Photo: @ARK via Twenty20
WRT #digitaltransformation success
In a world of rapid innovation around new cloud software and the reality of data gravity, how can businesses realistically improve their analytics maturity?
I think this is an important question.
Consider the ability to provision almost anything a company needs today to get a job done — there are many options, AI and otherwise.
Also consider that data gravity states that where most of a business’s data lives, is where the analytics work happens. Applications and information architecture are of course related and tied, but totally distinct from one another.
The reason to “improve analytics maturity” is because it’s a fundamental part of any successful digital transformation.
To resolve all of this starting with the C-suite, I think it’s healthy to look at the world like this:
- Best in class companies will figure out how to manage all their data and do so as an asset. Not any one type of data, but all relevant data.
- Best in class companies will develop plans to leverage this data using a use case roadmap that prioritizes the highest value and most quickly tested projects. I’m not speaking about BI, reporting and analysts here, but rather use cases aimed at improving the business.
- These projects may need to operate across the innovative cloud software layer and your information architecture. The smarts may live with the cloud software and the data it creates and manages or live nearer the data layer and be completely separate but require integration with the software. Either way, all of this needs to be easily activated and governed.
What I have described here is the necessity to ensure collaboration — not separation — between the line of business management and the IT or data management group. Both need to execute in unison on an agreed-upon use case strategy that is endorsed by the CDO, CEO or other executives, who are actively engaged in what’s happening.
This is totally uncommon but is totally necessary for #digitaltransformation to succeed.Â
Discussion Questions
DISCUSSION QUESTIONS: How should retailers and brands be looking to modernize their analytic infrastructure to reach higher levels of analytical maturity? What behavioral changes may be required?
Poll
BrainTrust
Jeff Weidauer
President, SSR Retail LLC
Cynthia Holcomb
Founder | CEO, Female Brain Ai & Prefeye - Preference Science Technologies Inc.
The need for cross-function collaboration can’t be overstated, and it must start at the C-level. Once begun, it will require ongoing C-level engagement to ensure that the collaboration continues, teams are formed and relevant insights can be derived. Most companies are awash in data; many have excellent analytics teams mining it. But the analytics must be guided by the business so actionable results are delivered. It’s not really about the hardware or the cloud – it’s about the people.
Fundamentally, digital and retail transformation plans are completely dependent upon the quality of data. It’s important for retailers to assess not only the quality of their data but also take a very comprehensive look at how they measure success and their KPIs across every single department of the company.
Far too often, different departments at a retail company not only manage their businesses uniquely but things as fundamental as gross margin, inventory turns, conversion, etc. are measured differently. So the first steps of the digital and analytic transformations are to cleanse the data and align across the entire company as to what KPIs are leveraged to measure success.
Analytic maturity is critical for retailers to get right, as we are living in a consumer insights-driven world. The culture of leveraging analytics to make purposeful business decisions has to extend from the top down, and there has to be a close partnership between IT and the business teams to remain agile, prescriptive and drive the transformation together.
Retailers and brands are drowning in data and most remain in the early stages of analytics maturity.
Behavioral changes that retail companies need to thrive with data analytics include:
Retail companies’ ultimate goal is to use data insights to create a superior customer experience by serving each consumer as a unique individual rather than a homogeneous mass market.
That’s why retail companies with the drive (and budget) to unify all their disparate data from across retail touchpoints – and apply their insights – will gain a lasting competitive advantage.
Only one aspect of this transformation is looking at the multiple roles of the CIO. Those roles should include keeping IT skills current for themselves, taking a tactical approach to building IT staff skills and fostering workforce transformation through employee engagement. All of this is critical while being a central conduit to the line of business to ensure understanding of the value for continuous digital transformation.
Gib Bassett’s commentary is on the money. Big Data and AI have the answers. The challenge is how to use them. I have read that only 0.5 percent of available data is used. When you look at data accumulation, year over year, you understand why. Ways have to be developed to visualize the data in an actionable way. Good visuals, infographics and animations can take hundreds of thousands of data points and put them into a meaningful communication.
As Bassett says, “Best in class companies will figure out how to manage all their data and do so as an asset. Not any one type of data, but all relevant data.”
There are two additional factors to consider when modernizing analytical infrastructure. First, beyond internal collaboration, there needs to be better external collaboration between retailers and manufacturers. Most retailers and manufacturers don’t have the depth of resources necessary to go down this path alone. Only through enhanced collaboration can both parties achieve the analytical insights that result in driving sales. Second, many retailers can’t afford huge IT projects and they struggle everyday with how to economically aggregate disparate data sources. The retailers also struggle with how to effectively use data analytics to find actionable insights. Analytical infrastructure solutions must be cost-effective for these retailers. Otherwise, the definition of “best in class” will only refer to retailers with deep pockets.
Brands and retailers are complex organizations with multiple systems and data existing in many silos in the organization. Many times an organization may think that they are modernizing their organization by buying and utilizing a new data visualization software or another cutting edge solution but the foundational data structures are not sound. Many data types may not be accessible for many analyst teams. Therefore they are not fully benefiting from all their talent since they can only use a portion of usable data.
With all of the changes in the past few years and new data sources now available, I would suggest that C-level leaders of retailers and brands conduct a thorough review of their data and understand if it is interconnected via a master data management system. Sometimes, data sets are brought in by departments and are not known by others that could benefit the business.
Another deep review would be to understand the various analytics being conducted throughout the organization. Bring the cross-functional analytic leaders together to talk about their analytic projects and identify gaps and opportunities. Make that a permanent group and meeting that is used to ensure communication and collaboration are ongoing and strengthen over time.
I keep hearing “data is the new oil” — something I fundamentally disagree with. Data is not like oil because oil is a finite resource, whereas data is continually increasing in volume. Extracting value is key to this — but you need the right tools, people and processes in place to do this. Data is the new soil may well be more appropriate!
Sifting through the soil is the challenge. There is a shortage of data scientists as we all know. There is also the challenge of taking a predictive model and using it operationally to make it useful in real-time.
Full disclosure: Software AG is in the business of helping customers with both self-service analytics to help business users build models, but we have also invested heavily in allowing predictive models to be useful with real-time streaming data from IoT devices and operational systems to allow a real-time response to a specific problem or opportunity.
This is a hard question, as retailers and brands long ago turned their data over to data analysts, as retailers did not want to deal with it. Today analysts with no experience in retail, product or the human decision to purchase, are tasked with evaluating and translating data into brilliant and actionable insights. Behaviorally, this is pretty much a mess. Structurally various data silos are strung out among disjointed walled data infrastructures. Wow.
The race to embrace data has lead retailers and brands to abandon common sense. Retailers and brands need a subject matter expert in the room, not linear-based, finite mathematical formulas trying to solve for human insights.
Retail is a digital business. Period. Someone needs to help the C-suite cross the knowledge gap between the physical and digital world. Only then will executives be able to leverage their business acumen, lacking in data analysts, into powerful insights able to crack the codes of knowledge buried in their data silos.
Effectively using data to make better decisions (over time and in real-time) is a perennial industry problem. To modernize, retailers need access to good quality internal and external data, effective tools to analyze, plan, predict and act, and event brokers to easily and securely move data points to where the need to be.
The culture of distributed decision-making starts at the top, requires collaboration across the organization and a modern IT infrastructure to execute.