The mega-success of Netflix is accredited to the company’s analytics journey. Thomas Davenport and Jeanne Harris, the authors of Competing on Analytics, call Netflix an analytics competitor. The book discusses 5 stages of analytics maturity. These stages help in determining an organization’s current position on its analytical journey.
Davenport and Harris also define a DELTA plus model, which we’ll discuss in today’s post. This model with the 5 stages of maturity serves as a well-known framework across the industries now.
What is DELTA Plus Model?
For an enterprise to achieve a certain level of maturity in its analytics journey, it should analyze 5 elements in the DELTA model.
1. Data
The first and focal element to achieve your milestones in your analytics journey is data. Your data should be quality-oriented, free of errors, accessible, and consistent to start decision-making with analytics. To achieve maturity in data, you need to move from siloed, unstandardized, and essential data to centralized, standardized, and advanced data.
By centralized we mean, your data should be uniform across the departments and is accessible to all. Standardization brings in data governance, security, and integrated good quality data, free of issues and duplicity.
Advanced means bringing in data from all sources—big and small, structured and unstructured. Once you collect data from all the sources, the assessment horizon automatically widens and provides in-depth information.
In a nutshell, basic data utilization with reporting starts with good quality data accessible to all. But analytics start much later with standardization and centralization.
2. Enterprise
A significant step for an organization in analytics maturity is to get rid of islands of data. Each department having its own data management system without sharing it beyond its periphery is a major growth hurdle.
Apart from centralization, enterprises at the matured analytics level also advocate for data democracy. However, data access to all is only one aspect of it. Your employees should understand and use data in decision-making without substantial assistance.
To achieve strong maturity in analytics, enterprises should set a vision and a roadmap to achieve that vision. You should let go of thought processes, which don’t align with the enterprise-wide strategy. Only with a coordinated approach can you get rid of islands of data.
Another important aspect of this pillar is to develop a culture of using analytics in making decisions. And for that, leadership plays a crucial role.
3. Leadership
Analytics-matured organizations show leadership that has embraced a data-driven culture. And such leaders also encourage their employees to acclimatize the same. This trait is not limited only to the top management. Each level bears the responsibility to adapt and encourage their subordinates.
The leadership element goes in tandem with Enterprise very closely. The authors mention these 12 traits of excellent leadership that has embraced analytics:
- Possess people skills,
- Push for more data and analysis,
- Hire smart people, and give them credit,
- Set a hands-on example,
- Sign up for results,
- Teach,
- Set strategy and performance expectations,
- Look for leverage,
- Demonstrate persistence over time,
- Build an analytical ecosystem,
- Work along multiple fronts,
- Know the limits of analytics.
You might like to read about technical analytics maturity, as well.
4. Targets
An organization should have an analytics agenda, which aligns with the organization’s broader vision. The analytics goals are usually a subset of the organization-level goals. And hence, the analytics efforts should go hand-in-hand with the corporate objectives.
Also, it’s nearly impossible to achieve analytics excellence in all business aspects. If you dip your legs in all possibilities—which in no way are limited—you might not float properly in any. So, focus on a few and most desirable use cases.
Leadership augments this pillar of DELTA. Leaders should have a good comprehension of the business and analytics to bring them together in consensus and achieve the targets. They should also vision the long- and short-term goals.
Once an enterprise achieves some level of analytics maturity, its processes get more streamlined and evolved. Eventually, it takes less effort to explore more opportunities and bring these analytics goals in sync with top-level business goals.
5. Analysts
An organization can achieve maturity in its analytics journey only when it has the talents to leverage the expertise and technologies. Agree that matured analytics stages need data literacy and democratization. But without data scientists and data engineers, achieving the vision is next to impossible.
You, as an organization, should have employees capable of working on basic spreadsheet activities. And you should also employ expertise capable of churning hidden information coming from social media, videos, images, and texts.
The book mentions 4 different categories of such experts:
- Analytical champions,
- Analytical professionals (data scientists),
- Analytical semiprofessionals,
- Analytical amateurs.
Amateurs could be the ones with negligible technical skills to understand and use tools and analytics literature. But they often possess the domain knowledge necessary to drive the business. Data scientists are highly skilled in analytics literature, tools, and technologies. These pros, however, might not possess the business acumen. Hence, a tandem between the various professionals is needed. And that’s when champions lead the initiatives.
The Plus to the DELTA
The DELTA model has evolved with time and now includes 2 more elements.
6. Technology
The authors call the evolution of analytics as 4 eras in 10 years. This started with Analytics 1.0 with reports and descriptive analytics and currently running at Analytics 4.0 with autonomous analytics driven by AI and ML.
Since analytics growth has been overwhelmingly fast and refined, organizations need to speedily upgrade their methods, infrastructure, and technologies to keep up with the pace. Big data became mainstream, and AI has brought a whole new gamut of analytics processes into the picture.
“Architectures must support experimentation and flexibility while making it feasible to integrate analytics with production systems and processes,” say Davenport and Harris.
With self-service BI and similar tools, companies rely on citizen data scientists and integrators to take care of many activities, which otherwise need experts. With niched tools and sophisticated technologies that don’t demand refined technical skills, companies can gain an edge with citizen data professionals. But this compels the need for governance and quality at granular data with data literacy and democratization.
7. Analytics Techniques
For many decades, the world was doing the same kind of analysis with data using spreadsheets. Descriptive analytics.
As the book mentions, the growth in analytics has been multifaced in just over a decade. First, with the shift from hindsight to insight and finally to generate foresight (prescriptive and predictive analytics), the technologies have got more nuanced.
Companies have moved from regression analysis and extrapolation to bespoke and sophisticated AI and ML algorithms. These deep learning modules consume all sorts of unstructured data like texts and videos.
As mentioned earlier, self-servicing tools make life easier for nontechnical people. But at the same time, need literacy and backend changes in the organization. The simpler the front end is, the complex becomes the backend.
Before We Wind Up…
The book, Competing on Analytics, also defines 5 stages of maturity in the analytics journey for an organization. We will round up soon on this excellent analytics maturity guide through our blog.
Meanwhile, if you need to assess your analytics journey, do tap us here. We would be happy to assist you.