Site icon IT World Canada

Startup harnesses big data to help big brands better engage with their consumers

big data analytics

Affinio is an award-winning marketing intelligence platform that leverages massive consumer data sets to understand today’s consumer through their interests. Founded in 2013, Affinio empowers the world’s leading brands, agencies, media companies, and publishers to redefine the way they research, plan, and execute marketing strategies.

“With Azure, we can now deploy in days what traditionally would’ve taken individual developers months or potentially years to build.” – Tim Burke, Cofounder and Chief Executive Officer, Affinio

The need for marketing intelligence

The Affinio marketing intelligence platform is changing the way marketers relate to consumers. By understanding audience interests, marketers can apply an audience-first approach to identify, reach, and engage high-value customer segments.

Powered by advanced proprietary machine learning algorithms, our Affinio platform automatically groups audiences based on what they care about most. The result is that you can better understand them through clusters of like-minded people, which allows you to discover hidden audience segments. Analyzing people through their interests provides knowledge about and cultural insights into your entire customer base. This sheds light, often for the first time, on whom your audience includes, what influences them, and what they discuss, share, and deeply value.

Understanding people on a deeper, richer level

Consequently, our customers can align their creative, content, media, and marketing strategies with their own consumers’ interests to better resonate with different segments. Affinio customers include BBC Worldwide, 20th Century Fox, Universal McCann, and VaynerMedia. Their strategists and analysts can understand what makes their audience tick—all with actionable data and insights delivered on a comprehensive visual platform in just hours instead of days or weeks.

We live in a content-filled world that’s driven by the so-called attention economy. To reach consumers today, you need to know what they are doing. Until now, marketers have relied on demographic analysis and broad assumptions, but this is no longer sufficient. People are unique, complex, and ever-changing. Static personas and “spray and pray” (sloppy, too widely targeted) strategies have trained people to ignore your content.

What sets the Affinio marketing intelligence platform apart is its ability to ingest and analyze billions of data points to deliver actionable insights so rapidly. And that’s where Microsoft Azure comes in. We chose Azure because it gives us the massive storage and the compute power we need to identify patterns in diverse information and to naturally cluster people based on similar passions. This process of unsupervised audience segmentation lets marketers visualize nuances that might otherwise go unnoticed, resulting in better, more efficient business decisions.

For instance, we can identify similarities and differences in the natural communities that connect to each other, including subcultures within those communities. With such rich data put into the proper context, strategists and analysts can create highly focused content and ad campaigns that go far beyond normal targeted marketing, and they can confidently determine what will and won’t resonate with the audience.

Marketing data that only the cloud can handle

The information Affinio accesses, stores, analyzes, and reports on for each audience segment represents vast data sets, so the flexibility, scalability, and reliability of Azure are foundational to what we do. Comparable traditional static servers would be impossible for us to invest in. The ability to spin up servers in Azure and scale up to whatever storage and processing capacity we require are critical for the type of company that we’re building.

Now, we never have to worry about scale. We deal with Fortune 500 companies and their global data sets that they themselves have a difficult time analyzing. We can walk confidently into their offices, give them our pitch, and know with certainty that we’ll be able to deliver because Azure is running on the back end.

Microsoft Azure and machine learning

Three and a half years ago, we were huge fans of Amazon Web Services, right out of the gate. But as we grew and were exposed to the Microsoft Accelerator startup program, we gained a deeper understanding of what we could do with Azure. So, we made the switch, and the decision to port our platform infrastructure to Azure made good economic sense. Additionally, the machine learning capability of Azure was a key development feature.

To amplify the latter point, we needed motivation before making a cloud-provider switch of that magnitude. Because Azure implements machine learning so well, it has become a major focus of our development efforts. Our Affinio DSaaS (data science as a service) platform has to recognize and extract meaningful information (indexing signals) from the noise. For example, we’re not just analyzing text, we’re also analyzing images and soon even videos—tens of thousands of postings in real time and at scale—to identify users’ pertinent areas of interest.

When you start talking about that amount of content, you absolutely need speed. The machine learning capability of Azure (and the unlimited compute power) is key to attaining the image-processing efficiency to support acquisition features like these. We’re also confident in the reliability of those features when we roll them out because Azure is ideal for prototyping and conducting proof-of-concept tests in the lab. With Azure, we can now deploy in days what traditionally would’ve taken individual developers months or potentially years to build.

Learn more at azure.com

Exit mobile version