Artificial Intelligence | News | Insights | AiThority
[bsfp-cryptocurrency style=”widget-18″ align=”marquee” columns=”6″ coins=”selected” coins-count=”6″ coins-selected=”BTC,ETH,XRP,LTC,EOS,ADA,XLM,NEO,LTC,EOS,XEM,DASH,USDT,BNB,QTUM,XVG,ONT,ZEC,STEEM” currency=”USD” title=”Cryptocurrency Widget” show_title=”0″ icon=”” scheme=”light” bs-show-desktop=”1″ bs-show-tablet=”1″ bs-show-phone=”1″ custom-css-class=”” custom-id=”” css=”.vc_custom_1523079266073{margin-bottom: 0px !important;padding-top: 0px !important;padding-bottom: 0px !important;}”]

Interview with Steve Nowlan, VP of Decision Sciences – Conversant

AIThority-MegaMind
Steven Nowlan

Innovative software product development leader with strong background in machine learning and analytics. Track record of turning innovative ideas into successful software products and services. Extensive experience with mobile and location based applications, complex software architecture, and with management of combinations of research scientists and software developers.

Conversant Logo
A leader in personalized digital marketing, transforming the industry through cutting-edge technology, bold creative and a staggering amount of data. Their roster of 4,000 clients includes 400+ blue chip brands and 65 of the Internet Retailer Top 100. Conversant is a division of Epsilon, the global leader in creating customer connections that build brand and business equity.

Take us through your journey in the tech industry.

My career began in the 1980’s, and I’ve been focused on machine learning since around 1986. As an undergraduate and graduate student, I studied the different categories of machine learning and AI, and the history of the different attempts and styles. My thesis advisor was Geoff Hinton, one of the originators of today’s deep learning-based AI.

After completing my education and post-doctoral research, I worked at Apple and Microsoft. I worked with Steve Jobs on the development of Knowledge Navigator, a concept that ultimately led to the iPhone. Over the couple of decades that followed, I worked at four different startups and at a number of large companies, including Motorola, in their research and development departments and their labs.

Today, I lead a team of PhDs in Conversant’s Decision Sciences department. I joined the company in 2014, and was tasked with building a team of bright people who enjoy working with data and machine learning, and know how to use it to solve real-world problems. We develop the advanced analytic algorithms that drive Conversant’s intelligence platform, which delivers personalized digital advertising to consumers on behalf of our clients. A few months after I started, the company was acquired by Alliance Data and became the digital media arm of Epsilon, a global marketing firm and another Alliance Data company.

Each stop along my career has given me a unique vantage point to witness the evolution of AI. The techniques that are being used today were actually invented in the period between 1985 and 1998, but fell out of favor for a time due to the lack of scale and data. It was only recently that we had enough data and computing power to finally demonstrate that the only successful path for AI would be to use systems based on adaptation and learning at a fine scale.

 

Define your ideal customer profile. What AI-powered solutions does Conversant offer the ideal customers?

Conversant and Epsilon work primarily with large Fortune 500 companies. Like anything that relies on machine learning, delivering personalized digital advertising at scale to millions of consumers requires huge amounts of data, which large companies have. That’s not to say we don’t work with smaller companies, but our ideal clients do tend to have a lot of data.

AI underpins everything Conversant does. We make extensive use of AI and machine learning to provide solutions such as customer engagement, customer relationship management, customer acquisition, media delivery, and more.

It starts with data. Conversant’s intelligence platform combines data from thousands of client companies worldwide. This network effect is one of the reasons Conversant has such a deep understanding of consumers. Our self-learning machines sense hundreds of billions of streaming inputs billions of times daily—then optimize, output and optimize again, in a continuous, controlled loop, more than a trillion times daily. Every time consumers act, across all channels, we learn from it. This allows us to make intelligent decisions about whom to message, what to say to them, how and when to deliver marketing and when to hold back – enabling smart, efficient one-to-one conversations.

We’re in our eighth year of continuous machine learning, so our clients get a long view of millions of consumers as their tastes, needs and categories have changed.

AI also plays a role in the measurement and reporting we provide for clients. For each ad shown to a consumer, AI measures the effect it had and then learns from it. We then report back to clients about their advertising performance and use that learning signal for real-time optimization that ensures clients always reach their goals.

Conversant also uses AI to make our operations more efficient. One of our company mantras is “automate to innovate”. This entails taking repetitive processes and replacing them with learning algorithms that complete tasks quicker, which allows our associates to focus on innovation projects that will drive meaningful outcomes for clients.

 

Read More: Interview With Field Garthwaite, Co-Founder And CEO, IRIS.TV

 

What role is AI playing in making Personalization a ubiquitous choice for marketers in 2018?

AI is at the heart of everything Conversant does, and more and more marketers are embracing it to keep up with the dynamic, ever-changing nature of consumers’ online behavior. The proliferation of emerging technology is happening more rapidly, and consumers, particularly younger generations, are changing alongside – responding to new opportunities, products and media channels more quickly than ever. Each online action and interaction from a consumer generates billions of new data points every second. Machine learning and AI make it possible to observe these inputs, learn from them, and optimize marketing in real-time.

 

What are the fundamental tenets of Conversant’s customer engagement platform? Where would AI fit in a CMO’s customer engagement stack?

Conversant has four fundamental tenets, which must work together for effective personalized advertising. Our relationship with Epsilon strengthens these tenets and makes our offering more robust.

  1. Identity: Conversant provides a comprehensive view of 200 million real people. Each person’s online and offline lives are seamlessly aligned to a Core ID, which is built on a foundation of verified transactions. The IDs strengthen over time as we continuously match them to 120 million transactions and 200 billion online interactions every day. This drives an unparalleled level of accuracy, stability and longevity, which means our clients stay connected to 80% of their audience on online channels, and nearly 100% of them on offline channels. When marketing is built on a foundation of truth, everything you do is enhanced.
  2. Personalization through data: We offer personalization with complete, actionable views of millions of consumers. Using our Core IDs as the basis, we attach years of real-time and historical data to build Core Profiles across 7,000+ person-level attributes. We understand consumers based on years of their transactions, web browsing behavior, offline activity, video views, emails, survey responses and beyond. This deep knowledge allows us to personalize every interaction and ensure that it reaches each in-market consumer in the best channel; that it has the most personally relevant content, product, title, format and offer; and that it’s seen at the best time, within the right context, in the right sequence and with the right frequency. The knowledge of what consumers care about and want to hear enables marketing decisions that are made at the person level, including the content a brand shares with each consumer.
  3. Optimization through machine learning: Our self-learning machines are always running and improving, with real-time optimization that ensures marketing is aligned with each client’s goals. The continuous feedback loop enables Conversant to anticipate each customer’s changing needs, and predict products and messages they’ll be most receptive to. For each consumer, we build a combination of 13 offline and online communication channels, so clients can reach them with personalized marketing only when and where they’re likely to act. Performance and efficiency are optimized in real time to drive business outcomes for clients.
  4. Return On Marketing Investment (ROMI): Conversant takes a holistic approach to our clients’ marketing investments, partnering with them to optimize both sides of the equation – revenue and marketing investment – to drive the best performance with the most efficient spend. We think of ourselves as ROMI advisors. Even as clients invest in internal teams and other marketing resources, we’ll work with them to maximize their efficiency and performance. Together, Epsilon and Conversant have 50 years of marketing leadership and 20 years in the digital space, so we know how to guide clients around the industry’s pitfalls. This gives our clients peace of mind that their marketing dollars are spent wisely and that their performance is continuously being fine-tuned for efficiency across all their marketing investments.

When it comes to the stack and where AI fits in, the answer is everywhere. AI is a key component of each and every aspect of Conversant’s platform. Everything from identity, which is actually a learned concept that’s constantly updated, to the way we use predictive AI algorithms to make real-time decisions about how to best take advantage of advertising opportunities.

 

Is it time that the CMOs finally take AI-as-a-Service seriously and start deploying them?

CMOs shouldn’t think about AI-as-a-Service; their time is better spent driving toward objectives such as customer relationship management, marketing investment, optimal allocation of spend across channels, etc. Instead of focusing on a standalone technology, marketers should focus on the business at hand and select the best end-to-end solution that drives the greatest efficiency. This is really how AI will transform industries; not by buying a third-party, out-of-the-box AI solution and expecting it to magically understand advertising overnight, but by starting with a complete end-to-end stack and understanding how AI is used to make it better and smarter.

Machine learning is not much different than human learning – the more experience you have, the smarter you are. In the case of AI, the more time and data seen by self-learning machines, the more intelligent they are in making the best possible decisions. Given that the algorithms being used across the industry are not that different, experience overrides most everything else.

 

Read More: Interview With Tomer Kashi, Co-Founder & CEO, SkyWatch

 

What industries and departments would benefit most from utilizing Conversant’s Customer Acquisition model?

Conversant offers a broad spectrum of solutions for marketers, with customer acquisition being just one of them. In its purest sense, customer acquisition entails finding new customers who have never shopped at a store or with a brand in at the last 3-5 years, and bringing them into the fold. These are referred to as “new-to-file” customers. We also do prospecting, in which we identify customers who are aware of a brand but are not buying, or customers who buy occasionally and have the potential to become regulars. Another program we offer focuses on reaching a brand’s best customers and expanding their loyalty, average order value or shopping frequency.

Customer acquisition is applicable to most, if not all, industries and business verticals that engage in advertising and marketing. Conversant works extensively with clients in retail, media and entertainment, automotive and manufacturing, CPG and grocery, financial services, and more.  At the same time, it’s important not to overlook existing customers and spend too much time and money going after prospects. We see brands aggressively pursue acquisitions, thinking it’s easier. But if they have strong brand identification and connect with consumers who have already demonstrated an interest in the brand, they’ll have an easier time making a connection for a repeat purchase.

 

What are your predictions for the AI and business intelligence technology industries in 2018?

We’re going to see numerous attempts at flashy AI projects that fall short, and numerous successes that have a more practical or behind-the-scenes application. A recent Harvard Business Review article, “Artificial Intelligence for the Real World,” addresses how AI is impacting various industries. The authors pointed to a survey which found that generally, the greatest AI successes occur in more “mundane” projects with expectations that are within reach. At Conversant, we have found that to be the case, too. We use AI extensively, but use it to constantly improve things we have been doing for decades, such as building individualized, anonymized consumer profiles that get leveraged for personalized advertising. It’s scenarios like these in which AI will have the biggest impact in the long term.

Thank you Steven! That was fun and hope to see you back on AiThority soon.

Leave A Reply

Your email address will not be published.