Greetings earthlings,

Here’s your regular dose of creative AI insight from me: practical, and built for agency and brand leaders navigating real change. Whether you’re getting ready for summer holidays or stuck in back-to-back calls, this edition is packed for those looking to get ahead and level up for H2.

June was a huge month on the events circuit. Jules ran a masterclass at the AI Summit London, while I chaired a panel with AI thought leaders from WPP, eight&four and Brave Bison. Then we hit up Cannes for a week followed by MAD//Fest, all while continuing to be incredibly busy running our AI Accelerator programmes. Yes, reader - we are still recuperating!

Inside this edition we bring you all the insight on how agencies and brands are already using AI, and how to stay credible, competitive, and ahead of the curve.

And today we are also launching our AI Accelerator for Small Agencies (fewer than ten people) - we’ve had so much demand for this and are really excited about it. If you want to join a cohort but receive the same incredible quality, with access to the same coaches who work with our large agency and brand clients then read on.

• 1. How to use AI for competitive positioning

• 2. Small agency, big ambition? This one’s for you

• 3. MAD//Fest: how agencies innovate with AI

• 4. Podcast drop: What's new in Creative AI July

• 5. AI News and market moves

1. How to use AI for competitive positioning

The message from everything we’ve seen this year is the agencies really moving forwards with AI don’t see it as a technology challenge, but a business transformation challenge. They align AI with their strategic goals and brand positioning, build structured workflows, invest in team culture, ensure strong AI governance, and focus on creating new things with AI — not just cutting costs.

As BDO recently put it: “AI maturity is starting to become a market differentiator” and Cactus just published in M&A in agency land this year, where they state: “buyers are increasingly looking at [an agency's] operational readiness for AI.”

Want a valuable agency - to a buyer, a client or even a prospective employee? It’s not about if you’re using AI - it’s about how well you’ve embedded it. AI-native workflows, trained teams, and strong governance are all starting to become must-haves now.

Innovation will set the pace

I see innovation as the real edge right now. Yes, automation can cut costs - but it can also lead to commoditisation. Augmentation, where people use AI tools to speed up and improve their work, is quickly becoming the baseline. By next year everyone will be doing it. It’s innovation that creates lasting differentiation - the kind that shows up in new services, distinctive experiences, and unexpected ideas. But innovation doesn’t just happen. It needs the right people, the right processes, and clear priorities to thrive.

Rethinking data - everything is now a strategic asset

When I talk about “data” these days, I don’t just mean performance reports or spreadsheets. I’m including client briefs, moodboards, voice notes, transcripts - even those scrappy whiteboard photos. Every email thread and project folder holds insights.

And the big LLMs are helping you leverage this. Microsoft, Google, and OpenAI now plug directly into your SharePoint or Google Drive. But if your data isn’t structured and accessible, it’s as good as lost. You need to treat unstructured data like a core asset - something to be captured, organised, and used to train tools, improve our outputs, and unlock your institutional knowledge. 

Don’t just use AI - restructure for it

The tools are here. The challenge is what happens when:

• AI generates 200 versions of the campaign

• Making things is now 10x faster, but thinking still takes the same time

• People’s roles shift from makers to editors

As much as you’re thinking about which tools to adopt, you need to be thinking about how your team is structured to use them. Who owns the outputs? Who quality checks? Who turns speed into strategy? Have you equipped them with the skills they need? Are you giving time and training to keep those skills cutting-edge? New roles, continuous learning, agility, and governance will define who thrives.

If you need some support on these areas of AI organisation transformation - that's exactly what we created AI Accelerator to help with. Find out more.

2. Small agency, big ambition? This one’s for you

Running a small agency means juggling a thousand things at once – and making space to ‘learn to use AI properly’ often falls to the bottom of the list. That’s why we’ve created a new version of our AI Accelerator, tailored specifically for creative agencies with fewer than 10 people.

You'll learn alongside other agencies facing the same challenges, building serious AI capability without breaking the bank or overwhelming your already-stretched team. From leadership to design, strategy to client service, this is about transforming how your agency works, competes, and grows – with expert support and real-world application every step of the way.

3. MAD//Fest: how agencies innovate with AI

Creativity rewired at R/GA

At MAD//Fest a couple of talks gave impressive examples of agencies innovating with AI. First up Nick Pringle from R/GA challenged the traditional notion of what it is to be a creative with a powerful message: the future belongs to those who don't just make things, but who build things that make things. What a brilliant way of putting it.

R/GA is embedding AI across all operations - it's now integrated into how they design, pitch, and collaborate. For instance, they aren’t delivering decks anymore - they’re building prototypes. Achievable in the same time, but far more impactful and engaging.

They've also developed ‘Smart Story Engines’ that generate personalised stories at scale, flexing creative in real-time across platforms using brand frameworks. And they've replaced static brand books with Living Brand LLMs - custom-trained models that generate copy in the brand voice, produce visual assets aligned to brand principles, and even respond to contextual prompts like "what would our brand say about this news event?"

In Nick’s view, everything is becoming adaptive. Websites, campaigns, and even brands are no longer static — they evolve dynamically, shaped by real-time data and user interaction.

His advice to agencies? “Learn, unlearn, and relearn” - and do it without shame. Being a master of your craft now means continuous learning.

OLIVER’s synthetic research with AI personas

Also at MAD//Fest I watched Alec Barr of OLIVER introduced Seance, OLIVER’s internal AI research system for AI focus groups. Here's how it works:

• Brief in: A researcher inputs a core question

• AI Recruiter Agent: Builds synthetic personas

• AI Moderator Agent: Facilitates dynamic, reactive discussion

• Live Sessions: Personas interact and respond to stimuli like real people

• Human Synthesis: Researchers extract themes and prompts

It’s an excellent example of how the real power of AI lies in orchestrating agents to work together as a team. And AI focus groups change the economics and pace of insights dramatically. They can gets you 85% of the way to the answer, and correlate closely to human focus groups.

How Rightmove's CMO is embedding AI 

Emma saw Matt Bushby, Rightmove CMO being interviewed and it was refreshing to see a brand given the bandwidth to innovate and do new things with AI and now simply focus on efficiencies and cost-cutting. Their approach covers:

• Avatars for virtual training

• Creative campaign engagement

• Productivity tools for partners

Some examples:

• AI Training Avatars (Internal Training)
  Creating video-based learning modules using AI-generated avatars of their own senior trainers. Shortform (1–2 min) content focuses on FAQ-style topics and is delivered via their LMS or onboarding flows.

• 'Dream Home' AI Generator (B2C Campaign)
  Users enter prompts like “flat in London, big windows, dog-friendly.” AI generates whimsical, shareable dream home images - combining aspiration with entertainment and boosting session time and social shares.

• Closed LLM Environment (Internal Teams)
  The 60-strong marketing team have access to Copilot and Gemini, with AI champions in each team tasked with experimentation, documenting use cases and failures, and sharing learnings, with a focus on both time-saving and creative support.

• Partner-Facing AI Assistants (B2B Tools)
  In pilot: tools for estate agents to streamline listing uploads, write better property descriptions, and create pitch decks - benchmarked against human output to track quality and efficiency.

This is a strong example of thinking through the automation, augmentation, and innovation opportunities — the AAI framework we use throughout Spark’s AI Accelerator. It shows how to spot efficiency wins, design better hybrid workflows, and create space for bold, differentiated work powered by AI.

WPP’s multi-agent systems

Chief AI Officer at WPP, Daniel Hulme has outlined how WPP are developing a network of collaborative AI agents that work together across brand, audience, and internal workflows:

Brand Brains: LLMs fine-tuned on tone, values, product detail, and positioning to generate content that’s fast, consistent, and on-brand.

Audience Brains: Simulated personas built from psychographic and behavioural data to test content resonance before going to market. E.g. “How would a values-led millennial mum respond? What about a disillusioned Gen Z creative?”

Digital Twins: Internal models trained on employee workflows (emails, schedules, feedback) to optimise project resourcing and talent allocation.

This is the future: AI not as a tool, but as a team of collaborators. “Marketing is becoming a choreography of intelligences. If you don’t have a position, you’ll be left out of the room.”

4. Podcast drop: What’s New in Creative AI July edition

Last week we recorded our July podcast and it’s a bumper one. In this episode:

• Our takeaways from Cannes Lions 2025

• More real stories from agencies experimenting with AI

• An early review of ChatGPT Record

• Tips on choosing the right LLM for your business

Whether you're walking the dog or sketching wireframes, it's a great way to stay up to date. Watch, listen, or skim the highlights — and tell us what you want us to cover next.

Spotify | Apple Podcasts | YouTube

5. AI News and market moves

Channel 4 launches AI-generated ads: More info

Google explores AI ad auctions: Read more

Grammarly acquires Superhuman: Read here

Getty vs Stability AI: Latest on the UK lawsuit: Details

Midjourney V1 video: Highly anticipated video engine launches: Read more 

US Clean Energy for AI: In Trump’s firing line: Explore

CapCut (same company as TikTok) T&Cs update: Significant IP implications: Review the terms

Thinking of taking one of our programmes later this year?

We’re now booking our tailored AI programmes for the autumn. So if Spark’s even a twinkle in your eye, give us a wave early—we really hate saying no when our team’s availability disappears. Book a slot in my diary if you’d like to have a chat and get on our radar.

That’s all for this week. We love to have your feedback - please hit a button below and let us know how you found todays issue.

See you next time!

Emma

Co-founder of Spark AI

About Spark AI

We help creative and marketing leaders turn AI curiosity into confidence—through hands-on training, strategic support and tailored roadmaps for your team.

Whether you're just getting started or scaling fast, Spark helps you lead with AI on your terms.

👉 Explore AI for Agencies | Explore AI for Marketers

Not a subscriber yet?