[et_pb_column type=”4_4″][et_pb_text admin_label=”Text”]With the development and growth of big data technology, the ease and cost-effectiveness through which companies can acquire and store data is high. With this ease comes the opportunity for companies to become more data-driven in their decision making. Data also helps companies compete on analytics, whereby the insight they glean enables them to differentiate themselves from competition and battle more effectively in the marketplace.
How can companies be more data-driven?
To answer this is to understand how analytics are applied in an organization. At a company macro level, internal analytics are used to measure effectiveness of strategies and progress towards business goals. Business goals and strategies must be decomposed into an analytics strategy that defines an agreed upon approach for how analytics programs are defined and executed.
To be more data-driven companies need to get value from their analytic programs.
has a process and framework for how companies get value from analytics programs to achieve the highest objectives of the business. The process and framework is represented through the Analytics Value Pyramid
, which breaks down the value components into segments, where each segment is fulfilled towards the highest value possible.
How the Analytics Value Pyramid Enables High-Value Analytics Programs
The highest value to achieve with analytics is towards business goals. Business goals are broad objectives that companies set out to accomplish each year that enable them to thrive and grow. These typically include revenue growth, cost stabilization, customer acquisition and retention, and productivity and efficiency increases.
Organizations decompose goals into business strategies for how they will accomplish them. Strategies vary by industry, company size, product and market.
Without analytics, a company cannot measure the effectiveness of its business strategy. To achieve greatest long-term value, it is well worth the time and effort to define an analytics strategy and decompose it into a set of analytics applications built from a user’s view with the data a company has or can acquire.
Launching from business strategy into the details of data and technology directly is where many companies fail in their efforts to become data driven.
What should you consider in your Analytics Strategy?
- KPI’s you will measure, lagging and leading indicators.
- Decision making analysis: this is closely tied to KPI’s and is the process of taking a business situation and converting it into an analytical use case that identifies beyond the KPI’s what action you will take to correct the business and bring the KPI’s into a range of acceptance.
- Who-Why-What-How: every analytics strategy should include background on who will be using the analysis, why they are analyzing the data, what data is available or can be acquired, and a high level of how they will use technology and engineering to analyze.
- Program strategy: this is the strategy for the technical implementation that will deliver the solution. It decomposes the analytics strategy into one or more detailed technical programs/projects delivered by project teams.
Analytics Applications – right data, right form, right user, right time
From Analytics Strategy, the next segment within the Analytics Value Pyramid
is the analytics applications themselves. One of the most important considerations for analytics applications is that the right data be presented in the right form to the right user at the right time. Analytics applications should be built from the user’s point of view based on the way they understand and consume business information.
Analytical applications include the obvious instruments, like reports, graphs, dashboards, scorecards, all of which should have drill-through capabilities for investigative analysis. Other types of analytical applications include exception alerting and mobility, and user data models, which are models of integrated data from various systems that provide a business user’s view (as opposed to a database physical schema view).
Data and Platform Technology
Data and platform technology are the building blocks for analytical applications and are at the base of the Analytics Value Pyramid
. They provide business value through each of the upper segments. Data without a user application and an analytics strategy doesn’t by itself provide business value towards achieving business goals.
This base segment layer includes:
- Original source data.
- Integrated data from multiple sources residing in a data warehouse or data lake
- Platform and visualization technology that stores integrated data, and queries and visualizes it for the user.
- Data feeds and data pipelines – engineered scripts and feeds that transform and pipe data into a data repository for the purposes of analysis through applications.
Continuous Feedback Loop and Information Flow
Rare is the company that takes an exclusive top down approach to analytics. In my experience the practice is a combination of top-down, bottoms-up, meet-in-the-middle, which works because of the top-down component. Companies that are operating purely from a bottoms-up approach expose themselves to the risk of analytics project failure and often produce applications that do not offer enough business value and have less adoption.
Feedback is a critical component of high-value analytical programs and is bi-directional within the Analytics Value Pyramid
. Business value is obtained when feedback from the top-down is incorporated into applications, because it enables the applications to continuously evolve with the business itself. Likewise, feedback up the value pyramid addresses issues with data, like integration and quality issues, and issues with the platform technology itself or how it’s implemented.
Companies wanting maximum value must design and maintain feedback loops
into their process for building and implementing analytics.
There are many variables that affect a company’s path towards business value from analytics and data-driven decision making. It is not necessarily company size alone, it includes process and program maturity and scope of how it wants to use analytics. Purpose specific and departmental analytics programs have shorter speed-to-value and delivery windows than sweeping programs that start from the top of the organization down.
Regardless of where a company lies on the data-driven and analytics readiness spectrum, the Analytics Value Pyramid
provide a framework for how to get the most value in the shortest time possible.
Andrea Amaraggi is Founder and Principal of Cohere Insights. She helps companies leverage technology and engineering to build business focused analytics applications that drive revenue, reduce costs, achieve operational excellence, and create customer successes. Contact her at firstname.lastname@example.org.