Marketing teams at many companies have launched multiple experiments with generative artificial intelligence, aiming to capitalize on the technology’s ability to transform how campaigns are designed, personalized, and delivered. Two years into adoption, these efforts have started to deliver significant results for some major brands.

Retailers, for instance, are using generative AI to create more precise customer segments, generate and test content quickly, and tailor recommendations to individual customer preferences. Those retailers experimenting with AI-powered targeted campaigns are achieving 10% to 25% higher returns on ad spending.

Consider craft website Etsy’s AI-powered “gift mode,” which incorporates user preferences to assign recipients one of more than 200 personas and offer tailored gift suggestions. Booking.com, a travel company, has expanded AI-powered features to simplify trip planning, making it easier to find the right place to stay, analyze guest reviews, and book with confidence. And consumer goods giant P&G, conscious of the gap between reported and actual consumer behavior, applies AI to analyze real-time usage data from smart products such as the Oral-B iO toothbrush, which helps it create new products or customize product lines to customers' preferences. Even companies in regulated industries such as financial services, healthcare, and telecommunications are realizing benefits consistent with legal and privacy concerns.

Still, scaling up is not easy, as the complexity of digital ecosystems and rising demand for personalized customer experiences challenge marketing teams. CFOs and CEOs, meanwhile, are pressing their CMOs to deliver more with fewer resources and to innovate faster. These factors make it urgent for senior marketing leaders to move beyond pilots and infuse their data, technology, and processes with generative AI (see Figure 1).

Figure 1

Where generative AI is paying off

Our work with leading marketers reveals that early adopters of generative AI are already reaping benefits, such as:

• Speed: Campaign time to market has been reduced by up to 50%.
• Efficiency: Content creation time has dropped by 30% to 50%.
• Return on investment: Hyper-personalized campaigns have boosted click-through rates by up to 40%.

Bain’s recent survey of more than 180 large US companies found that 27% of respondents said generative AI had exceeded or far exceeded expectations for marketing (see Figure 2).

Figure 2

For the next wave of deployment, four marketing areas offer the greatest potential, with a company’s priorities shaped more by its business goals than by the promise of the technology:

• Workflow simplification. Processes such as creative concept drafting, image production, content translation, brand compliance checks, and asset tagging are becoming more streamlined.
• Content creation and personalization. Generative AI automates copywriting, image production, ad versioning, and other creative tasks.
• Customer insights and intelligence. Real-time analytics and segmentation help marketers get more precise in their targeting, as generative AI simulates customer behavior and predicts future needs. Customer digital twins enable marketers to test concepts, products, content, and calls to action faster and cheaper than ever.
• Measurement and optimization. The technology can automate campaign performance analysis and integrate unstructured data.

These innovations are already reshaping how brands connect with customers, assuming large-scale deployment of generative AI (see Figure 3). Getting there, however, requires focused investment and resources behind a handful of high-impact battlegrounds rather than dispersing efforts across numerous proofs of concept. At the same time, marketing teams need to raise their AI literacy to achieve widespread adoption and modernize and simplify how they work.

Figure 3

Five ways to gain an edge

Our analysis of leading marketers finds that five practical steps will allow organizations to accelerate the next phase of generative AI maturity.

Commit to bold ambitions—and results.   

Too often, marketing organizations focus on individual use cases rather than broad, transformative goals. CMOs should set measurable, ambitious targets (whether operational, customer-centered, or financial) and hold their teams accountable.

One global financial services firm, for instance, aimed to reduce campaign time to market by 50%. That sparked innovations in AI-driven content creation, including revising the platform for its marketing technology stack, and embedding AI assistants throughout workflows. 

Similarly, a media company set a goal of deploying generative AI to scale up its marketing personalization capabilities and automatically generate communications with text and graphics tailored to customers’ preferences and broader shifts in the market. It achieved a 5 to 7 times increase in click-through rates for new campaigns, and the company is now embedding these practices into everyday tasks.

Focus on big wins rather than letting a thousand flowers bloom.

Early proof points build credibility and momentum, and getting fewer solutions into production will prove their scalability. This approach departs from marketers’ recent focus on experimentation and breadth of AI innovation. Start with manageable use cases that deliver quick, meaningful results, such as automating the production of direct marketing copy, social media posts, or landing pages. Another use case is tailoring content to individual customer preferences, thereby raising engagement rates. Advancing quickly to more complex use cases also leads some marketing teams to realize they need broader overhauls of the data or marketing technology stack to achieve the desired AI ambition.

These were considerations for a consumer bank that implemented a generative AI-powered creative assistant to accelerate creating personalized content for paid online search and social media campaigns. The initiative reduced production time by 75% and uncovered a 20% to 25% opportunity to increase the volume of new accounts by scaling up experiments. 

Design for users’ needs.

Adoption falters when solutions do not align with how teams actually work, or when those solutions are pushed by internal IT teams rather than marketing leaders. Broad adoption stems from marketers defining their workflows, identifying opportunities for improvement, and cocreating AI solutions with the data and technology teams. A financial services firm enlisted internal “super users” to guide development of a custom AI content-creation tool, ensuring relevance and buy-in. Those employees then helped train their peers and proselytize the technology.

At the same time, leaders should encourage their teams to reimagine entire processes and current roles rather than making incremental improvements to individual tasks.

Never stop learning and raising the bar.

While many marketers have a basic familiarity with what generative AI is, few companies have invested to equip their frontline marketing staff in using it effectively and at large scale. Training needs to be tailored to employees’ day-to-day work, showing where generative AI can complement and enhance specific roles, from creatives to data analysts.

One media company made progress with generative AI tools a standing agenda item in weekly team meetings. Marketers were trained on prompt writing and tasked with experimenting during the week, regrouping to discuss successes and refine techniques.

As teams come to understand the generative AI models, CMOs should continually challenge them to change their work behaviors. Simply issuing Gemini or ChatGPT Enterprise licenses without an ambitious playbook will not change long-standing behavior around how people get work done.

Expand the partner ecosystem.

Managing the array of marketing technology has become even more difficult as vendors jockey for position. Vendors can assume some work that used to be completed by creative and media agencies, but the vendor landscape is still messy.

The best response is to pilot marketing-tailored vendors, as many marketers have done with Adobe, Jasper, Synthesia, or Typeface. While the vendor landscape will evolve, it’s important to explore which ones currently excel in certain fields and identify best fits for scaling up. Trying a generative AI vendor should be much faster than selecting a new core marketing technology vendor. Agencies are also stepping up their innovation, and marketing teams need to keep up on partners’ innovations as most solutions will not be built in-house.

Generative AI has moved from novelty to necessity in marketing. Advanced teams are already considering what might be next: They are rethinking the partner ecosystem, preparing for the end of link-based search, and imagining what marketing to bots might entail. For those still stuck in pilot mode with a few use cases, it’s high time to scale up efforts and make meaningful improvements in productivity, personalization, and return on investment.

Related brief

Goodbye Clicks, Hello AI: Zero-Click Search Redefines Marketing

As consumers rely more on AI-based search and summaries, how will brands adapt their strategies?