We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.


Business Intelligence Programs for Healthcare

Originally published August 18, 2011


There are many reasons why business intelligence (BI) and data warehousing (DW) efforts fail. Many of my colleagues want to attribute it to just one thing, but I donít believe thatís true. I believe that business intelligence is an ecosystem that exists within the broader context of your organization. Just like in our environmental ecosystem, if one thing gets out of balance, the rest of the system must accommodate or shift in order to bring things back to equilibrium. If your BI program isnít prepared to do that, then it will likely implode. I have a hundred stories of failed BI programs, and each of them failed for different reasons, but it all comes down to this one simple idea:† All things are related.

To guard against failure, you must take some key steps. In earlier articles, we covered the need for a focus on ETL and data modeling. Now we must cover the other aspects of BI programs such as program management, project management, strategic roadmaps, data governance, user adoption and return on investment (ROI). I canít guarantee you wonít fail if you cover each of these aspects of BI programs, but it is less likely.

Program and Project Management

These two functions are actually quite different, but they are related and in some respects drive each other, so I will cover them together. Program management may not be critical the first day of your BI journey, but at some point you will be running multiple BI projects and it will become very important to have a big picture view of these projects, the resources that are participating and the cost for each. To keep your project managers focused on completing their projects, the program management function should exist at a different level with different resources. The program management function will exist even after you have fewer projects running because at that point you have created a set of BI capabilities that drive the business. While it may not be as large of an effort, it must still be managed and maintained.

Project management best practices can follow what your organization supports. Typical waterfall methods work well for traditional project management, but agile methods work well for business intelligence and data warehouse projects. It is imperative that project management be well accepted within your organization.

Strategic Roadmap

When you need to reach a destination, you look at a map to see how to get there. We do this every day without thinking twice about it. If you have never been to said destination before, trying to get there without a map is a foolís game. Yet, organizations do this all the time with business intelligence and data warehousing programs. Gartner released a research note in November 2008 indicating that much of the failure of BI programs could be attributed to the lack of strategic planning. No one should be surprised when their BI program fails if they have planned to fail by not planning at all. Business intelligence and data warehousing require specialized skill sets and a laser-like focus on business value deliverables. Without it, BI programs will not survive.

Building the roadmap doesnít have to be a six-month long effort. The most important thing you can do is understand the needs of your user base and organize the projects based on tangible business value. If you do that, ROI will follow. Deriving ROI from your BI program is critical, and not as daunting as it may seem. The key is delivering value to business users.

Roadmaps are good for about 18 months Ė†much shorter than that and they are just†really large project plans. Roadmaps lasting longer than 18 months are too reliant on predicting the future of a dynamic market place, which doesnít deliver value to your organization.† You will need to revisit the roadmap about every six months, but be careful to not let this devolve into a continuous roadmap exercise. You wouldnít plan a road trip without having some sense of where you are going; if you are constantly changing your destination (e.g., from New York to Florida), the trip will be more costly and deliver fewer benefits.

Data Governance and User Adoption

There are very few things as challenging as getting people to agree about data. If you donít believe me, ask ten people at your organization for the definition of patient. You will likely get ten different answers. Organizations donít set out to create confusing and conflicting definitions, but often we find that the function of a patient definition needs to vary for clinicians versus finance department. What is the solution?†There is no easy answer. You have to do the hard work of coming to some type of consensus about what your data means for your organization, and that means having a data governance function.

As we see in Figure 1, governance is the vegetation that keeps the soil together (i.e., architecture and metadata), and it is fed by the users and customers (internal stakeholders). The idea of an ecosystem is that no part of the system can exist without the other parts, and governance provides key nutrients to the entire system. You can get started without governance, but users will likely not adopt a system that they donít trust.


Figure 1: BI ecosystem


User adoption is the fuel of your BI program. Trust, ease of use and training are the trinity of user adoption. Users need support and training. Their job is not to run reports or analysis, so asking them to remember certain one-offs of your system on a monthly or quarterly basis is a recipe for disenchantment. Some of your users will easily adapt, regardless of the obstacles, but many wonít. Understanding your user base is critical, and the easiest way to do that is to create a set of standard personas that will help you develop training and support materials specific to their needs.

Simplifying your user interface is also very helpful. Research has shown that users are sensitive to the number of mouse clicks they have to make Ė anything beyond ten clicks and users will complain. Many BI products have folder structures that are tied to security, forcing organizations to bury reports in a deep folder structure for users to click through. Often itís addressed as a ďtrainingĒ issue to help users understand that they have options like favorites, but training must be repeated to gain the full benefit from it.

Key Tenets of Business Intelligence Programs

We have covered the critical aspects of BI programs that can help you weather the storm (trust me, there will be storms). Program and project management help you execute your project, a roadmap is a plan to reach your destination, and data governance helps drive user adoption. You donít have to complete this list in order; many organizations focus on governance first and bring program management on when they have grown to accommodate multiple projects. Regardless of the order, these tenets bring structure and efficiency to a program that can easily spiral into a level of complexity that leaves executives and users scratching their heads.


  • Laura MadsenLaura Madsen
    Laura leads the healthcare practice for Lancet, where she brings more than a decade of experience in business intelligence (BI) and data warehousing for healthcare, and a passion for engaging and educating the BI community.† At Lancet, she spearheads strategy and product development for the healthcare sector. She also works with key accounts across the country in the provider, payer, and healthcare manufacturing markets. Laura is the founder of the Healthcare Business Intelligence Summit, an annual event that brings together top hospitals, insurers, and suppliers in the healthcare business intelligence space. Laura is also the author of the popular book, Healthcare Business Intelligence: A Guide to Empowering Successful Data Reporting and Analytics (Wiley, 2012). You may reach her at lmadsen@lancetsoftware.com.

    Editor's note: More healthcare articles, resources, news and events are available in the BeyeNETWORK's Healthcare Channel featuring Laura Madsen and Scott Wanless. Be sure to visit today!

Recent articles by Laura Madsen

 

Comments

Want to post a comment? Login or become a member today!

Posted August 22, 2011 by Laura Madsen

Thank you for your comments.  I enjoyed reading your blog.  

Is this comment inappropriate? Click here to flag this comment.

Posted August 18, 2011 by wstepalavich@gmail.com

You are so right about all of this.  The "ecosystem" concept is something that I think is lost in the grand scheme of things.  Without all of those different parties, it is nearly impossible to be sucessful.

My longer take:

 

Is this comment inappropriate? Click here to flag this comment.