th, 2012 at 8:07 am

@Jeff: “Every feature added to the product is based on customer demand.”

When did the customer demand ineffective graphics of the type McCandless produces? Do all customers feel this way?

By Steve. September 30th, 2012 at 1:47 pm

I saw a quote recently that seems to sum this up nicely for me which was along the lines of “Just because you can, does not mean that you should…”

By Kathryn. September 30th, 2012 at 4:42 pm

I think it’s great that SAS are trying to improve their data viz! The quality of the analysis is not affected by the way the data is presented - it is the responsibility of the user to produce valid statistics. If SAS are trying to keep up with the times and help their users share information in a way more people are likely to understand and want to look at, more power to them!

By Stephen Few. September 30th, 2012 at 5:00 pm

Kathryn,

Unfortunately, embracing the work of David McCandless and adding 3-D and effects of light and shadow to their charts are not examples of improvements. Quite the opposite. I can assure you, based on a wealth of evidence and deep expertise in data visualization, that these efforts undermine effectiveness. People will understand the information less, not more. Making a chart sizzle in a way that makes the information harder to perceive and understand is hardly an improvement.

What’s interesting is the fact that the folks who have added these effects actually know this to be true. They have paid for my services to advise them and understand the merits of my recommendations. Nevertheless, they have knowingly added features that conflict with the needs of their customers because they know that these features are superficially appealing. In other words, they have opted for the initial win of a sale rather than the ongoing benefit of their customers. In my book, this is not something to celebrate.

By Jared. October 1st, 2012 at 6:53 am

I’ve seen this type of argument before in other tech areas. As respectfully as I can say it, the argument is typically made by ‘more seasoned’ programmers. Sometimes we feel threatened when new approaches start to take over.

A local user group I was part of had an website built using old HTML 3.0 coding. It was functional (to a point), but looked terrible. A bunch of us ‘younger’ folks took over the site, updated the HTML and dressed it up with some graphics and CSS. Same functionality and maintainability, but with additional benefits: easier to find content and more search engine friendly.

As we updated the site, threatened ’seasoned’ members complained, demanded strict xhtml validation and pleaded to avoid the use CSS. Well now look - most websites heavily rely on CSS and javascript. And xhtml is dead. Where would the site be today without these newer approaches?

We just wanted to keep good content current. Similarly, there is nothing wrong with dressing up good, solid statistics. It’s not undermining, but complimenting.

The argument made here is no different than gun debates. Guns don’t kill people. People kill people. Blah blah blah.

Perhaps we should also complain about “point and click” software such as Enterprise Guide? These also empower non-analysts to quickly generate improper results. So where do we draw the line?

Is the problem spoken of here actually caused by visualization tools? Where is the science behind that assumption? Perhaps the industry is becoming too good at romanticizing statisticians and data scientists to the point where non-analysts are attracted to try their luck at making a buck or two? Perhaps the root cause is the social media revolution has saturated the industry with so much data that opportunities are growing faster than the industry can handle?

I think you are onto something, but its the tip of the iceberg.

By Kevin. October 1st, 2012 at 9:18 am

I think it would be great IF SAS were trying to improve their data viz, but it appears they are following the wrong vendors in attempting to do so. Adding features like 3D effects and Speedometers may sell licenses to short sighted CIOs that know little about data analysis, but it does nothing to empower analysts and leaders to make decisions based on their data. Not only do these types of features not improve how you can present your data, often times they do the opposite in distorting the truth. In summary, simply adding features does not improve your product, adding the right features does.

You can have the greatest analytical horsepower in the world, which I firmly believe SAS has, but if you can’t present your data in a way that allows your decision makers to best SEE the story your data is telling all you have is data.

SAS has the opportunity to compete for best of breed in both the analytical firepower and the data visualization if they allow the right individuals to steer the course. From the sounds of it, they have those individuals on the JMP team. Very much looking forward to seeing how this plays out. Does Marketing win or do the experts get heard.

By Stephen Few. October 1st, 2012 at 3:10 pm

Jared,

Your argument is both logically and factually flawed. You began with something called a “false analogy.” My position is not a reaction against new technology. I constantly monitor the latest developments in this field. As a leading expert, it is my job to do so. I embrace what works and reject what doesn’t, not based on opinion, but on science. You will find no evidence to the contrary in my work.

Contrary to your opinion, the features that I criticized as harmful to data visualization do in fact undermine use of the data. Dressing up data with unnecessary lighting and 3-D effects works against the reader’s ability to perceive and use the data effectively. Again, this is not an opinion; it is the finding of research and it is quite easy to demonstrate. Your opinion to the contrary is an error of fact. You won’t find anyone from SAS arguing in a public venue that my position is incorrect or that my criticism of the new 3-D effects they’ve introduced in SAS Visual Analytics are in fact useful. These features were introduced, not because they are useful, but because people see them in other vendor’s products (dysfunctional products) and therefore ask for them, assuming that they’re useful. Rather then responding with integrity and concern for their customers by refusing to add harmful features, in this particular case SAS went for the quick sale. If you believe that technology should support the user well and refrain from undermining the user’s efforts, you shouldn’t be defending SAS in this particular instance.

By Jared. October 12th, 2012 at 9:15 am

I appreciate the response, Stephen. Thanks.

I re-read your post and realized I read too much into it. Where you said “…which obscures data behind 3-D effects of light and shadow”, I assumed (and you know what that makes me) that you were using that as an example to represent the many fancy graph and charting products that seem to be popping up lately (like the ones you find in google analytics or any analytics available in one’s profile from social media). After narrowing it to 3D effects and the like, I realize now I completely agree with you. I feel like a numskull now, and I really thank you for being patient with me.

By Jared. October 16th, 2012 at 8:40 am

Since your post was fresh in my mind, I thought I’d highlight an example. On this page, scroll down to a chart called “Top API Categories”: http://gigaom.com/2012/10/06/business-process-api-ification-the-lego-promise-fulfilled/

Not only is it a pie chart (which is a type of chart that is often abused), but the 3D effect and pastel colouring makes it difficult to interpret - especially the legend.