Igor Perisic, a long-time LinkedIn executive, has seen the company transition from a professional networking entity to a Microsoft-owned technology powerhouse. In his 12-year-long career at LinkedIn, Igor has worn many hats — Director of Engineering, Senior Director, VP, and now Chief Data Officer. In the debut Q&A feature — Ask the CXO, Igor shares his views with ToolboxOpens a new window team on the responsibilities of a Chief Data Officer (CDO)— a recent C-suite addition, the importance of ethicsOpens a new window in data science, and how leaders can win the digital transformation race with a focused reskilling program.
Key takeaways from the interview:
- Understanding the right approach to data management
- Insights on data quality controls to build a winning business strategy
- Tips on how to leverage AI for data analytics
- Tips on how to build a good reskilling program
Read the full interview below:
The role of a Chief Data Officer is a recent addition to the C-suite? Can you tell us what are the key responsibilities of a CDO besides addressing the core function of data governance, platform, and delivery?
The title is sufficiently recent, so it can vary from company to company. At the core, the role tends to be a little different depending on how each company uses data, how regulated their industry might be, and if they are a consumer or enterprise type business. So there is no one standard job description for a Chief Data OfficerOpens a new window but many of us share the same general goals.
Most often, CDOs are accountable for how data is defined, handled and produced. They also tend to be the point of contact internally for all data policyOpens a new window issues for their companies. For example, when data policy decisions are complex and there is confusion over who is accountable for what, I am expected to work with many different stakeholders at all levels of our company (and our parent company) to ensure that the right folks are in the right conversations, and to help us all get to clarity on who are the right decision makers for specific issues.
Learn More: How the Role of the CTO Influences Digital Transformation: Q&A With Deloitte’s Bill BriggsOpens a new window
From leading engineering to CDO at LinkedIn, how was your transition? Can you share what the initial 3-6 months were like when you stepped into this new role? Are there any best practices you would like to share with other CDOs on how best to transition to this role?
While the official title of a Chief Data Officer is relatively new at LinkedIn, the work has been going on for much longer. So at LinkedIn, and I suspect at many other companies, it is more of an evolution than a dramatic change.
I am in a somewhat unique position as a CDO, in that I wear two hats — that of an engineer, in addition to that of a data officer. The teams that I oversee are responsible for building LinkedIn’s core data infrastructure, overseeing the evolution of our AI systems, creating our AI models and inferences, and carrying out all data analyticsOpens a new window functions for our products and the business. Hence, for me, it wasn’t a transition rather than an extension of responsibilities.
I have previously worked with many people in product, security, and legal teams as an engineering leader. I did not come from any of those backgrounds, though and I had to spend time understanding how things like new data regulations could impact not only these groups but others, such as members of the sales team. I am glad that at LinkedIn, I had good mentors who were patient with me to learn it.
Learn more: Why the API Economy Is Booming: Q&A With Postman Chief Evangelist Kin LaneOpens a new window
Technology giants are increasingly coming under regulatory pressure these days. And as we enter the next decade, data privacyOpens a new window , security, and trust have become the key talking points. Against this background, can you share how CDOs have to strike a balance between using data as a strategic asset and leveraging it responsibly?
At LinkedIn, I am happy to say that the importance of trust is not a new concept. A core value at our company is “members first,â€Opens a new window or putting our responsibilities to our members as a top priority. TrustOpens a new window is also one of our major operating priorities (only second to Talent).
As a members-first organization, we keep the privacyOpens a new window and securityOpens a new window of our members at the forefront of our work. I’m actually very glad to see that the conversation about responsible use of data and all that it entails (ensuring member consent, clarity in communicating how data is used, giving members control over their data, etc.) is now top of mind for most people. CDOs need to have the same conversations with people in their organizations and find ways to “scale†the data culture they are building within their own organizations.
From a strategic perspective, data is the foundation of every product and feature at LinkedIn. From our recommendation systems and A/B testsOpens a new window that optimize a member’s feed to business analytics systems used by our employees — every member, product, and business unit within LinkedIn relies on our data and data infrastructure in some way.
This shapes how we think about just everything at the company; from the products we build, all the way to the final experiences that our members see. Using this data to create an economic opportunity for every member of the global workforce is an exciting and empowering prospect, but also carries with it a set of responsibilities that I mentioned earlier.
How important is AI in your day-to-day job role and how do you resolve the data issues around it?
We have been saying for a while that: “AI is like oxygen at LinkedIn.†If we did not have Artificial Intelligence, many of our products and services simply would not function. At our scale, our products would not be able to deliver on our vision of creating economic opportunities for every member of the global workforce without the help of AI.
While CDOs do not need to be experts on AI, they need to know enough so that they can ask the right questions. If you treat AI and data scienceOpens a new window as magical tools that are beyond comprehension, how can you manage them responsibly?
This is a topic that needs much more discussion at the C-level in many companies. That’s why we have shared some of our internal best practices in areas such as A/B testingOpens a new window , discussed the importance of ethics in fields such as data scienceOpens a new window , shared our commitment to engaging with outside researchersOpens a new window , and given examples of how our principlesOpens a new window help us further our company’s mission and vision.
Learn more: Why We Need Better AI Regulation: Q&A With Access Partnership’s Michael ClauserOpens a new window
LinkedIn introduced the Knowledge Graph — what are the other data products coming out of the networking platform. As a longtime LinkedIn executive, how has the networking platform matured over the last decade?
The Economic Graph is a digital representation of the global economy based on the mapping of every member, company, job, and school. Through this, we’re able to spot trends like talent migration, hiring rates, and in-demand skills by region. These insights help us connect people to economic opportunities in new ways. By partnering with governments and organizations around the world, we help them better connect people to opportunities. Although it would be hard to name them all the data products we create at LinkedIn, the insights we glean from the Economic Graph have transformed the way that we approach data-driven insightsOpens a new window .
With respect to the maturity of the networking platform, I’ll answer it from the perspective of a data engineer. It has simply become much more stable over the years. We managed to scale it through our growth and increased our availability metrics in just about all our core services. To get to where we are, we needed to be able to scale for that next level and the next and the next. Not jumping from where we were, say 10 years ago, to where we are today in one step, but to constantly challenge our systems to be able to take multiple steps to get to where we are.
With respect to the product, one aspect that jumps at me is how it became a global phenomenon. In the early days, we had some doubts about whether we should have “localized LinkedIn’s†in different countries, given the cultural aspects of the workplace and how professionals preferred to interact with one another online. I am glad that we stuck to our vision and managed to deliver it.
What are some of the biggest trends we’ll see in the data industry?
The two clearest trends we see in the industry are Artificial Intelligence and regulations.
Artificial Intelligence: Artificial Intelligence, although in the limelight for a while now, is still a growing trend. It is hard to minimize the influenceOpens a new window it has had, is having, and will have not only in the delivery of your value proposition but also on your software and hardware infrastructure as well the skills required for your organization to succeed.
Regulations: RegulationsOpens a new window are now holding corporations accountable for how they leverage their data. We tend to forget that GDPR is less than two years old. The state of California’s CCPAOpens a new window is officially less than a month old. It is a new ‘trend’ and will grow globally.
Learn more: Top 20 Technology Trends to Watch in 2020Opens a new window
Reskilling has become a C-level priority for all organizations. How is LinkedIn addressing the skills gap within the organization, and in your view, how do you think L&D programs help organizations retain talent and functional knowledge?
First of all, I don’t believe that reskilling is really a new trend. After all, learning and development (L&D) programs were always entrusted to create learning curricula, hence reskillingOpens a new window their workforce. What may have dramatically changed is how much more central or strategic it has become, rather than a “nice to haveâ€.
There are multiple parts to creating an excellent reskilling program. We have worked for a while on creating some transparency, internally and externally, about critical skills for a given job function. This not only helps motivate “students†but can also provide guidance to L&D teams. We didn’t just stop there but also created learning paths within our LinkedIn Learning product that our members, employees and the L&D team can leverage when thinking about building a curriculum.
Given this need, we developed an internal program run by the AI team called the LinkedIn AI AcademyOpens a new window . The goal of this program is to equip employees across the company—in areas like engineering, product management, etc.—with the knowledge they need to optimally deliver impactful and responsible AI experiences to our members.
About Igor PerisicOpens a new window :
Igor PerisicOpens a new window is the Chief Data Officer (CDO) and Vice President of Engineering at LinkedIn. As the CDO, he helps ensure that LinkedIn is investing in technology, policies, and controls that allow the company to rapidly and safely scale its portfolio of product offerings. The data products created by his team power personalized experiences for hundreds of millions of LinkedIn members around the world. Igor joined the company in 2007.
About LinkedInOpens a new window :
Founded in 2003, LinkedIn connects the world’s professionals to make them more productive and successful. With more than 660+ million members worldwide, including executives from every Fortune 500 company, LinkedIn is the world’s largest professional network. The company has a diversified business model with revenue coming from Talent Solutions, Marketing Solutions, and Premium Subscriptions products. Headquartered in Silicon Valley, LinkedIn has offices across the globe.
About Ask the CXO:
Ask the CXO is a brand-new interview series on ToolboxOpens a new window that features C-suite executives from across the domain and aims to highlight how they are leading digital transformation and optimizing their digital portfolio. The interview is geared at bringing insights and perspectives on specific use cases, IT strategy, and the skill set required for the evolving landscape.
How do you think the role of a Chief Data Officer will transform in the near future? Comment below or let us know on LinkedInOpens a new window , TwitterOpens a new window , or FacebookOpens a new window . We’d love to hear from you!