The Mindset of Leadership and Data
Driving Innovation through Business Intelligence
Unleashing the Power of Analytics in Academics
AI-enabled business transformation: Closing the gaps
Applying Deep Learning to Streamline Healthcare Administration
Sanji Fernando, SVP of Artficial Intelligence & Analytics Platforms, Optum
Tire Data, IT and Data Science: Revolutionizing the Tire Industry...
Tim Eisenmann, Chief Analytics Officer & SVP of Advanced Analytics, American Tire Distributors
Thank you for Subscribing to CIO Applications Weekly Brief
Approaches for Growing Your Organization's Analytic Capability
By Rich Huebner, PhD, Director, Data Science & Data Architecture, Houghton Mifflin Harcourt
Structure of Analytics Work
Consider how analytics work (whether this is data visualization, reporting, data modeling, data science) is structured. We have options on whether we structure the analytics work within a single centralized department, or have analytics work conducted in a matrix or decentralized structure. Choice of the structure largely depends on the type and size of the organization. Centralizing the analytics work can result in efficiencies and greater ability to standardize analytics work within the organization. Several organizations that I’ve worked with have established Analytics Centers of Excellence (ACoE), and I liken these to the project management’s version of a PMO – or project management office. Structuring analytics work in a centralized way may allow the organization to move the needle forward in terms of capabilities. In one company, each analyst within the analytics department was responsible for working directly with business unit leaders. For example, analyst Joe was assigned to work with the Marketing department and focused exclusively on marketing analytics. Before long, Joe became known as the in-house expert in this area.
Staff must see data and analytic capability as an organizational priority. If they don’t, they will focus their energies on other tasks
Benefits of having Joe work alongside other analysts within the analytics department allowed him to share this work with others, bounce ideas off other analytics professionals, and help find ways to standardize analytics processes.
How capable is the organization with regards to analytics? In other words, how “good” is the organization using analytics? One way to assess this is by using an Analytics Maturity Model (AMM). There are several AMMs available. AMMs a) assess current analytic capability, b) provide insight as to next steps and goals, and c) benchmark the organization compared to others. Most AMMs have either four- or five stages, with stage one indicating an apparent lack of data and analytics.One of the first steps for stage one companies is to get a handle on data and begin using it to improve business operations. Another critical step is to ensure the organization has integrated any data into a centralized data warehouse or data lake.
Conversely, stage five characterizes organizations that are well versed in analytics, and typically have many analysts, data scientists, and BI professionals, and are focused on advanced analytics and competitive models. Analysts in Stage five companies often work closely together and drive projects toward company objectives. Once you’ve determined which stage your organization is in, you can develop strategies and tactics for improving analytics maturity across the enterprise.
Creating a Culture of Data Exploration and Analytics
Staff must see data and analytic capability as an organizational priority. If they don’t, they will focus their energies on other tasks. Setting organizational priorities is left up to a senior leadership team, and communicating the importance of analytics is critical. Research shows evidence oftop leadership teams having a profound influence on analytics and data science projects. As leaders, we need to play the role of project champion for analytics. I have coached several leaders around showing enthusiasm for analytics. If you can start to demonstrate the possibilities of analytics (through using visualizations), people begin to get excited about the possibilities.
I know several VPs and CIOs that have rolled up their sleeves, created some data visualizations, shared them with staff, and have been successful in getting people hooked. Getting staff involved in analytics means allowing them to experiment a bit. At one former company, we placed Tableau Desktop in the hands of HR staff, who subsequently went to town ingesting data and created some very insightful visualizations that were then used to make some changes in their recruitment and selection process. Staff will inevitably share their work with others when they realize the benefits visualizing the data can bring. It’s also essential for staff to see leaders getting involved in some of the analytics project work – either by helping monitor project progress, providing expertise and suggestions, and removing any roadblocks to success.