Part 1 of a 3-part series on Transformational Analytics
Transformational: A powerful word that at a high level can be defined as "to cause important and lasting change".
This definition can certainly be applied to data analytics as well: leveraging data to drive impactful, lasting change to your business strategy and execution.
In Part 1 in our series on Transformational Analytics we will explore that this really looks like in practice.
What does Transformational Analytics look like?
In the Analytics world, Gartner describes transformational organizations as those where:
Business people routinely and easily obtain information that they can customize, which helps them run and continually improve the business.
The scope of Transformational Analytics is perfectly summarized in this one sentence... So, let's unpack it and highlight the FIVE(!) key points:
Business people: not data scientists or data analysts
Routinely and easily: independently via self-service
Customize: analytics that is relevant at an individual level
Run the business: empower decision-making with trusted insights at the speed of the business
Continuously improve the business: Drive business strategy
This is a powerful mindset - one that is radically different than the "old school" BI Maturity model:
This traditional BI Maturity Curve does a good job in clearly and concisely describing analytics CAPABILITIES.
Thanks to its simplicity and easy-to-grasp concepts, it has often been used as the "go to reference" in driving analytics strategy, hiring, organizational structure, and IT spending.
However, it clearly does not provide the richness of what a successful analytics organization looks like.
As a result of its one-dimensional view and focus solely on analytics capabilities (instead of the benefit to business teams), it has become the hidden root cause behind many under-performing analytics teams - driving actions such as:
Focusing on technology capabilities, and the corresponding hiring, to drive the analytics strategy (instead of focusing on business needs and benefits)
Hiring a data science team before the organization is ready to generate or even properly consume information generated from AI/ML
Implementing an organization structure that results in analytics silos (where the data scientists work separately from the data analysts)
In Gartner's Analytics Maturity Model, Transformative Analytics organizations are considered "Level 5".
Rather than representing the maturity levels as a curve, a better visualization is one that emphasizes that each lower level forms the foundation for the next level. This reflects the evolution process for analytics organizations and highlights a key principal that Transformational Analytics will not succeed without each underlying level being solid and stable.
In this model, the levels are:
Basic: ad hoc reporting, generally from spreadsheets or production systems (eg. SFDC) with limited business value
Opportunistic: Data is aggregated from multiple source systems that provides value to individual business functions
Standards: There exists a structured approach that enables analytics to span across business functions
Enterprise: Analytics leverages an enterprise framework to deliver consistent, predictable, and trusted insights that results in meaningful value across the entire organization
As you can see, each level depends on the "lower" level to be successful. It would not be possible to skip directly to an Enterprise-mature analytics function.
With a sense of what Transformational Analytics LOOKS like, questions will arise about how to get there and the amount of time (and money. and people.) to make this happen. Is it worth the expense? Is the timing right? Are our existing analytics capabilities not already good enough?
These are all fair questions. Be prepared to answer them. To help with this, Before embarking on the journey, let's take a look at some considerations and best practices for success - and debunk a myth or two along the way.
Consider: It is a journey. One that in all likelihood will not take you on a straight path from Point A (where you are today) to Point B (where you want to be).
However, there will, rightly so, be an emphasis from leadership on producing results quickly. The critical consideration here is that the plans must be "agile"- to deliver smaller chunks of results rather than a "big bang", and allow for adjustments as needed along way to accommodate changes in business needs.
Quarterly deliverables - what we like to call "rolling thunder" are a must. As is a quarterly readout to revisit and fine-tuning (not necessarily redo) the plan.
Consider: It is very easy to get caught up in the technology or capability hype that you will hear and see at conferences and in articles/Webinars. Transformative Analytics does NOT necessarily mean a powerhouse data science team and systems delivering prescriptive analytics.
Looking back at Gartner's definition:
Business people routinely and easily obtain information that they can customize, which helps them run and continually improve the business.
Nothing there about capabilities or technology solutions. It is about improving the business. This can be often be done through "traditional" BI capabilities and proper processes for business users to consume, customize, and action the insights.
Of course, for many organizations, data science, predictive models, and prescriptive analytics will be required to make meaningful improvements to the business.
Consider: Not every organization is the same. Your business vision, strategy, goals, culture, budget, data mindset, org structure and current priorities are different than any other organization. So the end-state and success criteria will, by definition, be unique to your organization. Do not rely solely on what you read and hear - there are no cookie cutter solutions. Make it your own! Just make sure it will lead to analytics that transforms your business.
Consider: Your organization needs to be mentally ready: if your business strategy, goals, culture, data mindset, or org structure are too murky or in a constant state of flux, the appetite for yet more change may not be there. Use your judgement: it may be best to let things settle before embarking on the journey to avoid constant struggles to gain traction and in executing on the plan.
Best Practice: The journey must be business-driven. Too often, such an effort is technology or solution driven. There should exist a business-facing function (regardless of where it sits in the organization) to ensure the end result meets business needs/goals and that proper change management happens. If not, the journey will face serious adoption challenges.
Best Practice: The journey must be an organizational-wide effort, with a clearly defined owner. Its span must include operations teams, field teams, product teams, finance, support, customer success, and IT. Ownership - and budget - must be centralized and ideally be separate from any of the stakeholders to minimize conflicts of interest.
Best Practice: C-level sponsorship, awareness, support, and commitment (of time, money, and influence) is absolutely critical.
Best Practice: Set a reasonable and realistic timeframe - as well as scope - for your journey. Your plan should not be less than six months (the minimum amount of time to make significant transformations to your analytics maturity). Nor should the scope be so large that it will take more than 24 months to execute. Scale the plan to ideally be in the 12 to 18 month range with clear expectations of where you expect the analytics maturity be at that point. At which time, you can revisit the strategy and plan for continuing on the journey.
Myth Buster: It does not have to impede, stall, or derail existing analytics efforts. In fact it MUST not do so. At first glance, it might seem like the equivalent of rebuilding the engine in your car while driving down the freeway, However, as we will explore in Parts 2 and 3 of this series, with proper planning, change management, and communications, keeping your analytics activities continuing to deliver business value IS entirely possible.
Myth Buster: It does NOT have to be expensive to define and implement a "business-need appropriate" level of Transformational Analytics. It certainly can be, and quite honestly often is. However, even an unlimited budget does not guarantee success. Significant gains can be had with reasonable expenditure on people and technology. Building a strong plan with an appropriate budget and full transparency on realistic expectations is a major key to success.
Comments