AI was a popular topic at the Spring 2025 MarTech Conference in late March. While we didn’t search through the transcripts of every session and panel discussion, AI was likely mentioned in every single one.
Three conference sessions were devoted entirely to AI, two of them were Coffee Talks where attendees could chat with the speakers.
The three sessions were:
- Real-world marketers sharing real AI success stories with Kendall Davis, global head of display, Google; Channan Sawhney, global Amazon customer leader, Kenvue; and Sarah Weiss, vice president, head of marketing, Qvest.
- Using generative AI tools for content creation with Lisa Peyton, an immersive and strategic communications instructor at the University of Oregon.
- Embracing agentic AI with Christopher Penn, chief data scientist, TrustInsights.ai and Constantine von Hoffman, managing editor, MarTech.
The combined knowledge of the speakers and the attendees for these sessions led us to five conclusions about AI and marketing.
1. Start small and scale for real impact
Rather than attempting to overhaul your marketing immediately, pick one specific challenge — maybe try creating targeted ad variations or automating repetitive tasks — and test AI-driven solutions there.
Why it matters: This lets your team learn quickly, demonstrate tangible ROI and systematically scale up the most successful use cases to more significant initiatives.
Example: Use AI to analyze a single high-priority campaign’s creatives, see what resonates most with your key segments, and optimize further from there.
2. Combine human expertise with machine efficiency
AI excels at processing vast amounts of data, surfacing trends and automating labor-intensive tasks. But the “human in the loop” is essential. Marketers must still provide strategy, creativity and an understanding of audience psychology.
Why it matters: Marketers worry AI might replace them, but AI works best when human intuition, brand knowledge and empathy guide the machine’s outputs.
Example: Have AI generate multiple versions of creative copy but rely on your marketing teams to refine tone, voice and brand authenticity.
3. Measure efficiency gains and performance lift
When evaluating AI’s effectiveness, don’t just look at performance metrics (like conversions or revenue). Track efficiency metrics such as time saved, cost per acquisition improvements or reduced creative-production overhead.
Why it matters: Stakeholders need business results and operational ROI. AI can free teams from manual work, channeling the hours saved toward higher-value strategy and innovation.
Example: Compare AI-assisted campaign outcomes to “business as usual” campaigns, quantifying the uplift in click-through or revenue and the production time and media spend saved.
4. Lean into hyper-personalization and real-time adjustments
Use AI to generate or adjust messaging, visuals and offers based on real-time signals — such as browsing behavior, device used or time of day — so that each consumer touchpoint feels uniquely relevant.
Why it matters: Consumers increasingly expect tailored experiences; AI can combine many data signals to deliver more timely, individualized offers than standard segmentation can.
Example: If a consumer shows repeated interest in a type of skincare product, deliver a dynamic ad featuring that exact product line, potentially with an immediate promo code during peak purchase hours.
5. Build partnerships and data strategies to stay agile
AI-driven marketing works best when you combine your data (e.g., on customers or past campaigns) with partner data (e.g., from retailers or digital platforms). However, success depends on having the right processes — clean data, robust APIs and organizational buy-in.
Why it matters: AI’s accuracy and relevance hinge on fresh, high-quality data. Collaborating with tech providers or retailer partners can supercharge your targeting and insight generation.
Example: Tap into your retail partner’s e-commerce behavior data — like purchase frequency — to create AI-driven re-engagement campaigns that nudge lapsed customers with curated product suggestions.
Learnings for marketing leaders
Integrating AI with business goals and measurable outcomes is key, whether you’re optimizing ad creative, testing new products or tailoring full customer journeys. Keep humans in the driver’s seat for strategy and brand voice, and let AI handle repetitive tasks and pattern-finding. By starting small, building cross-functional partnerships and focusing on measurement, you’ll see faster, more reliable gains that can scale across your entire marketing operation.
A critical insight from each AI session
We’ve made all three AI-focused sessions available below without registration. But if you’re short on time, we’ve chosen to share a critical insight from each session.
Using generative AI tools for content creation
Key Insight: Leverage “meta prompting” to let AI help write the prompts for you. In other words, have one AI model (like Claude) generate high-quality prompts for another model (like GPT-4). This approach saves time, produces more targeted instructions and yields better AI-driven outputs.
Embracing agentic AI
Key Insight: Distinguish between mere automations (“done with you”) and true AI agents (“done for you”). True AI agents handle tasks autonomously without you overseeing each step, whereas automations require direct participation. Knowing the difference prevents overpaying for “agents” that are just advanced automations — and helps you plan for where full autonomy makes business sense.
Real-world marketers sharing real AI success stories
Key Insight: Use AI to personalize at scale and measure efficiency and performance lift. The marketers in this session found success by running A/B tests comparing AI-optimized campaigns to traditional ones and quantifying improvements in conversions or ROAS and time/cost savings. Hyper-personalization efforts driven by AI can unlock better campaign outcomes and operational efficiencies.
You can watch these three sessions in their entirety below.
The post 5 things we learned about AI at the Spring 2025 MarTech Conference appeared first on MarTech.