With the headline, "Get inspired by the life-changing benefits of Dell Statistica," we linked to a new case study in the April/May issue of the Statistica Monthly News. This one is all about the rare diseases unit of Shire, a global biopharmaceutical company that seeks to ensure the robust and uninterrupted supply of quality medication to its patients. The implementation of Dell Statistica has helped them conduct statistical process control, monitor processes and identify areas for improvement.
Anything that reduces a multi-day analysis down to mere minutes without sacrificing quality has got to be good, right?
Given the nature of Shire’s business and the kind of help they offer to people all over the world, this is a very inspiring story, and Dell Statistica is proud to be part of it.
Read more in the April/May issue of Statistica Monthly News >
How do you empower more people with advanced analytics for greater impact on how they do their job? How do you embed analytics everywhere so you can make data-driven decisions? How can you use analytics to innovate faster?
These questions keep us up at night, just as they keep a lot of our customers up at night. We don’t have a silver bullet for them yet, but we’re moving Statistica closer to being one, little by little.
The press and analysts are starting to notice. They’re talking to you and finding out that Statistica is meeting your needs and then some. Many Statistica customers are commenting favorably on the ease of use, completeness of solution, integration efforts with other Dell products, and ongoing investment by Dell in the product and people since the acquisition last year.
That’s encouraging news as we get ready to launch this quarter’s Statistica version 12.7, with features aimed at helping you get advanced analytics into the hands of more of your users:
We plan quarterly releases to Statistica from now on. If you already use Statistica, keep an eye out for upgrade instructions. And if you’re tired of staying up at night figuring out how to get analytics into the hands of more of your users, keep an eye out for the free trial version to download.
What do your smartphone and your office air conditioning system have in common? Surprise—they top the list of IoT data sources.
In the May issue of the Statistica Monthly News, we shared an item about a new infographic that summarizes the results of a survey Dell commissioned from Enterprise Management Associates (EMA). It’s all about the Internet of Things (IoT) marketplace.
When it comes to the IoT, I tend to think of it as something that is still theoretical, probably because I didn’t wake up this morning to learn that Skynet was suddenly in charge of everything. But the build-up of the IoT is in full swing and has been for some time with machine-to-machine (M2M) sensors and real-time automation. Perhaps it is not at critical mass yet, but EMA’s infographic identifies some of the major industries pushing that swing. Review the infographic and see what other interesting information EMA uncovered.
You never know what you’ll find in the Statistica Monthly News email. The top item in last week’s issue was a blurb steering readers to the press release announcing Dell Software’s partnership with Datawatch that will take Statistica’s already robust data visualizations and dashboarding to a whole new level of interactive dynamics.
Historically, a common objection from prospects was that certain spreadsheet programs could display business data visualization with sufficient detail for all manner of reporting. So, they would ask, why upgrade to Statistica? Arguably, this might have been partially true to a point at one time, but there comes a day when market changes demand advanced analytics, and advanced analytics demand advanced visualization in order to do any good. And it is so much easier to conduct analytics with a comprehensive platform like Statistica that doesn’t require data to be output to a standalone visualization application.
I have seen impressive, real-time demonstrations of this new capability in Statistica and am excited about what the Datawatch partnership has to offer our users.
Read more in the May issue of Statistica Monthly News!
Most predictive analytics professionals would agree that the cloud has made its way into their IT organizations – certainly in the form of Software as a Service (SaaS) but likely beyond use of web-based software to cloud deployments in the data center. In fact, cloud computing is supporting, and in many cases driving, today’s technology trends. The cloud, then, provides a way for IT to deliver increased value and support the c-suite’s priorities as outlined in my previous post on Analytics in the Cloud
As funding sources shift from IT to line-of-business executives, analytics will be increasingly tied to business strategies to increase a company’s competitive advantage. In the end, business will be focusing more and more on speed to value.
Cloud analytics are an increasingly important component of successful business intelligence and analytic strategies. But more analytics could mean more complexity, if not managed properly. And while adding cloud analytics to the top of the “to-do” list might seem daunting, there are actually simple-to-implement techniques for taking full advantage of cloud analytics at your organization. It all depends on having the right tools.
We’ve lined up industry experts to help you gain a deeper understanding of analytics in your cloud and hybrid data ecosystems. Jacob Spoelstra, Director of Data Science, Azure Machine Learning Platform at Microsoft and Christopher Ray, M.D, Chief Technology Officer at AnesthesiaOS are leading an educational webcast to show you how to harness cloud analytics to gain a significant competitive edge, with less effort than you might imagine.
Is the cloud a part of your go-forward business intelligence and analytics strategy? Register for a complimentary webcast featuring experts from Microsoft and Dell to learn how to get a handle on your hybrid data ecosystem.
Want more? Following the webcast you’ll be invited to demo a free trail of Dell Software’s Statistica, one of the most intuitive, easiest-to-use, and comprehensive predictive analytics software platforms available in the market.
Sign up for the webcast now to learn how to take the complexity out of using cloud analytics for a competitive advantage.
Dr. Thomas Hill Executive Director of Analytics for Dell Information Management Group recently contributed a very interesting article entitled “When Smart People Won’t Use Smart Technologies”. Intriguing title, the article looks more closely at how different people respond when exposed to innovation and new technology options and how it affects people’s attitudes toward their physicians, healthcare provides, and personal health data. One would expect quick adoption and overall excitement towards new capabilities that combine data to enhance our care– apparently this is not the case.
Dr. Hill’s article provides insights on how patients are reacting to more analytically driven healthcare and how you can help your patients embrace the new age of technology and healthcare.
Learn how you can be smarter with your technology implementation by preparing your organization with governed and integrated data that is delivering results and value to the patient.
What do the 43rd President of the United States and I have in common? We like hanging out with healthcare technology professionals!! President George W. Bush gave the closing keynote at this year’s Health Information and Management Systems Society (HIMSS) event in Chicago and Dell was one of the Corporate Sponsors. It’s been a few years since my last visit to HIMSS and I have to say I was extremely impressed. The event is attended by over 42,000 people and seems to cover every square foot of the McCormick Center. Dell had an extremely strong presence at the program, hosting a charity in the booth, a dozen different Dell HCLS solution demo's and 3 live tweetups.
The technology themes were varied throughout the event and I was there to help lead a discussion on Population Health with Dell experts Dr. Gary Miner, Dr. Tom Hill and Dell's acting Chief Medical Officer Dr. Charlotte Hovet. We were also joined by Dr. Ken Yale, Vice President of Clinical Solutions, ActiveHealth Management. It’s interesting to see where data is playing a role in driving more consistent and higher quality patient care. Population health obviously benefits from data driven insights. Technology's like advanced analytics are helping us move beyond an understanding of large populations and to focus in on more personalized patient care via diverse data and insightful analytics. As we are able to leverage more data and a greater variety of information on specific patients the ability to personalize care and apply a customized level of best practices will result in much better overall patient care.
L-R Shawn Rogers, Dr. Gary Miner, Dr. Tom Hill, Dr. Charlotte Hovet and Dr. Ken Yale
The end result as advanced analytics drives patient care forward will be precision healthcare where care givers are able to execute specific regimes for each individual based on their specific needs, chemistry, DNA and other personalized markers and prerequisites. It's exciting to think that advanced analytics has the ability to enhance treatment and deliver personalized healthcare. Innovation isn't without its hurdles, connectivity to data and a new responsibility of patients to bring their own data into play will prove difficult. New trends will include device information on a patient’s exercise and activities, diet, location, travel history and more. Advanced analytic platforms will factor many new data points into models in order to achieve the highly specific care plans required by precision medical treatments. Look for care givers to push back a bit as the culture of human knowledge and instincts collides with automated and model driven best practices. I believe that ultimately both voices need to be heard in order to supply the best possible care. Dr. Hovet made the point that even though analytic platforms will supply a path for treatment Dr’s will still play a critical role in communicating, implementing and executing precision medical treatment. The days of the robot doctor are still way out in our future.
Having been in the data business for as long as I have, I found the themes at HIMSS to be exciting and full of promise for future and immediate innovations based on our ability to leverage greater amounts of data and a wider more dynamic range of information in order to add value to overall patient care. These are exciting data driven days for health care.
Many businesses pursue analytics projects with the idea that they can respond ever more effectively to dynamic customer, market and business demands. Thompson describes the critical factors of analytics agility necessary to make this happen.
Applied to the preponderance of connected monitoring, social psychology research suggests that increased public awareness of the Internet of Things around us will influence our daily choices for the better. Davis ruminates on this effect.
In her latest CMSWire article, Schloss notes that the very term, "Shadow IT," portends rogue employees and covert operations, but such negative connotations are unwarranted. Shadow IT will only grow larger in a world of advanced analytics, so she examines the common myths to explain why businesses should want Shadow IT to thrive.
One way to think of "advanced analytics" might be as "complex analytics," the kind that go beyond the traditional analytics employed to produce business intelligence. However, there is another meaning that is certainly apropos in the context of anticipating circumstances, predicting trends, and prescribing actions. In such future-facing scenarios, "advanced analytics" might suitably be thought of as "analytics in advance." This is where the analytics maturity model starts to make sense.
David Sweenor, Statistica product marketing manager in Dell Software's Information Management Group, provides a practical summary of this maturity model in his recent post, "5 ways to boost your business IQ." It is worth noting that the model he touts is layered like a pyramid, with each layer built upon the solid foundation of previous layers. There is no skipping ahead when it come to maturity: every level of analytical maturity must be earned and learned in sequence.
Accordingly, Sweenor helpfully provides a quick overview of five advanced analytics techniques that should be evaluated by any organization seeking to build up its maturity: segmentation, decision trees, predictive models, text analytics, and optimization/simulation. And he describes some helpful Statistica case studies that prove the value of advanced analytics in real-world scenarios. Read David's post and see where you are in the model. You can also find more Statistica case studies under the "Resources" tab here.
The need to be more agile comes up a lot in customer conversations, especially from frustrated executives who want to be more sure-footed and flexible in moving their businesses forward. Too often, they feel stymied by a lack of useful insight, which hampers their ability to respond quickly and effectively to changing customer, market and business demands.
Luckily, this gives me an opportunity to bring up one of my favorite topics: analytics agility. With the right mindset, tools and technologies, organizations can become much more adroit about how they use the power of analytics to improve decision making. As with most things, the toughest part is getting started.
According to Dell’s 2014 Global Technology Adoption Index, 61 percent of companies worldwide have big data waiting to be analyzed—and yet only 39 percent of those polled felt they had a firm grasp of how to go about extracting value from that data. What it takes is a mix of intellectual curiosity and intestinal fortitude to develop an understanding of how your business works from a data perspective.
In my experience, there’s usually a group of naturally curious intellectuals in every company that are eager to drill down into business facts and figures to discover trends, triggers and roadblocks impacting business success. Thanks to our increasingly connected world, these data miners have more tools and techniques at their disposal than ever before to look at data from all angles.
Also critical is having a supportive, equally curious leadership team that encourages the use of data to figure out the business. I met recently with a large sportswear manufacturer that invests heavily in analytics to support a variety of marketing initiatives. The challenge for the analytics team is that when the data supports what the marketers want to hear, it’s all good. When the analytics reveal a different outcome, the marketers claim that the data is bad and do what they want anyway.
Unfortunately, having the right analytics tools and smart people driving the process won’t make much difference if company leaders aren’t open to learning and following what the data reveals. After all, the point of analytics agility is the opportunity to quickly and nimbly change direction completely or make a minor course direction before it’s too late. As we all know, however, sometimes it takes a few lessons learned the hard way to realize the data was good to begin with but the business decision was flawed from the start.
Another critical success factor to increasing analytics agility is having the support of an IT team that continually and consistently collects, manages and exposes data for a variety of analytics efforts. Historically, this has been one of the biggest stumbling blocks as traditional, centralized IT teams often were too overwhelmed with “break-fix” tasks to respond quickly and efficiently to analytics requests. In the past, many early analytics efforts died as soon as the financial, sales and marketing people generated data from separate silos of information and nothing matched up.
Thanks to continued IT decentralization and increased data sharing, it’s much easier now for IT to build an infrastructure that brings different types of data together and delivers a single version of the truth that everyone can get behind. When that occurs, the journey to analytics agility becomes a shared experience that produces tremendous insights and business breakthroughs.
And in some cases, even medical miracles. I’m still sharing the story about Dr. John Cromwell at the University of Iowa Hospitals and Clinics. As reported in the Wall Street Journal’s CIO Journal, Dr. Cromwell is using Dell Statistica to better predict which patients face surgery risks and then expedite surgery room decisions on which medications or wound treatments will be most effective. Now, that’s a prime example of the power of analytics agility.
I also recently spoke with Danske Bank, the largest bank in Denmark and one of the leading financial institutions in northern Europe. This Dell customer is doing some amazing things with Statistica and various credit scoring tools to produce real-time insight that enables cutting their credit risk exposure nearly in half. By taking advantage of analytics agility, the bank can make up-to-the-minute decisions about which markets to serve to gain a competitive edge while mitigating risk.
Today, we have the analytics tools to drive fast, flexible business decisions. And, each day, I hear about another customer with a strong IT and leadership team backing efforts to push the analytics envelope. I’m encouraged to see more companies getting a firmer grasp on what they can do with greater analytics drive and dexterity.
I’m looking for more examples of how companies are flexing their data muscle with analytics agility. Drop me a line at email@example.com to share what you’re seeing.
Smart phones were the harbingers of the connected future as they transformed from merely portals for consuming information into sensors and location-based signaling tools. But the Internet of Things (IoT) takes that to another level where relatively inexpensive devices can be connected and managed through automation suites, then mined for new information. As I write this, my Nest thermostat is reporting my energy usage, and my Enphase Energy hub is recording my solar power generation. In April, in Northern California, it is almost ideal conditions for both, with no need for climate control and ample sunshine. Finally, my Kevo locks let me know when people come and go from the house or when a door is left open.
We expect this automation trend to accelerate, too. From water sensors to warn you when your plumbing has gone awry to smart pill bottles to help insure accurate daily dosing for an increasingly aged population, the range of information that we will soon have access to will likely become overwhelming. Indeed, just managing all of the devices, protecting privacy while allowing access, and figuring out how to merge related data streams together is already becoming a growth industry.
The latter may be the most interesting and valuable part of the technology puzzle, as well as the most challenging. Data analysis traditionally falls into several areas. The most common is alerting and descriptive statistics. The smart pill bottle needs to, at a minimum, alert the patient when the dosing is incorrect. It might also provide aggregate statistics concerning compliance. Where it becomes more challenging is when those statistics are merged together with other information sources. What is the correlation of a medication with the successful treatment of the condition or even off-label impacts and adverse effects? This has traditionally been the domain of retrospective analysis with large variations in the data sets due to incomplete and inaccurate reporting and the challenge of collecting together sufficient records. While medical privacy laws like HIPAA play a part in blocking effective access to this data, reporting is at least as great a challenge.
We can see similar patterns in terms of energy monitoring and consumption, like with my solar array. We recently used connected temperature sensors in refrigerators at Dell to monitor usage and performance of the devices. An odd spike occurred in one on Friday afternoons, showing the impact of an ice cream party on the energy consumption of the device. While a humorous outcome, opening and closing refrigerators consumes around 7% of the total energy used by the devices, and broad monitoring patterns has the potential to help reduce this waste.
These kinds of outlier patterns in the data, whether in off-label effects of medications or in energy consumption by appliances, have a broad social impact. Social psychologists have known for some time that the setting a person is in can strongly affect their choices. For instance, when people see others around them picking up litter from the ground, they are much less likely to litter themselves. We can guess that the awareness that connected monitoring will bring to our lives will have a similar effect. While we may try to avoid leaving the fridge door open, seeing the impact of such actions on the building, the company, the city, the nation, and the world creates a network of awareness and expectations that reinforce better behavior.
Understanding the architecture that supports Internet of Things (IoT) projects at first glance can seem overwhelming if not impossible to emulate. There is a wide array of technologies that support these projects and help customers to wrangle the data involved in these programs. Data management, integration and analytics all play a key role in delivering innovative and responsive IoT projects.
Enterprise Management Associates (EMA) recently authored a white paper titled Demystifying the Internet of Things: Implementing IoT Solutions. The paper explores the necessary strategies to implement a project and also highlights some of the solutions that Dell provides its customer to support these projects. EMA identified the most popular data sources that companies are utilizing for IoT projects. (See figure below)
Geo-location devices are the most popular representing 14.7% of the projects researched by EMA but with that said its clear there is a wide variety of opportunities ranging from Corporate infrastructure data, manufacturing data and consumer infrastructure just to name a few. EMA’s paper explores the challenges of integrating, replication, analyzing and securing a responsive environment that moves at the speed of the business to support greater business insights and enable action. Companies adopting IoT can benefit greatly from a smarter view of their business and create competitive differentiation by using this technology. Download a copy of the full paper here and share your thoughts on IoT with me. I’d like to hear how your company is taking advantage of this new and innovative approach to data.
What do 257 of your peers with 841 different cloud-based analytics and BI projects know about that you may still be missing out on?
Well, Shawn Rogers pointed out in his recent blog “Why Do Analytically Driven Companies Adopt the Cloud?” that there are several other business and technical drivers associated with cloud adoption besides saving money. From a business perspective, speed of implementation, flexibility and quick start templates top the list. From a technical perspective, data security, technical agility and software availability are foremost concerns.
Think of it like this:
If you’re feeling a little FOMO (fear of missing out), take a look at “Analytics in the Cloud,” a new report on research conducted by Enterprise Management Associates (EMA) into the state of cloud-based analytics in enterprises around the world.
Now that we understand a bit more about the motivating factors surrounding cloud based analytics projects, I was curious as to what types of business problems people were tackling.
If you’re curious to see what the top tool choices are, read the entire report (hint, hint….Dell Statistica ranked #1).
The companies surveyed reported a total of over 800 planned or active implementations of cloud-based analytics among them. If you haven’t started moving your analytics out of the data center and into the cloud yet, find out what these companies know that you don’t and use the report to start building up your business case for analytics in the cloud. And if your analytics projects are already in the cloud, find out how your organization’s effort measures up.
Still feeling like you’re missing out? Sign up for our April 9 live webcast “Myth or Reality – Learn How to Make IoT Analytics Work for Your Organization,” featuring Howard Dresner. You’ll come away with even more ideas about ways to introduce and grow cloud-based analytics in your organization.
You have nothing to lose but your FOMO.
As I shared in my last post, February’s TDWI event in Las Vegas held special meaning for me, since it marked the organization’s 20th anniversary and my fourth consecutive year as a faculty member. Dell Software also hosted a book signing for my latest publication, co-authored with Krish Krishnan and entitled Social Data Analytics: Collaborations for the Enterprise.
This book is the first practical guide for professionals who want to employ social data for analytics and BI. We offer a series of use cases and examples to help readers make the most of the five social data types: Sentiment, behavioral, social graph, location and rich media data. I shared some of the insights from the book in my class, aptly called “Social Analytics Driving Real Business Value with Big Data.”
Every time I teach this course, I’m struck by how the use of social data continues to mature. Early on, social analytics discussions were met with equal amounts of skepticism and cynicism. Executives, in particular, had a chip on their shoulders and couldn’t see where social data fit in their enterprises. Now, instead of challenging the validity or usefulness of social data, companies want to understand how they can take advantage of social analytics to be different and more innovative.
To demonstrate the continuing maturity of social analytics, I hosted my first “Level Four Maturity” student in Las Vegas, as his organization is participating, listening, integrating and analyzing data in a project. Each of these represents a level of my Social Media Maturity Model as described and depicted below:
Most organizations get stuck between levels two and three because it requires them to integrate siloed social and business data. However, my student, who worked for a retailer, shared how his company had cleared this hurdle to have a major impact on their supply chain. Not only did the company successfully integrate multiple systems with multiple data sources, they then went on to aggregate and mash it to produce actionable insights to better serve customers.
By listening and engaging with customers, the retailer developed a better understanding about which particular brands and products were liked best. This data then was integrated with geo-location data to determine regional preferences for different brands and products. Armed with this knowledge, decisions could be made about which models should drive the supply chain.
While this example went straight to the head of the class, the majority of students still were struggling to understand the best ways to listen and join the conversations around the five major social data types. And for good reason.
Each type brings a unique set of capabilities from an analytics standpoint, especially when you marry them. The possibilities for innovation are endless. Interactive billboards speak to your phone and engage you on the spot. Retailers capture in-store location data to produce behavioral insights designed to improve your shopping experience.
As social analytics mature, the line between innovative and icky gets blurry pretty fast. That’s why we spent time talking about the cultural aspects of maturing social models. Everyone agreed that companies need to invest in policies and best practices that address transparency and privacy.
The time is right to get schooled on the impact of social analytics because as it evolves, walls between siloed areas will come down and make it much easier to integrate data into wider decision platforms. Solutions, like Dell Statistica, analyze highly useful social sentiment data that helps customers create more enhanced views or highly detailed slices of their customers.
Actionable business insights are within reach of companies that effectively listen, engage and integrate social data from one platform to another. Then, they can bring all this information into their critical workflows for enhanced decision making. As this area continues to evolve, slews of real-world examples will surface that will inspire others to learn how to fulfill the promise of social analytics within their own organizations.
I can’t wait until my next class to find out which new innovative uses of social analytics make the honor roll. Until then, shoot me an email at Shawn.Rogers@software.dell.com with your best social analytics use case.
A few of us visited Dr. Cromwell at the impressive University of Iowa Hospitals and Clinics on a chilly day in late November to capture his story on video.
One of the reasons I find predictive analytics so intriguing is the sheer number and variety of use cases, relative to other technologies. It is, after all, the use cases and customer stories that bring us out of esoteric, theoretical territory and make a technology meaningful and relevant. That’s why it was so inspiring to hear about the University of Iowa Hospitals and Clinics’ compelling and real world use of predictive analytics.
Dr. John W. Cromwell, director of the Division of Gastrointestinal, Minimally Invasive, and Bariatric Surgery has been able to significantly lower the risk of surgical site infections for his patients using Dell Statistica. Great example of how advanced analytics software can really move the needle in improving patient outcomes!
By Dr. Thomas Hill, Executive director, Analytics, Information Management Group, Dell
Big-data predictive analytics offer the promise of better outcomes and lower costs for healthcare organizations, effectively allowing a patient to access the expertise of thousands of experts gained through treating millions of patients. But successfully deploying the technology isn’t always easy. How you plan for and introduce analytics is critical to acceptance by stakeholders and a willingness to take action based on the knowledge you generate.
The Locomotive from Nurnberg to Furth: The fear effect of technology
Disruptive technologies almost always elicit initial skepticism and even fear. When the first steam locomotive prepared to make its run from Nurnberg to Furth in 1835, people were concerned about noise and pollution and feared that human physiology might not support travel at speeds over 20 mph. Other useful-but-disruptive advances have also generated initial fear.
In research published over 30 years ago, I demonstrated how a lack of a sense of control generated fear when personal computers revolutionized computing at universities. Interacting with a black box that generates results as if by magic, without giving any control to the end-user, will always generate distrust.
Projects will fail if analytics technologies are perceived to usurp personal judgment and control over final decisions.
Presenting analytics to stakeholders as a tool they can use will empower them and pre-empt fear. Demonstrate how these new tools help healthcare professionals to quickly evaluate risks, potential outcomes, and what-if scenarios. Predictions, recommendations and prescriptions derived from analytics need to come with reasons why a particular risk is indicated and how recommended actions will affect outcomes.
Avoid alarm-fatigue with unambiguous, actionable alerts
People cease to pay attention when alarms are too frequent or information doesn’t present clear options for action.
Enhance rather than add to existing processes, screens and alarming rules and ensure that important information is unambiguous, actionable and consistent with existing work flows. Think through where analytic results will be used, identify benefits and ROI and make certain that information is actionable. For example, Dr. John Cromwell implemented a system at the University of Iowa Hospitals and Clinics which sends real-time, actionable risk information to the operating room that is helping surgeons avoid post-surgical infections.
Don’t add to the onslaught of computer work
General surgeon Jeffrey Singer recently noted in the Wall Street Journal that rigid electronic health records systems promote “tunnel vision in which physicians become so focused on complying with the EHR worksheet that they surrender a degree of critical thinking and medical investigation.”
Analytics technology should be entirely hidden, yet deliver reliable information about risks, best next action and alternatives. Don’t require medical professionals to complete yet another computer screen.
Know the end point and how to measure results
One of the most important things to consider before embarking on any IT project is to clearly establish what the completed project would look like and how to measure success. Avoid projects that are attractive because of the “cool” technologies involved without clear definitions of success and ROI.
Think about what ideal results look like; who would use them and how; and how something of value would be created. Involve key stakeholders and end-users to reflect their concerns and perceived barriers to success. Once you know how success is exactly and operationally defined, everything else follows, such as where to look for what data, how results are delivered, what level of integration, training, operational changes or new resources/personnel are required.
Decide what data you need
Data acquisition and preparation is always the most time-consuming and difficult part of any advanced predictive analytics project. EMR systems are mostly closed and data from different sources and repositories use different labels and metrics for the same measurements. For example, reports from different laboratories may use different formats, scales and nomenclatures.
Before you begin, think through whether you need immediate ROI for a specific project or a more complete analytics solution. A project-specific approach allows you to go after low-hanging fruit using data that are easiest to get and integrate; a longer term approach, to support diverse projects, requires building a robust general data-analysis warehouse with a Master Patient Index, terminologies and translation logic, and incorporates adapters to allow integration with relevant data sources.
Finally, there is governance, though ideally this would come first. Often overlooked, governance is important for two major reasons. First, many projects initially succeed, but then fold after the project champion departs, leaving nobody who knows and understands how it all works and where the data are. Second, regulatory oversight and scrutiny will become important when analytics affect real patient outcomes.
A role-based system with lifecycle management, version control, audit logs, approval processes, etc., will solve the issue of departing champions as well as the need to document how predictive models were built, validated, approved and implemented. Good examples of this can be found among pharmaceutical and medical device manufacturers, which have for years incorporated mature governance features to meet these challenges.
When advanced analytics projects fail and smart professionals decide not to leverage smart technology to improve outcomes, the cause is often project leaders who ignore critical steps when planning and implementing such systems. Future healthcare will inevitably rely greatly on advanced predictive and automated analytics to help health care professionals produce better patient outcomes more reliably and effectively. Getting it right from the start will create that future faster and benefit everyone.
I’ll be at HIMSS, April 13-15 in Chicago, and leading a tweet up discussion on the future of population health management on April 15 at 11 am. I look forward to hearing your thoughts on this topic.
About the author
Dr. Thomas Hill is Executive Director for Analytics at Dell’s Information Management Group. He joined Dell through the acquisition of StatSoft Inc. in 2014, where he was Senior Vice President for Analytic Solutions for over 20 years, responsible for building out Statistica into a leading analytics platform. He was on the faculty of the University of Tulsa from 1984 to 2009, where he conducted research in cognitive science. Dr. Hill has received numerous grants from the National Science Foundation, National Institute of Health, Center for Innovation Management, Electric Power Research Institute, and other institutions. Over the past 20 years, his teams have completed consulting projects with companies across various industries in the United States and internationally. Dr. Hill is the (co)author of several books, most recently of Practical Text Mining and Statistical Analysis for Non-Structured Text Data Applications (2012) and Practical Predictive Analytics and Decisioning Systems for Medicine (2014).
At the risk of aging myself, I’ve probably attended 70 or so TDWI conferences and executive summits, but the recent TDWI Las Vegas was different. It marked the organization’s 20th anniversary and reinforced the importance of analytics, which was a big topic among attendees, speakers and vendors.
As a TDWI faculty member for the fourth consecutive year, I taught a whole-day class on social analytics, which focused on driving business values with big data. But more on that subject in my next post. In this blog, I’d like to step back and address changing analytics dynamics.
This vital area has come a long way from its roots in data management, reporting and BI. I reminisced with TDWI president Steven Crofts about all the changes that have taken place over the years. It was fun to remember when TDWI, aka The Data Warehouse Institute, was about innovations in data processing and warehousing. Fast-forward two decades: We’re talking about big data, social analytics, machine learning and cognitive computing.
The same quantum leap holds true when talking about analytics, which has evolved into highly automated, somewhat transparent solutions for ingesting, integrating and leveraging vast amounts of information. As analytics mature, organizations of all types are looking at how they glean greater business value. Some industry segments, like pharmaceutical, manufacturing and retail, are ahead of the curve because market disruption has led to adopting new analytical capabilities and advanced workloads to produce real-time fraud and risk analyses as well as quality control and other complex insights.
Many companies I spoke with are in the midst of retooling their analytics foundations to be more successful. Others are making bold moves. The manufacturer of an industrial French fry maker is using sophisticated analytics and sensors to better monitor equipment vitals (e.g., the filter is dirty, heating element isn’t hot enough, etc.). In doing so, they can assure their restaurant customers that their equipment is performing optimally.
It doesn’t stop there. They also want to listen to the social signals of their customers’ customers—the folks eating fries—to better understand satisfaction levels through trend analyses. If they learn through social analytics that customers complained about substandard fries at their customers’ establishments, they could proactively help the restaurant take action. As a result, this company will be able to differentiate themselves by helping their customers before something hurts their brand. How cool is that?
Regardless of where companies are in the analytics adoption curve, there are major drivers accelerating change across the entire data landscape. As business analytics mature, there will be continued movement along these four pressure points:
New types of data can serve a wider community of users who will want to mix, match and mash up information to yield value in responding to business needs. This will lead to a more mature mantra, from any company looking to innovate: “Put the right data, on the right platform, at the right time, for the right workload.”
In my next post, I’ll offer more insights from TDWI by sharing experiences and more real-world examples from my class, “Social Analytics: Driving Real Business Value with Big Data.” Until then, drop me a line at Shawn.Rogers@dell.com to share your mantra for business analytics.
Innovative companies are adopting advanced analytics to take action and match the speed of their business. This is especially true in the world of manufacturing where complicated process driven activities benefit greatly from smarter, faster analytic insights and actions. Collecting and analyzing process data from sophisticated manufacturing processes requires a flexible and agile infrastructure that supports a wide variety of disparate data sources often spanning sensor and machine sources that are combined with instrumentation and testing data, machinery and production data, customer and market data, supply chain information, 3rd party benchmarks and a wide assortment of system data.
In the case of Pharmaceutical manufacturing systems can include laboratory information management systems (LIMS), manufacturing execution systems (MES), enterprise resource planning systems (ERP), supervisory control and data acquisition systems (SCADA) and last and perhaps most difficult to manage and integrate; paper documentation. Bringing this data together in an action oriented manner requires accurate planning and solid project management principles.
Sanofi Genzyme the 3rd largest pharmaceutical company in the world is embracing this challenge and utilizes the following criteria when planning process driven analytic projects.
Genzyme has built an agile architecture named APEX to bring data together to support advanced analytics. APEX is designed to bridge decentralized, heterogeneous data sources and provide a centralized, secure and validated data layer for analytics. Genzyme recognized early on that they needed to leverage real-time data as well as process oriented data in order to get the analytic insights they desired, APEX accommodates these functions through the use of a traditional data warehouse working in tandem with a process oriented data store. Genzyme also took into account how important data lineage, integrity and validation are to their compliance initiatives and maintains strict control over all data sources throughout the analytic process.
The framework that Genzyme has developed creates value across the company as it supports a repeatable and scalable environment designed to meet the evolving needs of their manufacturing processes. In the end, the following themes are crucial to the success of this and future projects.
For more details on process driven analytics for manufacturing and the Genzyme story, watch this webcast: "Business Analytics in Regulated Manufacturing".
Statistica users > Make your voice heard in Dresner's 2015 Wisdom of Crowds® survey by March 20 and you may qualify for a complimentary copy of the study findings.
We invite you to help represent Dell Statistica in this 6th annual examination of the Business Intelligence (BI) marketplace that covers BI deployment trends and related areas including cloud BI, collaborative BI, advanced and predictive analytics, cognitive/artificial intelligence, data storytelling, and an all-new section on enterprise planning. Statistica users in all roles and throughout all industries are invited to contribute their insight, which should take approximately 20 minutes. Take the survey yourself and share the link with colleagues and customers: www.dresnersurvey.com/TR8DLYN. We appreciate your support and thank you for completing the survey by March 20, 2015.
by Danny Stout
Manufacturers have a long history of successfully employing data - big data - to help make important and insightful business decisions. According to a recent article in CMS WiRE by Joanna Schloss, a subject matter expert specializing in data and information management at Dell, early adoption means the industry is set to be a primary benefactor of the big data analytics boom.
Schloss submits that as an early adopter of big data, with a ubiquitous presence in society, and unparalleled access to data collection, the manufacturing industry has a plethora of new revenue streams available to it. In her article, she outlines three:
The potential of big data is now a reality for every industry. Manufacturing just happens to be positioned to immediately begin delivery and reaping the rewards.
You can read all of Joanna's insights here.
by Uday Tekumalla
Predictive analytics are used by companies for everything from customer retention and direct marketing to forecasting sales. But at the University of Iowa Hospitals and Clinics, predictive analytics are serving a far more noble purpose - to decrease post-surgical infections.
By utilizing a number of different data points that were gathered from 1,600 patients, each of whom has had colon surgery performed at the University's hospitals, the medical teams have dramatically reduced the number of patients inflicted with post-surgical infections. In fact, over a two-year period, those infections were slashed by an impressive 58-percent.
That is an impressive feat. There are, after all, a multitude of variables that can lead to an infection. This analysis considered several different data points - patients’ medical history, data from monitoring equipment, data from national registries, and real time data collected while the surgery is being performed like blood loss, wound contamination, etc. The University built predictive models using Dell Statistica predictive analytics software to achieve these impressive results. Running this analysis allows the hospital to determine a patient's risk level for post-surgical infection, providing the medical team with clear insight into the medications and treatment plans to employ going forward to minimize the risk of infection.
Along with providing better patient outcomes the University of Iowa also has likely reduced medical costs. This is an exciting example of the potential of predictive analysis. Learn more about the university's results here.
Whittaker finds it increasingly difficult to talk about analytics without also talking about the role of cloud.
In her latest CMSWire article, Schloss describes three reasons why manufacturers are uniquely poised to be primary benefactors of the big data analytics boom.
Rogers also reacts to the recent Enterprise Management Associates (EMA) research study, reflecting on the competitive advantages that can result from “cloud first” thinking
At Dell’s recent Big Data 1-5-10 event, I kicked off my introduction by saying my goal is “to help customers use 100 percent of their available data all the time.” This remark caused a few heads to turn, and later prompted Jeff Frick, GM of SiliconANGLE and host of theCUBE live interview show, to ask me for more insight into what he called a “provocative statement.”
Shouldn’t we all be driving toward collecting, analyzing and utilizing data to its fullest? As I explained to Jeff, we’re nowhere near ready to deliver all the data, all the time, but we need to make steps in that direction so we’ll be ready to clear the hurdles and take full advantage of opportunities as they become available.
Technology is still siloed, unfortunately, which makes it difficult for people to build out all the analytical models today that can deliver answers to their most critical questions. Structured and unstructured information isn’t analyzed together, which creates another barrier to getting one single view of the truth. Another barrier: people doing the analytics address very specific, often narrow areas of focus.
Currently, most companies use only a subset of their data for a very specific purpose. But, you can discover so much more if you step back and take a larger view. For example, instead of only looking at revenue trends over the past 12 months, what could be learned if you look more broadly at the health of your company’s customer base or the social factors driving trends and behaviors that either accelerate or moderate a drop or move in your business?
Delving deeper into the data delivers so much more insight. At the University of Iowa Hospitals and Clinics, for instance, Dell Statistica is used to pull data from a wide variety of data sources to help lower the rate of infection for surgical patients. As reported in the Wall Street Journal’s CIO Journal, the University of Iowa takes information from patients’ medical records, and surgery specifics, such as patient vital signs during operations, to predict which patients are face the biggest risk of infection.
Armed with this valuable insight, doctors can create a plan to reduce the risk by altering medications or using different wound treatments.
Thanks to the evolution of analytics, other organizations will be able to follow University of Iowa’s lead in more fully utilizing their data. We’re at a tipping point—compute cycles now are affordable enough and can keep pace with data proliferation while plentiful bandwidth and cloud services make ubiquitous data access a reality. Today’s infrastructures enable us to do things that weren’t possible five years ago.
While environments now are ready to accommodate a more holistic view and broader conversations about data, most companies are just starting to buy-in conceptually. Sure, companies want access to all their data, all the time, but most folks I speak with see this as an aspirational goal still to be achieved. When it comes to the here and now, they’re pretty pragmatic and taking the first steps to realizing their data’s full potential.
Since focus is the hallmark of success, I recommend putting customers first. Start by taking all the steps you can to get all the data on your customers. Then, gather all the data on your product areas, supply chain, manufacturing, etc. In each respective area, there likely will be a dozen different data sources that are interconnected and interrelated. For instance, in compiling data on customers, you’re likely to encounter exposed interfaces that take you to product, which can be integrated with manufacturing, and so on. It’s kinda like assembling LEGO blocks or deciphering fractal patterns as all the data elements are nested and interwoven.
Another major step is determining how best to empower your data analysts by providing them with the right tools for producing everything from simple reports and visualizations to complex analytics. But don’t stop there. If your data is locked away and only useful for PhD modelers and data scientists, you’ll only solve part of your problems. Getting data into the hands of your subject matter experts and line-of-business decision makers is crucial because they too must be empowered to build their own analytical models.
The day when employees become their own data analyst isn’t too far out on the horizon. Once everyone has access to all the data, all the time, they can create their own hypotheses. Training your employees to think more analytically is something every organization should already be doing to stay ahead of the curve.
What steps are you taking to ensure your company gets the most from all its data, all the time? Drop me a line at firstname.lastname@example.org to exchange ideas on how to unlock the power of your data.
Why do analytically driven companies adopt cloud? It seems like a pretty straight forward question and more often than not most people assume the answer is centric to the economics of cloud. From the very beginning cloud has had a reputation for being a cheaper alternative to traditional on-premises solutions. It’s not an absolute truth in every implementation but cloud can often deliver an economic upside in comparison to on-premises solutions. Cloud helps companies avoid the risky capital investment often necessary with IT projects.
Not surprisingly IT and Business users see advantages to cloud through different lenses and they drive growing adoption of cloud for analytic use cases across the enterprise. A recent research study executed by Enterprise Management Associates (EMA) determined there was a wide variety of drivers and influencers behind cloud adoption in the enterprise. (Click here for a free summary of the research.) In companies 500-5000 employees in size “team leaders” were mentioned as the most likely to control cloud based analytic budgets and these executives engaged cloud solution as an operational expense, side stepping the more traditional capital expenditure that is commonly part of IT projects.
The EMA chart below details the business drivers for cloud analytic adoption and as you can see monetary concerns are overshadowed by the need to implement analytic projects faster, create a more adaptive and flexible analytic environment. As IT projects come under greater control and sponsorship from the business these cloud advantages are helping IT match the demand of their end users and move at the speed of the business.
The technical drivers for analytic cloud workloads differs greatly from the business drivers the line of business (LOB) is most interested in speed and flexibility while technical drivers for cloud adoption look to improve data security, technical agility and improved software availability according to the EMA study.
Regardless of technical or business drivers cloud based analytics is a fast growing solution area for smart innovative companies. Workloads like sales analytics, risk management, marketing analysis are all serving critical needs within companies and leveraging the cloud and analytics to create competitive advantage over slower less agile competitors. The EMA study examined 831 cloud analytic projects and found that 41.5% of the responding companies had 3-4 cloud projects underway. If your company remains tentative towards cloud analytic adoption the data indicates you are falling behind the market trends and perhaps missing a critical opportunity to improve your company’s performance.
This blog will be where members of the Statistica team--a dedicated group of trouble-shooters, project managers, subject matter experts, sales engineers and thought leaders--can freely connect with each other and with members of the broader Statistica user community.
We are excited to share our thoughts about issues and trends and products within the dynamic realm of statistics, business analytics, big data, predictive solutions, and information management. This community is also intended to facilitate your use and comprehension of the award-winning Statistica analytics platform, through the eventual sharing of blogs, webcasts, media, whitepapers, tools, solutions, and more. We appreciate your patience as we roll out relevant content here, and we recommend you visit the excellent resources already available via the Statistica product page. Meanwhile, come back and come back often to learn what’s new in the world of Statistica.
Or better yet, join the community if you’re not already a member and visit our main blog page to email subscribe to this blog. (The link is under "Options.") Feel free to give us feedback and suggestions by posting your comments on a blog post or a how-to wiki article. We want to provide tools and solutions that are practical and helpful for you.
-- The Dell Statistica Team