In this digital age, artificial intelligence (AI) has the power to transform businesses — and is already transforming business operational models. However, AI, and AI-driven analytics, are only as good as the data that powers them.
That’s why businesses that succeed at digital transformation have two things in common. First, they have detailed strategies for managing data according to its role and its value, and recognize that different types of data need to be managed in different ways. Second, they have a data platform that makes data discovery, data integration, and data management as fast, easy, and automated as possible.
AI is Transforming Business Operations
An amazing example of how AI is changing the face of business is digital native Ant Group (formerly Ant Financial), the world’s highest-valued fintech. The Alibaba affiliate’s mobile and online payment app Alipay has more digital users than the biggest US banks, yet the company operates with just a fraction of the staff relative to that of those banks.
How does Ant Group do it? As Marco Iansiti and Karim Lakhani describe in their book, Competing in the Age of AI, the company has “the ability to leverage data to learn about users’ needs and respond with digital services to address them.” For example, Ant Group has completely automated its loan processes, transforming them into fully digital services with no human involvement.
Unfortunately, it’s not so easy for legacy organizations that are still using legacy operating models. Before these businesses can become good at AI — and produce the most meaningful analytics — they have been to be good at data. In Analytics at Work, distinguished professor, author, and thought leader Tom Davenport says, “You can’t be really good at analytics without really good data.”
Going forward, winning organizations will have the capability to turn data into value faster through analytics, and to turn those analytics into value and action through better business processes and improved experiences for customers, employees, partners, suppliers, and prospective customers.
What is Needed to be Great at Data?
As Davenport says in What’s Your Data Strategy?, “Even with the emergence of data management functions and chief data officers, most companies remain badly behind the curve. Cross-industry studies show that on average, less than half of an organization’s structured data is actively used in making decisions — and less than 1 percent of its unstructured data is analyzed or used at all. And 80 percent of analysts’ time is spent simply discovering and preparing data.4”
Davenport also states that “Having a CDO and a data management function is a start, but neither can be fully effective in the absence of a coherent strategy for organizing, governing, analyzing, and deploying an organization’s information assets. The plumbing aspects of data management may not be as sexy as the predictive models and colorful dashboards they produce, but they’re vital to high performance.”
He goes on to suggest :
- Data defense, which ensures compliance with regulations, the integrity of data flowing through a company’s systems, and the governing authoritative data sources including a single source of truth
- Data offense, which enables customer insights via data analysis and modeling, integrating disparate customer data, and market data to support managerial decision making
To learn more about how you can accelerate your data transformation journey, read our brief “Transform Data Into Business Value Faster.”
Data-Driven Companies Industrialize Their Data Processes
According to Iansiti and Lakhani, success here involves “industrializing data gathering, analytics, and decision making to reinvent the core of the modern firm, in what we call the AI factory.”
They suggest that an AI factory is built upon a data pipeline which gathers, inputs, cleans, integrates, processes, and safeguards data systematically. This includes, increasingly, a publish and subscribe methodology for APIs. The purpose is to make clean, consistent data available. Once this data is available, algorithm development and experimentation can take place.
But where are data leaders at in terms of delivering this vision?
In a recent #CIOChat, former CIO Isaac Sacolick suggests that “Many CIOs can, unfortunately, tell you more about the box data lives on than what’s inside the box and how it can be used to accomplish business outcomes.” And former CIO Tim McBreen adds that “Organizations try to fix data too late in the data usage chain. They fix it where they notice it instead of finding the source of the error. This wastes time and leaves the root causes untouched.”
The fact is, today, too many chief data officers (CDOs) and IT organizations are forced to take a time consuming, “swivel chair” approach to data management, relying on a collection of disparate applications for discovering, integrating, and analyzing data. As a result of this scattershot approach, data scientists end up spending too much of their time collecting and cleaning data, instead of analyzing it. No wonder AI initiatives aren’t yet paying off.
CDOs Remain Challenged to Make Improvements
Typically, a CDO’s goal is to ensure their organization is leveraging data as a strategic enterprise asset for data-driven decision making. But our research shows that they often are too focused on fundamental blocking and tackling. They have to act as data evangelists, keeping data in the forefront of people’s minds. They also need to educate people on the fundamental value of timely, accurate data. This means that most CDOs spend a lot of time explaining the value of data versus guiding their organization’s data pursuits.
Yet, digital disruptors like Stitch Fix are built on data and the value of analytics. The retailer’s 3,400 stylists work with an AI recommendation engine to suggest apparel for customers based on each customer’s personal preferences and information provided — and this is just one example of innovative data-driven personalization.
For this reason, today’s CDOs say they are investing much of their energy and dollars in data defense, including data dictionaries, enterprise data models, standardized data flows, and data quality tools. If only it could happen faster…
Parting Words
Data matters. And as Davenport says,“Good data is the prerequisite for everything analytical, it is clean in terms of accuracy and format.” Given this, CDOs and CIOs have their work cut out for them. However, the journey is worth the effort, as only organizations that are great with data — and have great data — can thrive in a digital, AI-driven world.
To learn more about how you can accelerate your data transformation journey, read our brief “Transform Data Into Business Value Faster.”