Generative AI isn’t just ready for primetime—it’s already making significant contributions to marketing and generating an enormous amount of attention along the way. This has led most brands to rightfully ask not “Should we use generative AI?” but “How?” and “How quickly?” While generative AI has enormous potential, it also carries significant risk. It can be costly to implement, it can significantly disrupt business processes that have been in place for decades, and it raises many questions about privacy, intellectual property and equity that will take years to resolve.

This is why ZS takes a relatively pragmatic view of generative AI in marketing. It’s here to stay, but brands should approach it methodically. We think of generative AI adoption in three phases. The initial phase, now underway, is a period in which brands should begin to enable existing marketing processes with more off-the-shelf generative AI capabilities. Success in this phase is primarily measured by productivity gains as brands experiment and explore what works best for them. As marketers gain experience and traction with the technology, we see brands entering a phase of generative AI-powered marketing. Here they may begin build more of their own capabilities internally and focus on more disruptive use cases that would drive meaningful growth. In the third phase, generative AI will unlock new marketing models, promising significant competitive gains with proprietary applications and tools that echo the transformative impact of digital marketing two decades ago.

What does this mean for marketers? Where classical AI has been successfully applied to tackle routine tasks like media plan optimization, today’s professionals assign generative AI to more complex duties. It can adapt and respond to human prompts, sense and understand intent and create text, images, videos and speech. The technology’s sophistication is expanding daily, challenging preconceived notions about its limitations. Generative AI allows marketers to focus on being creative and spend less time on tedious tasks.

Recent ZS-led surveys found that while 94% of senior marketers are moderately familiar with generative AI capabilities, only 15% report regularly using it. The primary obstacles include perceived complexity in getting started, residual skepticism about how much difference generative AI can make for their teams and concerns about having the right IT systems, policies and governance in place.

To effectively implement generative AI, we advise marketers to assemble a toolkit tailored to their consumers, categories, competitive set and internal team needs. In developing a 2024 toolkit, we evaluated use cases against several key criteria that prioritize:

• Return on investment with growth in top-line revenue and bottom-line profitability
• Proven methodologies that can deliver near-term results and adapt over the long term to integrate advances in generative AI
• Opportunities to work with multiple potential partners with industry-specific insights and skill sets
• Integration within the existing corporate structure vs. an organization-wide transformation

Guided by this criteria, we identified four focus areas where generative AI can drive marketing impact.

In the modern digital world, consumers leave massive volumes of feedback across traditional channels and e-commerce platforms. With generative AI, marketers can summarize, analyze, categorize and draw conclusions, at scale, from this vast array of data. They can, for example, analyze consumer reviews to assess intent and sentiment and assign keywords to various analytics-friendly buckets. This nuanced understanding of consumer satisfaction helps brands tap into previously untapped data sources to address growth opportunities, craft unique brand positioning and identify consumer trends to support product development. In the past, these efforts might have required lengthy and costly research projects, but generative AI offers a more streamlined yet comprehensive approach.

An established chocolate brand faced a challenge from Feastables, a smaller competitor launched by YouTube influencer Jimmy Donaldson (aka Mr. Beast). Traditional retail data and conventional research methods fell short of assessing the newcomer, even as it gained momentum. The brand used a generative AI model to analyze consumer reviews, finding that while Mr. Beast was an exciting persona that led to sales, the chocolate bars themselves were disappointing. The competitor was not a long-term threat; rather it revealed a missed opportunity for the brand to explore partnerships with other pop culture celebrities.

Efficiency gains are substantial. Insights can be delivered up to four times faster than traditional market research methods. Because fielding times aren’t needed, analysis is accelerated and costs can be reduced by as much as 50% because external data is relatively inexpensive. Moreover, generative AI solutions often yield better quality insights, with increased sample sizes, even up to the millions, that provide more granularity. The absence of a structured questionnaire allows for uncovering unknowns and reducing bias, making the insights more operational.

Generative AI “agents” represent a significant advancement in accelerating product development. These agents are large language models trained on extensive data sets and customized to deliver user-friendly information and a conversational experience. They can be applied at various points in the innovation cycle to accelerate and improve outcomes.

For example, an AI agent could be trained with data about consumers, categories, competition, product attributes and packaging qualities. This agent could then be used during sessions with multifunctional teams to mimic consumers or competitors in roleplay to project future actions they might take. Also, an AI agent trained to conduct a preliminary legal or regulatory review of claims ideas based on U.S. Food and Drug Administration guidelines could help weed out nonviable options and suggest alternatives.

The versatility of generative AI is evident. One agent can perform multiple tasks with the right data set and training, and real-life usage has demonstrated substantial efficiency gains in the innovation cycle. The overall process moves three times faster due to reduced time spent gathering and analyzing data, more productive brainstorming and faster evaluation of ideas. Costs are reduced by up to 30% because of less rework and retesting. The end result is a product that better meets consumer needs.

A protein bar brand was losing market share to key competitors. Consumer research identified issues with its products and packaging that were leading to sales loss at the shelf. With generative AI, the brand equipped its marketing team to develop product innovation ideas, including 30 pre-validated concepts for testing that were created four times faster than in the past. The brand not only reversed its declining sales trend but also gained insights that informed future product strategy.

Digital marketing ushered in the need for more content that could be tailored to the channel and consumer microsegments. Resource constraints, including time and cost limitations, however, stood in the way. Generative AI solutions now exist that can create content tailored to specific brand requirements, for example, with a tonality that will appeal to Millennials vs. Gen Z consumers or adhere to a specified copy length. The intent isn’t to replace human oversight but rather to enhance existing creative processes. Generative AI content production applications span the myriad communication channels marketers use—automated creation of text, images, videos and interactive outputs such as avatars and three-dimensional worlds.

The results from clients using generative AI for content production are impressive. Marketers can use generative AI to increase the speed and quality of agency briefings, enabling them to iterate and test creative content more effectively. This leads to an 80% faster delivery to the market, potential cost savings through less rework and reduced external dependencies. While human oversight remains crucial to ensuring quality, creativity and representation, generative AI allows for substantially greater content personalization.

A quick-service food retailer achieved significant revenue growth by using generative AI to personalize email content based on consumer segments. Previously relying on a small assortment of fixed email messages, the retailer leveraged generative AI to create a large database of custom headlines and offers based on consumers’ online history and interactions. The result was a substantial increase in click rates, offer redemptions and a revenue boost of more than $5 million.

Generative AI tools promise to improve personal productivity by automating routine administrative tasks. Several off-the-shelf tools already are available and more are regularly introduced—including ChatGPT and Microsoft Copilot. Customized solutions tailored to an organization’s data and unique work processes also are becoming available, and the potential applications are vast. These tools can assist with a variety of assignments, from drafting presentation slides to managing calendars and writing spreadsheet formulas.

On average, employees spend two days a week completing routine administrative tasks, leaving two-thirds of employees feeling they don’t have enough time in the workweek to do their jobs. Generative AI tools can increase personal productivity by four times, improving the speed of writing tasks by 60% and increasing quality by 20%.

Consultants often engage in extensive desk research. A common task, for example, involves reviewing industry financial reports and analyst call transcripts for clients and their key competitors to better understand trends and opportunities. This process had traditionally been a manual one requiring a significant investment of time to read the material and write summaries. At ZS, we deployed a generative AI application that resulted in a 70% reduction in activities considered tedious by our teams and a 50% faster turnaround time. We not only streamlined the process, but we also freed our people to focus on value-added, meaningful work, contributing to increased satisfaction.

At this stage, we encourage marketing teams to consider the productivity-enhancing use cases that will create the foundation for more revolutionary work in the coming Generative AI-Powered Marketing and Generative AI-Transformed Marketing phases. While generative AI’s long-term role in marketing is still coming into focus, its ability to significantly enhance productivity across core functions is already clear. Success will come to the organizations that best equip their teams with the right capabilities and then give them the freedom to explore. 

We hope you’ll think of this toolkit as a compass, guiding your marketing teams toward innovation, strategic growth and efficiency. Adopting generative AI represents more than a technology implementation. It presents an opportunity to transform the work of marketing and open a new era of creativity led by real-world insights and realized through more personalized and engaging content. Marketers who integrate generative AI into their processes—along with a readiness to test, learn, iterate and adapt—position themselves not only for competitive advantage but also the chance to lead their industry.