This is an extraordinary time to be a marketer. The need for positive, authentic connection has never been greater than it is now. So, too, the opportunity to shape it. Marketers have always played a key role in unlocking sources of growth. But delivering top performance means owning and enhancing the entire customer journey across the enterprise. This requires a degree of digital maturity that many organizations are still trying to master.
The silver lining to periods of disruption is the permission it gives organizations to shake things up and emerge smarter. CMOs can lead that transformation—helping companies become more reflective, responsive, and relevant. But they don’t need to do it alone.
Artificial intelligence (AI) can be an enabler. Because AI can understand, reason, and learn from massive and diverse data inputs, as well as facilitate natural connections between humans and machines, it can stimulate marketers’ creativity and help them make faster, better decisions so they can thrive in their expanded role.
CMOs can lead in this moment, and AI marketing can help
Recent years have seen a steady stream of digital adoption. In 2020, that stream became a tidal wave. E-commerce growth jumped 30% in the US alone, accelerating the shift to online shopping by nearly two years. A recent IBM Institute for Business Value (IBV) report found that business leaders’ needs for speed and flexibility have amplified dramatically, and 59% of organizations accelerated their digital transformation because of the COVID-19 pandemic.
This altered landscape is challenging marketers to revisit fundamental truths about who their customers are and how best to engage them. Segments have changed, buying patterns have shifted, and consumer behavior has evolved. For example, customers who are navigating the complexities of shopping during the COVID-19 era say they are more likely to select a product based on convenience, health, safety, and purpose than they are brand loyalty. To unearth salient behavioral indicators in time to act on them, marketers need real-time data from a variety of sources and at a far more detailed level.
In 2020, e-commerce growth jumped 30% in the US, accelerating the shift to online shopping by nearly two years.
Even before COVID-19, marketers were contending with an explosion of data, rising customer expectations, and fierce competition—all driven by the exponential growth in digitization. What the pandemic has done is raise the stakes, putting a premium on digital strategy and reducing the margin of error. If digital transformation was a priority before, it is essential now. According to IBV research, more than three-quarters of senior leaders expect changed customer behavior to continue after COVID-19, with consumers trading face-to-face contact for more shopping and customer service interactions online.
Fortunately, marketers now have AI-powered tools to keep pace with these shifts. However, many have yet to employ them effectively. IBV research reveals the majority of marketing departments have not moved beyond the AI evaluation stage, and fewer than one in five have implemented AI in any one particular core process. But, when deployed as part of a comprehensive digital strategy, AI can help CMOs and their teams operate at the velocity of today’s marketplace demands, giving marketers the freedom to focus on the work that matters most.
Changing what we know: Enabling enlightenment at the scale of digital
AI makes short work of monumental tasks humans can’t do on their own—merging a vast number of data sources onto a single interface and scanning text, online footprints, social sentiment, video, and other forms of structured and unstructured data at scale. Instead of marketers combing through oceans of data, machines can now do that for them.
Consider the good news, bad news plight of a North American multimedia retail company. More customers were flocking to the company’s various digital channels than ever before, but they were bouncing from channel to channel before purchasing, making it extremely difficult to discern buying signals. To develop audience segments, teams were reduced to the time-consuming task of manually sifting through gigabytes of data in Excel. The company needed more data, more granularity, and more efficiency, so it implemented an AI marketing platform across its channels. The solution collects browser and search histories, customer profiles, device type, and other information, then organizes the findings in a searchable format. Marketers can now easily query the platform for desired demographics and instantly create targetable microsegments.
In addition to interrogating data in nimble ways, AI can help make sense of unstructured information. For example, marketers know that social media chatter contains invaluable insights. But many of those insights can fly over the proverbial heads of most data collection engines since the algorithms have no way of decoding slang, emotion, mispronunciations, contractions, and other conversational elements.
With natural language processing (NLP) and understanding (NLU) tools, marketers can make sense of unstructured data quickly. Instead of poring over massive amounts of content to identify patterns and insights, marketers can turn that work over to AI. With NLP, AI can ingest and categorize information across common themes, topics, or tones and help teams make decisions about content creation, SEO optimization, and page optimization at scale.
Natural language processing tools help marketers more accurately identify audience preferences on an individual level.
Tone analyzers that use NLP technologies take things a step further, helping marketers sense what customers are feeling as they interact online. A French bank, for instance, employs NLP as an email analyzer. On a daily basis, the tool scours more than 300,000 emails and detects customer intent with 80% accuracy. At a time when contact is driven by faceless interactions such as emails or virtual chats, marketers need new ways of gauging nonverbal customer sentiment and forming genuine connections throughout the customer journey. Tone analyzers can meet that need, helping marketers address customer concerns at a faster rate and bridge the gap between digital and physical worlds.
In these ways, AI helps marketers understand their audiences’ habits and preferences on an individual level and with higher accuracy.
Changing how we work: Empowering decisive action
Marketers also need the ability to make decisions with lightning speed. While digitization has exponentially increased opportunities for marketers, it has also significantly multiplied marketers’ work. Building customer awareness, interest, desire, action, and ongoing loyalty requires marketers to choreograph end-to-end customer engagement, not just discrete pre-sales elements. Success now relies on marketers’ ability to make many more decisions on many more workstreams in a very short timeframe.
Knowing which action will have the most impact on any one customer becomes terribly complex when dealing with an audience of millions, each at different stages of the buying process. To be effective at scale, marketers need a way to automate decision making without compromising the individual touch.
Intelligent workflows can look at that audience of millions, analyze similar customer journeys made by similar individuals, and help marketers determine their next-best move with hyper-personalization. While marketers go about their day, AI-powered models work alongside them, crunching through customer analytics and purchasing and performance data to anticipate not only customers’ habits and preferences, but also their motivations in the precise moment and how they prefer to engage. Based on this analysis, AI can tell if, for example, the opportunity is ripe to schedule an event and who should be invited or if a new email campaign should be created, which customers to target and what content to use. With this context and evidence-backed recommendations, marketers can quickly select the appropriate response and deliver a successful experience.
AI has also helped marketers adopt an agile approach to their work by enabling them to focus on human-centered outcomes. AI systems can be trained to flag key indicators, alerting marketers when something needs their attention, such as content with low engagement rates or skewed audience targeting. Knowing that AI is working in the background cuts guesswork, freeing marketers to be more responsive and, ultimately, more effective.
Insight: Accessing a world of intelligence through a single window
At IBM, we know that streamlining processes is one way to move at the speed of digital, but what about streamlining thinking steps? Recognizing that marketing teams might easily spend all their hours looking for customer clues in pools of information and hundreds of workstreams, we wondered: What if we turned our reporting system into an analytical service? The result is “Pearl,” a platform that provides at-a-glance intelligence on campaign effectiveness from a single window. The platform synthesizes real-time visitation, email, event, and CRM data into a centralized view that lets marketers see everything in one place. It also sends alerts to marketers throughout the day, feeding them performance metrics and recommendations so marketers can stay up to date while on the go.
Changing how we engage: Engaging authentically with precision
Customers want authentic connections with their brands and interactions that feel human and empathetic, especially these days. But while personalization has been the promise for decades, only 18% of consumers said ads “often” seemed to understand their needs. Blunt-edged retargeting and other poorly executed personalization efforts not only result in missed opportunities, they can turn customers away entirely. Brands must create trusted exchanges that protect customer privacy—and they will need to do it without the assistance that mobile identifiers and third-party cookies have provided in the past.
AI can help brands break through the noise and create inclusive, authentic interactions that enable customers to feel known and understood. Yet, according to the recent IBV AI survey, while many top marketing executives are evaluating and piloting AI to personalize customer outreach and offerings, only 10% have moved forward to officially implement it, and a mere 4% are operationalizing AI for this purpose.
It’s time to change that. Instead of one-way conversations, AI solutions can help brands create conversational and reciprocal exchanges. In addition to employing virtual customer assistants, chatbots, and apps, companies can create advertising that enables customer interaction, such as prompts that allow an automotive company to ask customers about their dream car.
The makers of the BEHR® Brand, for example, offer nearly 2,000 colors in the company’s paint collection. But consumers can be overwhelmed by choice, struggling to choose colors, and as a result can be hesitant to start their own painting projects. To reach consumers and help make the “do-it-yourself” process easier, Behr used IBM Watson Advertising Conversations, which leverages Watson’s machine learning and natural language capabilities, to enable real-time, one-on-one dialogue with consumers and deliver unique paint color recommendations.
Thousands of consumers engaged with the ad, spending on average over one minute with Behr branded content. In addition to driving increased brand preference and loyalty, Behr has leveraged Conversations as mini focus groups, gleaning real-time insights such as the finding that consumers are most interested in relaxing, comfortable, and friendly colors. Findings like these can help the brand improve its ongoing product and marketing strategy.
AI marketing solutions can help brands create more engaging, conversational, and reciprocal exchanges.
Brands can use commercially available AI to identify and segment audiences and build creative elements in real-time—assessing past targeting and ad performance, comparing that against desired performance, and identifying new audiences that are likely to buy. AI-powered systems can also personalize ads based on what works best for the team’s goals and suggest which ads to run. Smart algorithms backed by natural language technologies can even assist with writing ad copy. While some of these systems are nascent, the field is expanding rapidly.
Insight: AI can be an advertiser’s best friend
At IBM, we are using self-learning AI platforms like Watson Advertising Accelerator to help media buyers discern which creative elements resonate best with specific audiences, continually refining the visual and messaging mix in response to signals and triggers in the environment. Brands use Accelerator’s sensing capabilities to be more attuned to subtle shifts and help customers feel heard while boosting campaign performance. We tested this technology to drive more installations for IBM’s Storm Radar weather app and found that Accelerator helped increase the rate of customer app installations by three times in just three weeks. Organizations using IBM Watson Advertising Accelerator since the launch of the solution in January 2020 have seen a minimum of a 25% increase in campaign performance from the start to the end of each campaign.
No matter how exciting the technology may be, in the end, customers need to feel their needs or concerns are addressed in the exchange, or the experience will be discordant. As high-profile incidents of social injustice in the US have made clear, systemic bias remains a persistent problem. Brands have an obligation to create inclusive experiences that honor and reflect the diversity of their audiences. Establishing that trust requires best practices that can guide the safe and ethical management of AI systems including alignment with social norms and values, algorithmic responsibility, compliance with existing legislation and policy, assurance of the integrity of the data, and protection of privacy and personal information. The effective use of AI requires governance protocols that help monitor model fairness, ensure explainability, and uphold transparency—crucial steps that allow marketers to be responsible stewards of the trust customers place with them.
Digital transformation is an inflection point—and CMOs can lead the way. With the help of AI, marketers can turn digital leadership into customer leadership and help ensure key voices are heard. Here’s what marketers need to do now.
Start with an outcome and scale what works
– Determine the outcomes you need to achieve. Do you need to get to know your customers better, automate complicated processes, or optimize your teams’ time? Identify your priorities and then work backward.
– Establish a baseline. Assess what is working and what needs to be changed. If you’re not sure where to start, find a partner who can help simplify the process and assist you in your AI journey.
– Create intelligent workflows. With clear priorities and a baseline, you can identify where marketing automation can be integrated into your processes to boost productivity and performance across your business.
Build a strong, trusted data foundation
– Be transparent. For teams to confidently rely on AI to prescribe or automate decisions, they need to trust the technology. Being transparent about how the models are trained and which data sets are used is a key ingredient in building trust.
– Make sure the inputs are as important as the outputs. For AI to function properly, you need to invest in access to reliable data. Are the data sets inclusive and unbiased? Are the models explainable?
– Support a 360-degree customer view. Make sure the data that fuels your AI solution is sufficiently comprehensive to offer a full view of the customer journey as people engage with you across your enterprise.
Drive an agile culture focused on learning
– Prioritize speed of response, flexibility, and continual iteration. AI is constantly evolving and adapting as customer behavior changes. Traditional waterfall approaches make it difficult to work at this pace, but agile teams enable marketers to work at the speed required.
– Build an empowered team. Respond to changes fluidly by building a team that is able to make quick decisions based on data.
– Promote a culture of experimentation. Facilitate continuous development, allowing your teams to consistently upskill in this dynamic world, with your permission and support to test, refine, and execute at scale.