A robust data analytics operation is necessary for healthcare systems’ survival. Just like any business, the analytics enterprise needs to be well managed using the principles of successful business operations.
This article walks through how to run an analytics operation like a business using the following five-question framework:
1. Who does the analytics team serve and what are those customers trying to do?
2. What services does the analytics team provide to help customers accomplish their goals?
3. How does the analytics team know they’re doing a great job and how do they communicate that effectively to the leadership team?
4. What is the most efficient way to provide analytics services?
5. What is the most effective way to organize?
Download
According to the Small Business Association (SBA), 30 percent of new businesses fail within two years of operating. Fifty percent fail within the first five years, and 66 percent fail within the first 10. Hundreds of books and articles have been written about the lessons learned from these high failure rates and the principles that, when followed, increase the likelihood of success. These lessons and principles are equally relevant when establishing and growing an analytics operation within a healthcare institution.
A robust data analytics operation is no longer optional for healthcare organizations; it’s necessary for their survival. While most health systems have invested heavily in their analytics operation in both time and money, many aren’t getting the kind of ROI they need. However, underinvestment in an analytics operation is more common and problematic than over-investment.
In many health systems’ analytics operations, there are many overtaxed resources that can never satisfy the demands placed on them by the organization. Analysts have long queues of requests for data and reports. Then, data consumers frequently have complaints about the data – it’s often incorrect, irrelevant, or takes a lot of back and forth to get what they want. Customer satisfaction among healthcare executives and clinicians is often low because the data they get isn’t timely and isn’t contextualized in a way that allows them to make important decisions.
For an analytics enterprise to be successful, it needs to be run like a business. And, just like any business, it will succeed or fail based on factors like customer satisfaction, profitability, return on investment, and potential for growth. This article will walk through how to run a healthcare analytics operation like a business using the five-question framework below (figure 1). This framework was developed by Eric Denna, Professional Services Senior Vice President at Health Catalyst, and is used by healthcare data analytics teams across the country. Analytics directors who do this successfully have the biggest impact on healthcare outcomes and ensure both their team’s and organization’s survival.
Who does the analytics team serve and what are those customers trying to do?
In order to provide products or services in the most effective way, any good business has a keen understanding of who their customers are and what they want. Similarly, the first step in running an analytics enterprise like a business is to define the customer.
In most analytics operations, this will include several different types of customers, such as executives and frontline leadership, clinicians, and improvement teams. For this reason, it may be helpful to segment the customers into different categories according to their needs. In a typical healthcare analytics operation, that segmentation might look something like this:
This is not a comprehensive list, but it’s a good example of how to define and segment customers of the analytics operation. The key is to ask not only who are they, but what are their goals? Answering these questions will help with the second step in running a healthcare analytics operation like a business: determining the best way to provide the service.
What services does the analytics team provide to help customers accomplish their goals?
Any good business has a plan, complete with a clearly defined offering. An important part of that plan is a roadmap that shows how to systematically improve or grow the business.
An effective analytics strategy also relies on creating this roadmap in order to clearly define the offering and the way forward. There are several important components of the road map as it relates to the analytics strategy that the team will need to think through and determine:
While these are just some examples of issues the analytics team needs to think through to define their offering, Mike Noke, a Senior Vice President in Professional Services at Health Catalyst, explores each of these success factors in his article on driving strategic advantage through widespread analytics adoption.
How does the analytics team know they’re doing a great job (and communicate that to healthcare leadership)?
Once a business is up and running, the next important step is measuring success. How does a business know if it’s growing or if it’s plateaued? How is success quantified? Just like a business owner, an analytics director will need to determine what defines success in a healthcare analytics operation and how to measure it. And be warned, if success isn’t defined clearly and confidently by the analytics director, someone else will do so. A few common methods for measuring success are listed below.
Customer satisfaction is the cornerstone of success for most businesses. Unfortunately, many analytics directors are not focused on this metric because they are not thinking like a business owner. Too often, analytics teams are buried in work, and are more concerned with keeping their heads above water with requests than delivering timely, actionable data to the individual teams and customers they are serving. To ensure the team’s success, it’s important to ask not only “How happy are people with the service?” but also, “How do I know?”
An August 2018 Gartner article focuses on this very topic: why and how to make customer success a focal point of the healthcare analytics operation. However, despite the value customer satisfaction can provide, many analytics teams fail to approach or measure it in a consistent manner, making the path to improvement unclear. Doing so includes knowing how satisfied customers are in a measurable way, whether they are asked to fill out a satisfaction survey each time a request is filled, or another system that produces quantifiable results.
In a technology-enabled services enterprise, the people are the most important asset, hands-down. Do you know how engaged, how loyal, how satisfied your team members are? Do you proactively measure and monitor this and are you open to make difficult adjustments when this data reveals challenges?
Another possible measurement for how well a business is doing is measuring the number of customers. Is this number going up or down over time? Keeping track of these simple statistics can help analytics directors know how they’re doing or if they need to make changes.
Unlike most businesses, an analytics enterprise is not receiving a payment for the services they’re delivering, so defining profitability may take some creativity. The expenses of an analytics enterprise are visible to the CFO, but what about the value provided? When it’s time to cut costs and the only thing visible is the expense, analytics teams are likely to see cuts if they can’t effectively communicate the value they’re providing to the organization.
Results With Improvement (RWIs) can help measure things such as patient outcomes, increased efficiency, and cost savings in order to provide a look at the true financial impact of an analytics enterprise. How many patient’s lives were impacted? Did the team help reduce readmissions? Did they help cut costs from supply chain? Did the team get more impactful work done in less time?
Analytics directors need to have a vested interest in defining upfront what success looks like and then capturing and recording the results. Not only is it necessary to measure success, it’s also necessary to articulate and communicate the value generated.
For example, the team might share their successes in a measurable, succinct way: “This year we saved the organization 15 million dollars, we saved 10,000 lives and we impacted 100,000 patients.” Being able to show the true significance of an analytics enterprise helps ensure the operations’ viability and growth. Ask yourself, is your operation a cost to the business or an investment?
What is the most efficient way to provide analytics services?
Successful businesses must determine the best way to provide their services to customers. This includes defining both the intake and fulfillment process in order to operate efficiently and effectively. A healthcare analytics operation also needs a clearly defined process that is transparent and easy for customers to use.
Defining the intake process is an important first step in ensuring an effective business operation. This might include answering questions such as, “How do customers interact with the business?” “Do they shop online for services or show up at the doorstep?” “Do they know the intake process?”
Answering these simple questions in order to define the mechanism for analytics access will help improve the intake process and the expectations for both the team and customers. Referring back to Step 1, depending on the customer, they might need a tour guide, or they might just need access and training as a self-service user.
The second component of this step is defining the fulfillment process. Let’s say a business selling widgets had more demand than they could satisfy at any given time. How would the business prioritize those customers? Is it simply a first-come, first-serve basis, and if so, is that the best way to operate?
In a free market, price and a willingness/ability to pay play a critical role in determining who gets access to services and products and in what priority order. So, if your customers aren’t paying to utilize your products and services, what is your mechanism for prioritizing? What plays the role of price in your business? In the business of analytics, the value exchange works a little differently. The payment you collect is the opportunity afforded to you for value creation. The higher the opportunity for value creation, the more appealing the customer. One customer might come in with a $200,000 problem and one might have a $1,000,000 problem. Realize, however, that these problems don’t always have a price tag that can be defined in dollars and cents. Sometimes patient safety is on the line. Other times, regulatory penalties or institutional reputation is on the line. Or, patient and team member satisfaction may be the goal. The value of these efforts may not be easily quantified. As such, it will be important to establish decision-making rights and process through effective governance (discussed briefly later and in various other articles produced by Health Catalyst).
Defining the process for prioritization is essential for success and goes hand-in-hand with the ability to measure success. For instance, if an analytics director can say he has both a $1,000,000 problem and a $200,000 problem in the queue, but not enough staff to tackle them, he can then take that information to the executive team in order to ask for more resources.
What’s the most efficient way to organize?
Most businesses spend a lot of time and energy on finding and keeping the right team. Before they do that, they need to know how many people they need on their team and what the functions of each position are. For analytics operations, defining the organization in order to maximize effectiveness is critical for success. Deciding how to do this is open to a lot of debate.
One question to start with is how specifically to define the expertise on an analytics team. This could include data modeling skills, data science and predictive analytics, statistical modeling skills, etc. An analytics enterprise might organize teams around each of those areas of expertise and each team would provide their expertise separately. This model is not very efficient. On the other end of the spectrum, each analyst would be a generalist on one large team and over time, each analyst gains expertise in every aspect of specialization. This model would be ideal if it were possible to hire and retain these people. The reality is likely somewhere in between.
It is recommended to err on the side of generalization and only specialize if and when there’s enough demand for that specialization to be self-sustaining. For example, if there’s a team of five data analysts and one data scientist, and one out of every 20 requests involves a data scientist, then the demand for that data scientist is going to periodically spike while the rest of the time he will experience a lull. During those down times, the business still has to pay his salary and all the associated expenses.
After deciding how much to specialize, the next step is to right-size the team. The organization model might change over time depending on the segmentation of customers and the categorization of demand coming from those customers. If and when there’s a steady demand for machine learning, then it makes sense to hire a data scientist to fulfill those requirements.
As part of defining the organization, it’s important to establish stakeholders that will approve funding, growth, and strategy. Just as many businesses are responsible to their investors and a board of directors, an analytics enterprise may also benefit from an invested group of stakeholders, such as an Information Management Executive Council.
I have previously written about this very topic in my article, Governance in Healthcare: Leadership for Successful Improvement, the first principle being to engage the right stakeholders. An effective data governance strategy includes both data stewardship and establishing the leadership needed to get the most out of an investment in analytics. Because leadership buy-in is so important, it’s important to tap the right stakeholders to form the healthcare data governance committee who will play key roles in establishing, growing, and sustaining a successful analytics program.
In 1988, Michael E. Gerber released the best-selling book, The E Myth: Why Most Businesses Don’t Work and What to Do About It. Today, the book is still relevant and popular because the basic principle is true: just because someone is good at something doesn’t mean he can turn that into a successful business. The reality is that most people are not good businessmen.
In looking at creating a successful analytics enterprise, an important consideration is who to have in the leadership and accountability role for this business. That person isn’t necessarily a great technician or analyst that moved up through the ranks. The person in this role needs to be a visionary that is closely connected to the customer.
A common downfall in healthcare organizations occurs when analytics falls under the accountability model of the CIO, and many CIOs are not the visionary that is needed. That’s not to say the CIO is never a visionary, but don’t make the mistake of assuming that is the right person to lead an analytics operation because it’s the most obvious. Instead, be thoughtful about who is in that leadership role, and, ideally that person should be elevated to be a member of the leadership that reports directly to the chief strategy officer or CEO. Ryan Smith, Senior Vice President of Professional Services at Health Catalyst and former healthcare CIO, makes a compelling case for executive sponsorship of a healthcare analytics program as a key strategy for its success. In the article, Smith says, “This strategy involves identifying the right executive sponsor, at the right level in the organization, to provide oversight of an enterprise data management and analytics program.” If run properly, there’s no other asset more valuable to a healthcare organization’s viability than a data analytics enterprise.
Just as many businesses don’t succeed, a healthcare analytics operation that fails to understand its customers, provide services in the most effective way, measure success, or organize efficiently is also likely to fail. Running a successful analytics operation involves many of the same steps for success as running a successful business. This includes understanding the unique needs of each analytics customer, creating a roadmap for success, measuring the success of the operation and communicating that clearly and frequently to hospital leadership, making sure the process is transparent and efficient, and organizing in a way to maximize effectiveness. Taking these steps helps ensure a successful analytics operation that can improve healthcare outcomes, save the organization money, and impact patient lives.
Would you like to learn more about this topic? Here are some articles we suggest:
Would you like to use or share these concepts? Download the presentation highlighting the key main points.
Click Here to Download the Slides
https://www.slideshare.net/slideshow/embed_code/key/hY4leQ2gsmrBSz
