Skip to content

Microsoft Industry Blogs


With the deluge of marketing content consumers are exposed to daily, companies must constantly compete for consumer attention. While most companies have access to vast stores of customer data, those who convert it to insight and develop effective, resonant campaigns will stay ahead of the competition. Luke Shave, Sr. Industry Marketing Manager, CPG & Retail Industries for Microsoft, sat down with Tim Burke, CEO and co-founder of Affinio, to learn more about this evolving retail and marketing landscape, and how Affinio is using leading-edge technology to help solve customer challenges.

Can you tell us a little bit about Affinio and how it started?

Affinio was built six years ago when my co-founder and I realized that many of the analytics associated with social listening were only based around what people were talking about, not what they truly cared about. We decided to take a different approach and see if we could go deeper to extract customer behavioral patterns by looking at what people are liking and following.

What we discovered was a massive amount of data that could be leveraged to better identify the passions and interests of key consumers. By applying our core technology, we’re able to consolidate this consumer data and make sense of it through automation and machine learning, enabling us to identify the signals that draw people to a certain company. We also uncover patterns in that data, empowering advertisers and marketers to hone their message and make sure their brand stays relevant to the audience segments they want to attract.

What marketplace challenges are you helping to address with the Affinio solution?

Marketers and advertisers currently limit themselves by relying on weak signals and low data volume from single surveys and focus groups. As a result of those guesses and hunches, targeting and segmentation becomes predominantly demographic in nature, but not necessarily reflective of what people are actually interested in (or who a brand’s true audience may be). They can identify and analyze all they want around the individuals they’ve already identified as customers, but what they can’t do is analyze customers whose data they don’t have.

Another challenge is consumer buying behavior shifts so fast, making traditional personas no longer useful. Marketers must be more dynamic in how they approach campaign strategies and product development, and change campaign strategies based on new data that’s being aggregated on the consumer side. However, reconciling flexibility and detailed data analysis isn’t easy. You can’t keep up with the volume of data you’re bringing in and still identify these opportunities while shifting gears on the fly. In the market today, we’re seeing a massive hiring surge of research teams, data scientists, and data analysts who comb through large amounts of consumer data. Unfortunately, that’s not a scalable model, and certainly not one that can be achieved within organizations or enterprises that don’t have the volume or resources to build those very expensive teams.

How does your solution address these challenges using advanced technologies like AI, machine learning and advanced analytics?

Our solution provides end users with strong signals that identify “white space”—unexplored areas of opportunity within their market—leading to entirely new campaigns directed at audiences who were previously unmarked in their datasets. The platform uses an engine that continuously and dynamically adjusts to new data to identify opportunities. This wasn’t possible before the advent of advanced cloud technology.

We also see AI and technology as a vehicle to deliver the capability of building highly valuable models, automating the results from those models, and providing simple visualization layers that anyone can use. We want end users who may not have math, data, or statistics backgrounds to interpret and make decisions around customer data on behalf of their brands and enterprises. We’re placing data-driven personas, built at the speed of culture, in the hands of strategists and planners so they can understand what audiences they want to reach and what campaign, copy, and content to create.

You mentioned the solution was built for marketers. How did you integrate the marketing end user into what eventually became the Affinio solution?

From the beginning, we placed a heavy focus not only on collecting and analyzing consumer data, but making it easy to interpret and accurate as possible. We’ve iterated, evolved and expanded a visualization layer to such a point that we’re heavily recognized as a company. The back-end data clustering is extremely complicated and advanced, but its simplification into an easy-to-interpret visualization makes audience analysis actionable for everybody – it democratizes the data.

What we’re seeing in the market is a shift towards this approach becoming the single source of truth within enterprises. The visualization layer is something that everybody in an enterprise can collaborate on and discuss relative to strategy and opportunity. When data is formatted in Excel sheets, pivot tables and custom graphs that change from meeting to meeting, you never get that sort of commonality that you can lean on and leverage.

Why did you choose to build the solution on Microsoft cloud technology?

We saw very early on that we had a massive global enterprise opportunity. When deciding what the core stack of our technology would be, we wanted to work with a partner that was known for enterprise-ready, fully trusted, and highly secure solutions. Azure was the natural choice, especially due to the opportunity to work closely with Microsoft. Now, we have the benefit of working with all the technology they continue to add into Azure, as well as the ability to work as partners.

What are some upcoming functionalities that you’re working on with Microsoft?

One of the unique things that we’re working with Microsoft on is related to the translational functionality that’s embedded in Azure. Our goal is to be able to have any team, from any location, speaking any language, interpret and leverage the solution to collaborate in their own language, with their own interpretation. When you have teams in global office locations that all speak different languages, building a centralized marketing strategy is a massive task. What we’re seeing is an opportunity to keep marketing control with the local regions while also collaborating at a global level, which has never been possible before.

We’re also working on image and video analysis. We’re analyzing hundreds of thousands of videos and images across different channels to figure out what signals they have, so we can start being more programmatic in what kind of content we create. We don’t ever envision a space where the creative aspect is entirely replaced by machine learning, but we see a massive opportunity where it can support the ideation of content in a very prescriptive way. The signals are all there, but traditionally there’s been too much to absorb and act on.

What transformations are you seeing in customers that use your solution?

At the end of the day, any marketing intent is to consumer behavior in your favor. From our perspective, when you’re able to start detecting that signal, you’re also able to establish a baseline at the beginning of a campaign and see the shifts in those behaviors over time. We believe letting the data tell its own story will lead to companies buying in and changing how they operate their business. We’ve seen customers like Unilever and Danone change the way they create campaigns by identifying “white space” opportunities and targeting entirely new audience segments. They’re generating new campaigns and using our platform to identify what channels and formats drive those campaigns, because they now understand the audience affinities through audience analysis. By the end of their campaigns, they have true measurement that wasn’t possible before.

We’re also seeing our customers transform their thinking around brand loyalty and realize it’s built gradually over time. For example, if someone’s intending to buy a car or house, those decisions should be influenced at a brand level, which means connecting with consumers months, if not years, in advance. Companies need to keep demonstrating relevance so when consumers are making decisions, their brand is always top of mind. They can’t simply put an ad in front of potential customers at the last minute and expect a sale, they must connect with people at their interest and passion levels, which our solution empowers them to do. This new approach drives affinity and loyalty, and changes consumer behavior when it’s time for them to make a buying decision. I think the shift towards this vision is probably the most exciting and tangible outcome we see today.

Affinio Integrated Audience Analysis, built on Microsoft Cloud technology, is now available on Microsoft AppSource. Learn more about the solution by trying the demo today.