Blog Post:

Tim Burke, CEO of Affinio, Talks Containerization

5 minutes

Affinio Inc

Recently Affinio announced it has created a containerization model for the company’s marketing strategy platform. What is a containerization model, and how does it address the chief concerns facing brands today? To answer these questions, Jon Baron, Affinio’s CRO, sat down with Tim Burke to talk about containerization, and its implications for the mar-tech world.
JB: Marketers are accustomed to a whole range of tech terms, but even for the most seasoned CMOs, containerization is a mouthful. What is it?
TB: The easiest way to understand it is to think of it as a modern twist on on-premise software, but instead of sending physical servers to run technology, the tech is packaged in a contain and later deployed in a private cloud instance, controlled entirely by the enterprise. So rather than uploading enterprise-owned private data to a third-party vendor for processing, the vendor sends its tech to the enterprise to run within their firewalls. In our case, we send the technology that powers Affinio’s AI-segmentation and visualization tools to the enterprise, which they deploy within their own private Microsoft Azure cloud instance.
JB: Is this a brand new technology or has it been used before?
TB: Containerization isn’t new. IT began using containers as a way of ensuring software runs reliably when moved from one computing environment to another (e.g. from a developer’s laptop to a test environment or from a physical machine in a data center to a virtual machine in a private or public cloud). Containers incorporate everything an app needs to run — the app (obviously), configuration file, dependent libraries and so on. This makes them easily transportable to an enterprise’s environment.
Companies like Docker have been offering enterprise-class container solutions since the early 2000’s, and containers are used throughout the enterprise, across multiple sectors.
JB: Why should marketers adopt the containerization model?
TB: The mar-tech sector is under intense scrutiny when it comes to how brands use first-party data, and for good reason. Consumers and regulators alike don’t want consumer data shared without explicit consumer approval. This is the heart of GDPR.
Meanwhile, data breaches are also a huge problem, as we’ve seen with Cambridge Analytica’s covert scraping of Facebook data. The fact is, the more brands share their data with vendors, the more risks they’ll face that their data will be breached.
Every CMO, CIO – and CEO for that matter – is worried about first-party data compliance. But at the same time, most organizations lack the technical resources and data science teams to develop deep insights from their own data. If their legal departments tell them they can no longer send their first-party data to their data analysis vendors, what are they to do?
Containerization solves this problem by allowing the brands to bring the technology inside their own private clouds, like Microsoft Azure. This way, first-party data never leaves the enterprise, and the brand’s marketing team gets to analyze the massive amount of consumer data stored in their data lakes in a privacy-protected way to unlock the insights they can then activate across a wide array of marketing, advertising, and customer care initiatives. The best part is that this model allows the enterprise to own their own insights, not a third-party vendor.  This has long-term competitive advantages to global enterprises who are fighting hard to stay relevant and profitable.
JB: That brings up another concern…how do brand marketers currently access first-party data for insights and strategy?
TB: The harsh reality is that information security risks prevent most marketers in global enterprises from having direct access to their company’s first-party data. In addition, most don’t have the technical skills to extract insights from this data even if they had direct access. If they are fortunate, their organization has invested in a data science team or insights team who can access this data, but unfortunately many of these teams are overwhelmed by the number of requests they receive across the organization, which means marketers must find other resources of data to build their strategies.  Many rely on survey or panel data or reports from consultants, but these insights often lack nuance, scale and speed.
Our vision is to put the power of AI at the fingertips of marketers around the world so they can answer their own questions and build data-driven strategies that scale. Our containerization model allows the enterprise to control who within their organization can access the Affinio technology. In this scenario, Affinio serves as an “intelligence” layer overtop of the first-party data within the enterprise itself, allowing authorized users to access aggregate insights into their audiences, without accessing the actual PII data, adding a new layer of first-party data security for enterprises.
JB: You mention Microsoft Azure…has that platform been used by marketing departments before?
TB: No, the primary use of containerization is dominantly for IT applications.  We believe that Affinio’s announcement is unique for the Microsoft Azure Marketplace and is a valuable addition to Microsoft’s core offerings by embedding our technology at the front end of the enterprise marketing strategy.
JB: Will container-as-a-service replace SaaS?
TB: I don’t think this will happen overnight, however with the changing market dynamic, I do believe that containerization will become a strategic deployment offering to many major global enterprises in the coming years, allowing brands to fully benefit from third-party vendor technology, like Affinio’s, without releasing any of their own data. I can totally envision of world where, for example, AI companies containerize their applications and provide them to clients who pay “as-a-service” monthly fee to access that technology.