Watson Calling: Can Cognitive Computing Improve Ad Effectiveness?


The following post first appeared in the IBM “Building a Smarter Planet” Blog on May 21, 2013.

===============================================================

It’s often said that marketers are drowning in data, but here at Nielsen, we have an idea for turning that harrowing experience into an insightful deep sea exploration.

For decades, Nielsen has been in the business of empowering brands and agencies with the information they need to understand and build connections with consumers.  We do this by measuring the advertising and media content people watch, the goods and services people buy – and very importantly, by spotting connections between the two. Why are these connections important? If you know what ads people are exposed to, and you know what those same people buy, you can get a lot smarter about how well your advertising is working and how to make it work better.

IBM Watson Calling - Can Cognitive Computing Improve Ad Effectiveness?

IBM Watson Calling – Can Cognitive Computing Improve Ad Effectiveness?

However, despite vast amount of data at their fingertips—and often because of the complexity of the data at hand—our clients’ ability to form these vital connections is oftentimes hindered. In fact, our advertiser clients tell us that despite the ever increasing amounts of data they have access to, they still have problems answering basic questions such as: How much should I spend on advertising? How should I allocate my spending across media platforms and within them? And, how do I measure my advertising performance “in-flight,” and make course corrections to improve results?

To help answer these questions, Nielsen created a simple “end-to-end” framework for measuring advertising and media effectiveness: the 3R’s. The 3R’s try to answer three simple questions: First, is my advertising reaching my intended audience (Reach)? Second, is it breaking through, being remembered and changing consumer opinion about my brand (Resonance)? And third, is it driving a behavioral reaction – e.g. sales (Reaction)?

We’ve been working hard to create the data sets and tools to help brand marketers answer these questions. We’ve gotten much better at answering the fundamental questions about advertising effectiveness, but we still have work to do.

Like never before, we and our clients face a continuous and, sometimes overwhelming stream of data generated by consumers and our digitally enabled measurement tools.  Just envision terabytes of structured and unstructured data pouring into our offices – trends, metrics, sentiments and perspectives – that express who consumers are, which programs and ads they’ve seen, how those ads perform by TV program, genre, web site, placement, number of times they’ve seen the ad, exposure to social media, where they’ve shopped, what they’ve bought, etc.

Within this data, there exists a range of opportunities to better understand consumers. But the challenge is clear, and for many, overwhelming – due to the data’s volume, scope and growing complexity, it’s almost too much to decipher.

A new form of technology is needed to dive deep into this sea of data, and come up for air with actionable advertising and media insights that can help brands understand their campaigns’ reach, resonance and reaction, and in turn, improve their advertising campaigns.

That’s why we were so intrigued when we began discussions with IBM about an entirely new and unique category of technology known as cognitive computing.

Cognitive computing systems can understand the nuances of human language, process questions akin to the way people think, and quickly cull through vast amounts of data for relevant, evidence-based answers to their human users’ needs. And very importantly, they learn from each interaction, to improve their performance and value to users over time.

Seems like a custom fit for the data-driven challenges today’s advertisers face.

Building on this excitement, we’re embarking on a new collaboration between the Nielsen Innovation Lab, which we founded in 2012 to advance research in advertising effectiveness, and IBM to harness the power of Watson, the sole player in this exciting new era of cognitive computing.

IBM Watson

IBM Watson

As part of our own brand’s continuous effort to advance understanding around advertising effectiveness, we’ll be exploring ways to use Watson for helping our agencies and their client brands engage more effectively with consumers across all devices – from TV to tablet to smartphone – while improving the impact of their advertising and media plans.

We believe this collaboration will open up a realm of possibilities for our clients, so they too can uncover value from new and exciting data. It’s time to stop drowning, and start exploring.

Get free updates of Randall Beard’s Blog by RSS reader

Get free updates of Randall Beard’s Blog by e-mail

Advertisements

One Response to Watson Calling: Can Cognitive Computing Improve Ad Effectiveness?

  1. Tejinder says:

    The ultimate challenge lies in how to marry all the data together. With shifting trends from Frequentist to Bayesian methodologies of predictive modeling, algorithm development and execution have become relatively easier. The big data explosion however has special needs and needs special treatment. Once all these multiple dimensions of data are housed under one roof with a string of indices attaching them to each other, analysts can find treasure houses of information from these data.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: