Human AI Partnership in Action: Team Reactivate Collaboration

2 weeks in Karachi. 7 global brands. 1 AI-powered mission. For the past few weeks, I’ve been on the ground in Karachi at Team Reactivate Headquarters, deep-diving into the heart of Unilever’s Beauty & Wellbeing (B&W) portfolio. Our mission is clear: leverage ImagineArt to move AI from a "cool experiment" to the literal engine of production. We are currently leading a pilot program designed to develop scalable workflows for creative content production across some of Unilever’s most iconic brands, including Sunsilk, Dove, Pond’s, and Tresemme. When working with such iconic brands, defined by comprehensive guardrails and distinct visual guidelines, the traditional "prompt and pray" methodology isn't enough. Random prompting might yield a one-off result, but it fails to build a repeatable, reliable system. To bridge this gap, we’ve been utilizing LLM-backed copilot nodes within ImagineArt's Workflows. Powered by tailored system prompts, these pipelines generate assets that are inherently conscious of funnel stages, brand voice, and specific visual guidelines. Overall, it has been an excercise in building creative pipelines for scale, while ensuring every output feels like it belongs to the brand. Beyond the tech, it has been incredible to collaborate first-hand with Team Reactivate to develop SOPs where AI acts as a true collaborator; proving that AI can handle the heavy lifting of design execution, allowing in-house creatives to reclaim their time for high-level creative polish and the strategic concept direction that defines a successful campaign. I'm excited to continue this work over the next month and share more insights as we turn these workflows into a new standard for enterprise creativity. 🚀

AI is reshaping the future of digital marketing. Our Digital Group Head, Sung Ju Park, shares his perspective with Campaign Indonesia on why success in the AI era is no longer about launching more campaigns, but about building smarter, integrated systems that align data, media, technology, and brand strategy. Proud to see Cheil Indonesia join the conversation. Read more here [ https://bit.ly/4b7vMG7 ]

Last week, we onboarded a shoe ecommerce brand in Nairobi. As part of our onboarding process, we run a full audit before touching anything, and one of the things we now check is how a brand shows up across AI engines, not just Google. So I asked four of them to recommend where to buy shoes in Nairobi. Google AI Overviews. ChatGPT. Perplexity. Claude. Four different answers. Four different lists of brands and stores, and our new client wasn't mentioned on a single one. And this is a brand with a decent website, real products, and customers who genuinely love what they sell. But to every AI engine fielding that question, they simply don't exist. This is the part of search that most brands in Kenya haven't started thinking about yet. Google rankings still matter. But a growing number of people are skipping the search results page entirely and just asking an AI directly. The AI recommends what it knows, what it's seen referenced, and what the broader internet has talked about enough to consider credible. If your brand isn't part of that conversation, you're invisible to an audience that's already looking for you. We've been building and testing our approach to this internally, and we're excited to keep pushing it for the brands that trust us with their growth. It's called Generative Engine Optimisation. And it's quickly becoming the thing that separates brands that get found from brands that don't.

Harvard Business Review published an article this month (Preparing Your Brand for Agentic AI) on how AI agents are changing brand-consumer relationships. The article is worth a read. One aspect made me dig deeper. Pernod Ricard, the liquor company, found out that LLMs were misclassifying one of their whiskey brands. An affordable product was being described as prestige. To get LLMs on the right track, they updated their website copy and ad messaging to be more consistent. Over time the AI models picked up on it. But why? Google is retrieval-based. It matches your pages to a query. So if Pernod Ricard had pages optimised for "affordable scotch whisky," they'd probably show up for that search regardless of what anyone else was saying about them. You control your own pages, you control your keywords, you rank for what you target. From what I understand, an LLM doesn't retrieve pages. It synthesises an impression from everything it's ingested. Every review site, every lifestyle blog, every retail listing, every Reddit thread. Whiskey as a category already carries prestige associations in culture. If most of the internet discussion about their product sits alongside that narrative, or if review sites describe it in language that sounds premium, the model absorbs that pattern. It doesn't matter that Pernod Ricard's own website says "affordable" if the broader internet says otherwise. It's a weighted average of everything written about the brand. And the brand's own communication was either too quiet, too ambiguous or too vague to outweigh the surrounding noise. Which means the fix isn't just "be clearer on your own website." It's also about shaping the broader conversation. PR, media coverage, partnerships, how retailers describe your product. All of that feeds the model. Carnegie Mellon research cited in the article makes this even more tangible. They found that swapping a few synonyms in a search prompt changed brand recommendations by up to 78%. With Google, that's manageable. You can optimise for keyword variations, buy ads against synonyms and build landing pages that target different phrasings. With an LLM, you get none of that. It gives the user an answer. Often a single recommendation or a very short list. There's no page two, no ad slot (not yet at least) and you don't have analytics that show you what prompt the user typed. Which means you have to be clear about what you say. You have to build up your ‘Brand Voice’ and train everyone on it. Because even though your own communication is only one input among many, it’s the one you control. The more consistent you are, the more consistent the articles, posts and ads that pick up on you will be. And that will ultimately lead to the LLMs to pick up on it, too.

The 2025 Global Brand Power Shift AI Leaders, Platforms, Consumer Icons The latest Kantar BrandZ 2025 insights reveal a decisive shift in the global brand landscape. Two powerful developments are shaping brand value today: the emergence of new technology driven leaders as well as the accelerated rise of platform based ecosystems. One of the most notable entries is ChatGPT (#60) with an estimated $43.6B brand value, marking one of the most significant brand debuts in recent years. Its appearance reflects the rapid global recognition of generative AI across consumer environments, enterprise solutions, moreover digital services. Alongside ChatGPT, several other companies enter the ranking as newcomers. Stripe (#85) highlights the growing importance of financial infrastructure powering the digital economy. Chipotle (#86) demonstrates how strong consumer engagement can translate into substantial brand value. Booking.com (#89) reinforces the scale of global travel platforms, whereas Hilton (#95) shows the continued strength of hospitality brands. Uniqlo (#97) further underlines the global relevance of apparel innovation, while DoorDash (#99) represents the continued expansion of delivery platforms. Several brands also return to the ranking as re entrants. Spotify (#76) reappears as one of the most influential digital media platforms worldwide. Meituan (#80) highlights the scale of integrated local service ecosystems in Asia. Morgan Stanley (#90) illustrates the enduring strength of financial institutions, likewise Agricultural Bank of China (#93) reflects the importance of large global banking brands. Xiaomi (#96) again confirms the influence of consumer technology manufacturers, while Adidas (#98) returns as one of the most recognizable global sports brands. Beyond the new entries, the Top Risers category reveals extraordinary brand value acceleration across technology, retail, financial services together with digital media. Leading the ranking is NVIDIA, recording an exceptional 152% growth, emphasizing the central role of AI infrastructure. Huawei (142%), VMware (119%), also Xiaomi (103%) further demonstrate the expanding influence of technology companies. Major digital platforms continue their growth trajectory. Instagram (101%) along with Facebook (80%) illustrate the scale of global social ecosystems, whereas Netflix (54%) reinforces the strength of digital entertainment platforms. Retail brands remain equally influential. Walmart (72%), Costco (67%), Sam’s Club (57%), Amazon (50%), additionally Mercado Libre (52%) demonstrate how scale, logistics excellence together with strong consumer reach translate into brand value growth. Enterprise technology also shows strong momentum through SAP (66%), ServiceNow (57%), Dell Technologies (53%), moreover Oracle (48%). Financial services brands including American Express (66%), J.P. Morgan (57%), likewise Chase (54%) further highlight the importance of trust combined with digital innovation.

What if we cast FMCG products like the heroes of a Bollywood blockbuster? Take a look at this concept. We didn’t just put three brands next to each other; we mapped them to raw, explosive human emotions: 🟢 हौसला (Courage): Mountain Dew. For taking the leap. 🔵 ईंधन (Fuel): Red Bull. For surviving the grind. 🔴 बदला (Revenge): Jolochip. For waking up your senses (or daring your friends). It’s loud, it’s cinematic, and it got me thinking: Is this the future of how we sell in India? Here is why this "State of Mind" conceptual marketing is a game-changer for the Indian market and the Quick Commerce space: 🔥 1. The Indian Market Runs on Emotion, Not Just Utility We are a culture that thrives on narratives. The modern Indian consumer is navigating a high-pressure, high-ambition environment. When you package products as an "Archetype" (The Hustler, The Daredevil, The Night Owl), you aren't just selling flavored water or spicy chips. You are selling a psychological armor. You're giving them the exact vibe they need to conquer their day. 🛒 2. The Next UX Frontier for Quick Commerce The 10-minute delivery problem is largely solved. The next evolution is in the discovery experience. Right now, Q-commerce apps categorize by "Beverages" or "Snacks." It's functional, but boring. Imagine the user experience if categories were built around intent. Instead of searching for three different items, what if a user could just add a pre-curated "Midnight Code-Crunch Kit" or a "Match-Day Adrenaline Bundle" to their cart with one tap? It reduces cognitive load, feels highly personalized, and completely transforms the shopping journey. 📈 3. The Ultimate Business Hack: Bundling for Margins Creating these trend-driven, emotional bundles is a massive win for business metrics: Skyrocketing AOV: You turn a ₹40 impulse click into a ₹400 basket instantly. Trend-Jacking: When a viral trend hits (like a massive cricket final or a new gaming release), agile brands can instantly deploy these themed visual bundles on delivery apps, capturing the hype in real-time. Cross-Pollination: The Mountain Dew drinker gets introduced to the Jolochip challenge. Brands win, platforms win, and the consumer gets a memorable unboxing experience. We need to stop selling shelf space and start selling mind space. What do you guys think? Would you be more likely to buy a generic "Snack Combo" or a high-octane "Endhan + Hosla" bundle on a Friday night? Let’s debate! 👇 #BrandStrategy #QuickCommerce #UXDesign #IndianMarket #MarketingTrends #ConsumerPsychology #CreativeDirection

Waikay introduces Topical Presence, a groundbreaking metric to assess brand visibility in AI-generated recommendations. - Waikay's Topical Presence metric gives brands a competitive edge by revealing gaps in AI associations compared to competitors, enabling strategic optimization of AI-generated recommendations. - This tool helps brands improve AI visibility, ensuring they contribute accurately to AI-generated conversations and foster more reliable information sharing in digital spaces. - Waikay's Topical Presence works by analyzing depth, breadth, and concentration of topic associations in AI training data to measure brand visibility across commercial topics. - Waikay's new metric maps what subjects AI associates with your brand, offering a novel way to understand influence beyond traditional search analytics.

AI slop—low-quality, mass-produced automated content—is one of the greatest threats to brand trust, according to a report from Klaviyo. “The honeymoon phase with AI is officially over for shoppers across Asia Pacific. Although consumers in the region lead the world in AI adoption, they have one of the highest bars for authenticity,” said Marcus Rossato, head of marketing APJ at Klaviyo. “For younger audiences and daily users, generic AI content isn’t just ineffective — it actively damages brand equity.”

A comment I came across recently about brand research made me think. The argument was that asking people questions about brands might have made sense a few years ago, but that today it is essentially obsolete because AI can tell us everything we need to know. I found that a rather curious conclusion. Of course, AI is transforming marketing and research. We can now analyse behaviour at an extraordinary scale, model potential scenarios, identify patterns in data that would previously have taken years to uncover, and even build synthetic panels that behave remarkably like real consumers. Anyone working in marketing today who ignores AI is quite simply missing a major part of the toolkit. 👉 But...the idea that this somehow removes the need to ask people what they think or feel about brands misunderstands what a brand actually is. A brand is not just behaviour. It is not simply a transaction, a clickstream, or a purchase. Brands live in people’s minds, in the associations they form, the trust they build, and the meaning they attach to companies and products. Behaviour can tell us what people do, but it rarely tells us why they do it, what they feel about it, or how the brand behind the action is positioned in their thinking. That dimension still requires human input. None of this means AI is a passing trend. Quite the opposite. AI will profoundly change how research is done, and it already has. Used well, it allows us to combine behavioural signals, modelling, and human insight in ways that were simply impossible not long ago. In other words, it makes good research better. 👉. But...it should augment science, NOT replace it. Every technological wave tends to produce a moment where we assume the previous methods must now be obsolete. We saw this when the internet arrived, we saw it again with big data, and we are seeing it once more with AI. In reality, what usually happens is something more interesting: the tools evolve, but the underlying questions remain remarkably similar. In branding those questions are still about perception, meaning, and trust, and that means that understanding how people experience a brand will remain central, regardless of how advanced the technology around us becomes. Much of the work now happening in modern brand measurement, including what we are building with brandr Index, is really about finding the right balance between these worlds: combining behavioural signals and AI capabilities with structured human insight to understand how brands actually live in people’s minds.

Brands are losing customer trust. Customers want authenticity. Not exactly surprising, but people can spot AI-generated content. I'm not saying AI is bad for content generation. It's just bad as a total solution for content. I certainly use it. But I use it as a starting point, not an end product. Use it to save time, not to be a cut-and-paste solution. Everyone knows what Claude sounds like. They need to know what you sound like. https://lnkd.in/e8Q_MDMp