Featuring RK Maniyani, CTO, Anteriad

Artificial Intelligence, or AI, is a hot topic in B2B marketing. As with all buzzwords, most press coverage of AI focuses on cutting-edge, or even futuristic applications of the tech that aren’t entirely proven, at least in real practice for most companies.

At Anteriad, we emphasize what we called “Fact-Based” analytics – carefully examining historical data to find marketing tactics that are actually working.

We believe in our proven approach. AI is essential in analyzing the massive volumes of data, now available, about potential customers and your marketing channels. This is true of the information you collect via your own systems, as well as that found by third-party intent monitoring data services, such as Anteriad’s InsightBASE, that have been identified as an essential B2B marketing component by no less an authority than SiriusDecisions.

In fact, a smart overview from the Content Marketing Institute suggests that having insufficient or poor-quality data is a key risk factor with deploying AI. Without a large enough pool of information, AI can out-think itself, and come to the wrong conclusions. However, with enough data, AI is becoming an essential tool in differentiating yourself from the completion.

What is AI, really?

At its most basic, AI describes machines that are smart enough to perform tasks and make decisions that we typically think of as requiring human intelligence. It never rests and is constantly available to think through new data and personalize your marketing efforts on a per-contact level.

AI is not always easy to spot at the surface level. For example, some chatbots may seem incredibly intelligent, but behind the scenes they may actually be accessing data based on a ruleset that’s no more complex than an Excel pivot table. This type of “intelligence” will fall short in spotting meaningful trends with any significant volume of data.

Machine Learning is the most prevalent form of true AI in use today. Machines “learn” what to do, and spot trends based on analysis of historical data. Machine Learning has been in use at services like Google Adwords since 2008 for both historical and predictive analysis. Anteriad, for example, employs Machine Learning AI to analyze the intent signals gathered daily from across the internet.

Neural Networks are another model of AI, which mimic the functionality of the human brain, by passing information through layers of processing, and collectively interpreting the signals, to spot or predict patterns.

How AI Is Being Used Today

You’ll find AI applications in almost every segmentation and personalization marketing technology. Some use cases, such as PPC ad spend optimization, have been around for years, while others are now only on the verge of real operational benefit.

Email campaign targeting

• Choose the right audience: The matrix of targeting factors that makes a for good B2B Marketing email campaign is incredibly complex. Past response history, demographics, and frequency of contact all remain key factors; but now, intent signals from across the internet are among the most critical of your audience segmentation criteria. Finding meaningful patterns in that data requires AI.
• Choose the right time to send: Not every Monday morning is created equal. Cultural and social events can greatly impact response rates – good luck getting a U.S. recipient to open a mail the Monday following Super Bowl Sunday. AI solutions analyze and correlate signals from social media to pinpoint the best time window to grab your prospects’ attention.

PPC Targeting and Optimization

• Choose the right platform: Google AdWords and Facebook control approximately 60 percent of the pay-per-click ad market in the U.S., and they are typically a safe bet when choosing where to spend your PPC dollars. But there are dozens of other marketplaces that may result in higher ROI. As I mentioned earlier, Google uses AI in its tools to help optimize your PPC spend; third-party or in-house AI systems can perform the same analysis across all platforms, and also automate bidding in competitive marketplaces.
• Optimize audience profiles: As with any other marketing communications, PPC audience profiles are based largely on demographics and direct on-site interactions. But AI systems can incorporate psychographic factors, for example: does the prospect place more significance on practicality or value in their purchase journey – as well as past communications in real- or near-real time targeting segmentation.
• Optimize geo targeting: Events in various parts of the world can greatly impact the success of your geo-targeting efforts. This goes well beyond obvious cultural and large-scale news events. Intent signal monitoring and correlation with spending trends and brand awareness in various regions can lift your PPC performance.
• Optimize the content: PPC ad content is terse, and optimizing it amounts to a fairly exact science. AI systems can consider not only click rates but also deep keyword search data and even referral source as factors for tuning your PPC ad content.

Personalized Content Curation and Creation

Once you successfully engage a user, AI can personalize their experience through highly curated content, and even content creation. Most email systems and web site CMSs support these functions, that are now based on a handful of criteria. AI can evaluate hundreds of factors to determine the best CTA or related content to present to a specific user. A recent survey found that 33 percent of marketers already use AI to personalize content, and the trend will surely continue to grow.

Some current examples of AI-powered content optimization include:

• AI systems can monitor your site traffic and source patterns in real-time, pair them social media and intent monitoring data, and select the main content for the home page, based on macro-trends. It’s important to remember that you don’t always want to speak to a user based only on what they’ve done in the past – sometimes, you want to jump into the broader conversation. (The now-legendary Oreo blackout anecdote is a prime example). Your human staff simply doesn’t have the bandwidth to constantly monitor these factors, but remember, AI never sleeps.
• Natural-language generation is one of most buzz-worthy of AI technologies, and back in 2015, Gartner predicted that by 2018 (that’s 3 months from now, by the way), 20 percent of all business content will be AI-generated. That prediction was a bit aggressive, obviously, but AI can help out in creating highly structured “business content” like real-time stock graphs, annual reports and even terse recaps of fairly predictable events [think sporting events – who won, who was the leading scorer, where did they play, etc.] Full-length marketing copy remains as much art as science, at least for the time being. But, as already noted, AI is showing positive results in optimizing and even generating very terse content like email subject lines and PPC ad copy.
• Chatbots are almost as buzz-worthy as AI these days. Our Craig Weiss has some interesting thoughts on how truly intelligent bots will impact B2B marketing. Clearly, the ability to react in real-time with the massive, complicated data set that is a frustrated human customer, will be enormously valuable, when it fully matures.

AI Has Been Learning for a While, Now

AI has been in use at larger data-driven marketing and advertising companies for years, and as the technology matures, it will create almost unlimited optimization opportunities for your own B2B Marketing efforts. But AI will require a lot of high-quality data to fuel its analysis, and a touch of human intuition to prioritize key factors in engaging with your audience.