This post was created in partnership with Treasure Data

In a moment when AI holds the promise of instant, always-on personalization, marketers are racing to deliver relevant, responsive experiences in real time. But driving business outcomes requires more than just plugging in new tools.

During a Brandweek 2025 session co-hosted with Treasure Data, Rafael “Rafa” Flores, chief product officer at Treasure Data, joined Ryan Joe, editor-in-chief of ADWEEK to break down how brands can get the most out of AI and drive proactive, autonomous engagement.

Fragmentation is the first barrier to effective implementation

Flores acknowledged that many assume the fear around job replacement is a barrier to AI adoption. But in his eyes, the true obstacle is rushing to implement AI technologies in an uncoordinated way, siloing operations and muddying outcomes.

“There’s a cost of fragmentation, not just with data but also with the agents themselves,” he said. “The biggest hold up is, how can I bring it all together so I can use it all in one place?”

So, how can brands move past the internal disconnection?

First, Flores advised brands to align on the specific business outcome they want AI to solve. And second, he said, companies need to collaborate cross-functionally so teams aren’t split across different tools (e.g., ChatGPT vs. Gemini).

“For change management to take place in the agentic-by-design era, everyone needs to know what role they play and what role the AI plays within their role,” he said.

Moving from predictive to proactive marketing

Once teams align on the problems and tech, the next steps should be evolving from projection-based approaches to always-on experiences.

Combining machine learning, rich data inputs, and agentic AI can unlock proactive marketing, said Flores. As an example, he shared Treasure Data’s experimentation with creating synthetic personas to gain insights about how to best target its audience segments.

The synthetic persona his team developed—Emma—helped them pressure test a customer journey to identify key touch points and opportunities to pivot that are worth considering in client marketing.

It starts with walking through Emma’s customer journey, Flores explained.  “First thing in the morning she wakes up, she grabs her coffee, and she shops,” he said. “So, how can we get in front of her? How can we get a VIP lookbook right to the top of her inbox so she can be on our website and actually click it?”

Using the data generated by the persona, Flores shared that the team was able to pinpoint when and where it was most effective to deploy push notifications and discount codes to increase the chances of conversion. AI can then apply a layer of personalization to ensure the tailored messaging based on past consumption patterns is seen.

That level of targeting and nuance, powered by autonomous AI decision-making, creates an end-to-end experience design that feels intuitive to the consumer. But, Flores noted, the learnings a brand can get from experimenting with synthetic personas are only as good as the data driving them.

Designing backward to achieve business outcomes

The discussion ended with a reality check on the guardrails and strategy required for autonomous AI marketing systems to succeed. Most importantly, Flores argued, brands should define desired results and design backward, building systems that allow their team members to focus on the pieces of the work that need deep focus, collaboration, and human expertise.

“When it comes to using AI for that collaboration, you need to sit down with all those different campaign stakeholders in the room and say, ‘OK, here’s the 80% of stuff that none of us here want to do. How can we put that on autopilot so that the rest of the time we can focus on that 20%?’”

 “It’s about actually having the time to focus on the things that matter that make a difference in your day-to-day,” he continued.

And staying realistic about a few targeted goals may be more effective than doing everything at once, he added.

“Just because AI is hyper-intelligent, it doesn’t mean that it’s going to solve every single problem you have,” he said. “So my biggest advice is have a purpose for why you want to use it.”

This post was created in partnership with HubSpot

Marketing is undergoing one of its most dramatic shifts in decades. As AI upends traditional discovery channels and rewrites the rules of creativity, brands have to rethink how they connect with audiences—and how much control they really have over those connections. What’s emerging is a new model of marketing that blends human insight with machine intelligence to build trust, authenticity, and long-term value.

During a Brandweek 2025 fireside chat co-hosted with HubSpot, the radical evolution of marketing took center stage. Jonathan Hunt, VP of media and head of The Hustle at HubSpot, explored how the relationship between marketers and AI is reshaping creativity, distribution, and brand authenticity while in conversation with Mike Beyman, chief of staff at ADWEEK.

Managing the dark moment AI created

Hunt set the stage by acknowledging the intense pressure marketers face. Traditional discovery channels are faltering, forcing a strategic crisis as brands struggle to connect with their audiences.

“It’s becoming harder and harder to be discovered by or to reach your customer. You see it through search trends and the headwinds there. You see it through rising paid media costs,” Hunt explained.

This was accelerated by the arrival of generative AI, and specifically ChatGPT. Recalling his reaction to its debut, Hunt said, “That was a very dark moment for all of marketing, especially media.”

He framed this shift not as a temporary disruption, but as a fundamental turning point. “If you look at the trend line of big technological shifts in media and marketing, ChatGPT precipitated what I think is now the current state of marketing.”

The new model is a direct response to how marketers traditionally moved in very linear and campaign-focused ways. “You can no longer afford to be that way,” Hunt explained. “You have to be a lot more iterative.”

Part of surviving this new era, Hunt argues, is breaking free from the very channels that are faltering. “We call it loop marketing, which is humans and AI coming together to create but also personalize your messaging and your content on a one-to-one level.”

The value of owning your audience

HubSpot has rejected the playbook of rented audiences on social platforms and search engines.

“What a lot of marketers are starting to find is that there’s a lot of value in owning your distribution channels as opposed to renting them and then having to invest in Meta or LinkedIn every single month and see that money evaporate,” he said.

But when companies do acquire and own their own media, he pointed to one common mistake they make.. “The instinct is, OK, I’m acquiring this media asset now, it’s going to become a megaphone for my product.”

HubSpot’s strategy is the opposite—the company actually preserves the assets it acquires. “What we see whenever we’re building out media assets or acquiring media assets is that you’re acquiring the trust in the audience,” Hunt said. “The very worst thing you could do is erode that, because then why did you buy it in the first place?”

Each HubSpot media asset is completely editorially independent, he said. “They’re focused on a specific customer profile that maps back to a product within the HubSpot portfolio, but we let our video producers, writers, editors, podcast producers, and creators take those insights and then translate that into what they feel like is most resonant with their audience.”

Still, owning the channel is only half the battle—marketers also need a new approach to what they put on it.

Using AI to improve human creativity

This new model requires collaboration between AI and humans.

Humans remain vital to HubSpot’s content-first approach. Readers are “going to people because people inherently trust people,” Hunt said. This insight drives HubSpot’s focus on video and creators. “Video as a format, it’s the currency of every single channel.”

When it comes to AI, instead of fighting it, optimize for it, particularly in search.

“What we were finding is that while LLM traffic is still just a drop in the bucket relative to referral traffic from Google, the intent is a lot higher,” explained Hunt. “The value of a click back from an LLM like ChatGPT is 3X higher than what we were seeing on search.”

The framework is simple: “How we think about it is everything has to be human-driven, AI-assisted,” said Hunt.

Coca-Cola is doubling down on generative AI. A year after its first AI-generated holiday campaign drew mixed reactions and anger, the company leaned even further into the technology this holiday season. 

Pratik Thakar, the Coca-Cola Company’s global vice president and head of generative AI, said there’s no retreating from AI.

“The genie’s out of the bottle,” he told an audience Monday at ADWEEK’S Brandweek 2025 event in Atlanta. “You either lead with it, or you keep crying about it.”

While last year’s AI-generated Christmas ad spurred backlash and anger from parts of the creative community, it turned out to be one of Coca-Cola’s most effective holiday campaigns with consumers. According to Thakar, research firm Kantar ranked it as the top-performing Christmas ad globally last year across all categories—a result that encouraged the company to continue experimenting with generative tools.

“It worked for our business,” he said. “That’s what motivated us to keep going.” Thakar added that while some in the industry criticized the ad’s visuals, such as the way “wheels weren’t moving properly,” consumers didn’t see it through that lens.

“We took that as positive feedback,” he said, noting that this year’s campaign benefitted from major advances in AI-generated cinematography, physics, and rendering quality.

Embedding AI throughout

Despite early skepticism, Coca-Cola has emerged as one of the most visible case studies of how a global brand can operationalize generative AI beyond marketing experiments.

The global beverage brand partnered with OpenAI in 2023, shortly after it launched ChatGPT, Thakar said, adding that Coca-Cola’s approach has been to embed AI into existing teams rather than build a separate department.

Thakar’s comments underscore a broader shift among brand leaders who see generative AI not as a passing trend or cost-cutting measure, but as an inevitable force reshaping how products are made, marketed, and managed. It also signals how legacy marketers are trying to turn early creative wins into long-term business transformation.

“The global Christmas campaign is our year’s biggest business opportunity. [Would we] want to do cost cutting with your biggest business opportunity? No, at least not at Coca-Cola,” Thakar said. 

Coca-Cola is also experimenting with AI through more than a dozen incubators with major tech companies, gaining early access to models to test how the technology can drive creativity and efficiency, he added. Beyond marketing, the company is now applying AI to research and development, supply chain, and knowledge management—but without creating a separate AI department. 

“We don’t want to, we shouldn’t [separate AI] because then it becomes a silo,” Thakar said.

In this special edition episode of The Speed of Culture Podcast, host Segun Oduolowu sits down with Suzy founder and CEO Matt Britton live from the Generation AI book launch in New York City. 

Together, they explore how artificial intelligence is transforming consumer behavior, personal branding, marketing strategy, and everyday life.

This is an eye-opening, forward-looking discussion on how the next generation—Generation Alpha—will grow up as true AI natives, and what that means for the future of business, creativity, and society.

Key takeaways:

(00:00) The Future of AI with Generation Alpha — Matt Britton shares insights from his book Generation AI, explaining how the first truly AI-native generation will redefine creativity, communication, and work. Segun and Matt discuss the tension between fear and opportunity as AI begins to outperform humans in certain areas, and why adaptability—not resistance—is the ultimate competitive edge.

(13:14) Parenting and Global Impacts of AI — This segment explores what it means to raise children in a world surrounded by AI. Britton emphasizes balanced guidance that encourages curiosity while setting healthy boundaries. The two also unpack how global AI competition, particularly between the U.S. and China, shapes innovation, security, and the global economy.

(26:12) AI Book Interaction and Future Trends — Britton reveals how Generation AI blurs the line between author and algorithm through its interactive QR-code chatbot experience. He and Segun discuss how AI will revolutionize how we read, learn, and tell stories, and why embracing AI creativity is essential for the future.
(29:59) Personal AI Management and Prioritization — Closing out the episode, Britton shares practical advice on incorporating AI into everyday routines, from automating workflows to creating personalized projects. His message: Intentionally engage with AI now, so you’re empowered—not overwhelmed—by what’s ahead.

The numbers

$33 billion: How much Amazon made in the third quarter from its cloud-computing business, Amazon Web Services—up 20% year-over-year and beating analyst projections. AWS now has an annualized revenue run rate of $132 billion.

$17.7 billion: How much Amazon made from advertising in the second quarter, an increase of 24% year-over-year.

13%: The amount that Amazon’s total net sales increased compared to last year’s Q3 earnings, to $180.2 billion.

$1.8 billion: What Amazon expects to pay in severance fees related to layoffs, which started this week when the company announced cuts of around 14,000 corporate jobs. Reuters and The Wall Street Journal both reported that the total number of affected jobs is expected to hit 30,000.

-69%: The drop in Amazon’s free cash flow—down to $14.8 billion versus the $47.7 billion reported in Q3 last year—largely due to capital investments related to artificial intelligence.

60%: Increase in the number of rural communities that have access to Amazon’s Same-Day and Next-Day Delivery over the last four months.

250 million: How many people have used Rufus, Amazon’s generative AI shopping assistant, so far this year. The company claims people who use Rufus are 60% more likely to complete a purchase.

$2.5 billion: The settlement Amazon had to pay this quarter after the FTC proved it misled people and made it exceedingly difficult to cancel subscriptions.

The watercooler talk

CEO Andy Jassy kicked off Amazon’s Q3 earnings call by discussing AWS, the company’s cloud-computing business, noting that the 20% growth rate is a return to a pace the company hasn’t seen since 2022. He also teased “several” new AWS deals signed in October that haven’t yet been announced.

“AWS is where the preponderance of companies data and workloads reside, and part of why most companies want to run AI on AWS,” Jassy said.

The AWS growth numbers outpaced analysts’ expectations. The company’s shares jumped as much as 14% in after-hours trading.

The company still has more demand for AWS services than it has capacity for, however. Currently, the bottleneck is in power, Jassy said, though it could at some point shift to chips. Amazon has added 3.8 gigawatts of power to its AWS infrastructure in the last year, with another gigawatt coming in Q4.

For its advertising division, Amazon highlighted deals with Netflix, Spotify, and SiriusXM. The ads business saw 24% growth from the year prior, to $17.7 billion in revenue.

The key quote

Agentic commerce is coming.

While it’s become very easy to buy things online when you know what you want, a physical store with a physical salesperson still has some advantages over ecommerce when you don’t know exactly what you want, Jassy explained in response to an investor question about agentic. But that’s going to change as AI agents get better at guiding online shopping experiences, he said.

“AI and agentic commerce solutions are going to expand the amount of shopping that happens online,” Jassy said. “That’s really good for customers, and I think it’s really good for Amazon, because at the end of the day, you’re going to buy from the outfit that allows you to have the broadest selection, great value, and continues to deliver for you very quickly and reliably.”