Learn

audience.ai 2.0 – Using Conversational Analysis For Audience Data and Research

What we SAY is a great indicator of what we FEEL, what we INTEND and what we WANT. By analyzing the written word, we elevate the understanding that brands can have about their audience.

For more than a decade Networked Insights has been driving forward the capabilities of what is possible with conversational data. Based on what people say online, we have helped cosmetic brands know what shape to make their mascara brushes, we have given insight into how consumers will feel about emerging technologies, and perhaps my own favorite, we were able to predict electoral behavior by analyzing emotions.


As we head into Q4, we have brought together our suite of Audience Research and Audience Data capabilities under the umbrella of audience.ai.


We Speak Conversations

Translating what is written into actionable data has taken a considerable amount of time and effort, and an evolved taxonomy of over 32,000 attributes. We collect data from over 1,300 sources, and we analyze millions of conversations every day. From that data, we run Audience Research reports, and we provide Audience Data segments for targeting.


Audience Research

We pride ourselves on answering the tough questions that brands have a hard time answering, as well as questions that could only be addressed with the type of data we have.


Below are a number of common research reports we get asked to run, but the message is “what can we help you answer?”. Chances are, we can do it!


— Discover Your Audience: 
who is your audience, what are they motivated by, what do they like, where are the opportunities and gaps?


— Persona Deep Dive: 
the top affinities of your audience(s)/segments and what media they engage with (TV, Web, Apps)


— Brand Ambassador: 
discover which celebrity/public figure will resonate with your ideal audience or your brand specifically


— Instant Surveys: 
get quick answers to questions that would take months to do


Audience Data

Typically, for online and offline targeting, the advertising industry still largely builds segments based on demographic data, or as a reaction to a small number of behavioral events. You can be classified as an “auto intenders” from a single page visit to a car auction website or be seen as “in market for a home” just from checking out the price of your new neighbor’s house on Zillow!


The written word is different. The written tells us what you might be in market for, how strongly you feel about what you intend, and how you plan on going about solving that problem.


Step 1: The starting point 
—You have a specific use case, such as “I need to reach…” / “I want to target…”, or give us a seed such as “my competitor’s social followers”.


Step 2: Data where you need it
 — Once the analysis is complete and the data is ready, we load your custom segments into the DSP or DMP of your choice.


Step 3: Transparent scaling 
— When producing segments we typically provide 3 layers; (i) the most accurate / core defined audience, (ii) closest match, more scale, and (iii) growth segment using broader look-alike


(and predictive leads)

In addition to data for digital targeting, we also translate the data into individual records for offline targeting. Uniquely, a lot of our data tells us what someone is going to do a long time before they actually do it — we call this ‘Predictive Data’.


A great example is the home mover. Commonly a marketer will buy leads from USPS who provide a file of movers in the last 90 days. Each of those marketers is then competing against the others to get that individual’s attention.


However, with our predictive capability, we are able to provide leads that have a high likelihood of being a move up to 90 days in advance, giving an obvious advantage to the marketer.

Our dedicated lead team can help answer any questions you might have.





Author


Avatar