After establishing the building blocks of data and AI, what does an organization need to do to differentiate its data strategy?

Powerful visualization and strong governance can take a company's data strategy to the next level, members of Protocol's Braintrust say.
Good afternoon! With today's Braintrust question, we wanted the experts to tell us about how to elevate a data strategy from good to great. Getting the first layer of data and AI in place can be difficult enough to get right, but in this question, we asked the group to focus on what comes after those cornerstones are in place and highlight the tactics that can take an organization to the next level. Questions or comments? Send us a note at braintrust@protocol.com
Chief Strategy Officer at DataStax
Data strategy must elevate the company's core strategy. No matter what business you're in, there are a few measurable goals that really matter: the ones that represent your strategy and your progress and investors use to understand your business. Using the lens of data to focus on one of these — like your renewal rates, on-time percentage or partner-led revenue — will lead you to reimagine your company's benchmarks for speed and coherence.
Three essential lenses are customer-centric, operations-centric and ecosystem-centric. There's a defining question worth pursuing endlessly that pairs with each one.
Large companies are typically good at many things, but only world-class in one. Pick the lens and question that matches your world-class ambition. Then explore "What might we do if the company were a single thinking entity?" followed by "What's the maximum imaginable impact that would have on our goals?" At Google, the strategy mantra is "10x": target achieving a new order of magnitude. This is a powerful provocation for data strategy, and a meaningful test of your definition of world class.
You'll end up with a clearly articulated data strategy that is built into your core strategy, with the power to change how your company sees itself — and change how your CEO allocates capital.
Global Chief AI Officer at IBM
Too often, the data and AI building blocks that organizations put in place don't result in a sound structure. Why? They lack an integrated layer that holds all the blocks together. Many organizations have the right building blocks in place, but if their overall strategy doesn’t aim to get data in the hands of every user, can their data strategy be successful? The success of an organization relies on building a robust data management system that's accessible by everyone across the business and ready for business exploration.
Businesses should aim to get data to anyone who needs it at the right time, but until recently, most have lacked a clear strategy to do so. An approach that centers around a data fabric architecture helps to solve that challenge. Data fabrics allow an organization to connect and access siloed data across distributed environments without ever having to copy or move it, all with governance and privacy embedded.
A data fabric puts the people, and not the technology, at the core of the data strategy. It helps make sure all your data is exposed and usable so the business doesn’t have to spend time centralizing and finding the data and instead can focus on understanding what the data mean. Organizations that embrace a data fabric strategy will take a major leap forward towards delivering on their business goals.
CTO at Scale AI
Production-grade ML is tricky. Each model often solves multiple problems; it doesn't just detect stop signs, it also must detect stop signs in the fog, or a stop sign that has turned yellow from years in the sun. You need to live and breathe your data set to really understand how your model is performing, and develop an intuition for how your data is driving your model performance. Maybe you just don't have enough data in the fog for your model to get good at that scenario. But to really understand that, you need great visualization, powerful search and great model validation tools. Building these tools can take far more time than getting the first model trained, but luckily, cutting-edge tools are becoming available to help companies in this next stage. For example, Scale’s Nucleus product enables customers to understand, curate and manage data sets, producing high-quality models that continue to improve over time.
Global Director of Data Science and Analytics at Coats
A winning data strategy not only requires the creation of a strong data foundation, but it demands placing a premium on non-technical factors such as analytical agility and culture.
Three critical actions are key in distinguishing a standout data strategy.
Senior Managing Director and Global Lead at Accenture Applied Intelligence
To remain (or become) a leading company — or even to survive in the future — organizations need to transform multiple parts of their business simultaneously, at unprecedented speed. We call this compressed transformation. And in today’s world, having and executing on a sound data strategy that is aligned to business priorities is the difference between organizational success and failure. As we’ve seen throughout the last year, becoming a data-led company quickly emerged as the newest boardroom and CEO priority, and is critical to driving sustainable growth. It allows for an adaptive business, and innovative culture and processes that reinvent the organization using data and AI. And we’re seeing this across industries and geographies.
After establishing the data and AI foundation, which must include a governance policy and framework to use data and AI responsibly, organizations must focus on three areas to not only survive, but to lead. These are: to identify the use cases that will drive the most value and deliver insights that support better decision-making; knowing the customer better and differentiating through new products and services; and finally, to create a value realization framework enabled by the right operating model across the enterprise.
Just as important, a data-led transformation must be powered by a people-first approach. Transformation can have huge impacts on people and culture, and the organizations that prioritize nurturing their talent and culture shifts will ultimately be most successful in scaling their data strategy and building a sustainable workforce.
CEO at Informatica
Every company today aspires to be "data-driven." Research indicates over 99% of companies have invested in data initiatives, but only 24% call themselves “data-driven.” The biggest barrier to data-led transformation is culture. Investments in best of breed data management technologies powered by cloud and AI will not reap the rewards of data-led transformation without a data-first culture. Data literacy is what sets leaders apart in how they achieve business value from their data.
In all my conversations with our customers, the ones that are truly driving innovation and a competitive advantage by leading with trusted data are also those that invested in building their people and processes to think data-first. CDOs must invest in a continuous learning model where users beyond IT are well-versed in operationalizing their data strategy across all levels within the organization. A robust training and certification program to stay on top of data management best practices is the GPS organizations need on the road to managing data in a highly fragmented, multicloud world.
Last year, partly in response to changed business conditions due to the pandemic, Informatica launched its Foundation Level Certification 100 Series of on-demand training and over 13,000 data leaders worldwide are now Informatica-certified professionals at no cost. We are also offering the next level of training and certification, the 200 Series, which includes a deeper dive into product architectures, use cases for cloud application modernization, data “democratization” including data governance and privacy, and 360-degree business views for finance, supply chain and customer experiences.
Data literacy is critical to building a data-first culture without which no technology can realize its true potential.
Kevin McAllister ( @k__mcallister) is a Research Editor at Protocol, leading the development of Braintrust. Prior to joining the team, he was a rankings data reporter at The Wall Street Journal, where he oversaw structured data projects for the Journal's strategy team.
To give you the best possible experience, this site uses cookies. If you continue browsing. you accept our use of cookies. You can review our privacy policy to find out more about the cookies we use.