Podcast Summary: The Dave Gerhardt Show (Exit Five)

Episode: Inside Ramp's Marketing: Creative Bets, Measurement, and AI Agents (with Drew Pinta)
Date: April 13, 2026
Host: Dave Gerhardt
Guest: Drew Pinta, Director of Growth Data at Ramp

Episode Overview

This episode dives into the innovative marketing engine at Ramp, a company widely recognized for both creative stunts and rigorous, data-driven growth. Dave Gerhardt sits down with Drew Pinta to explore how Ramp balances bold brand bets (like putting an accountant in a glass box in NYC) with disciplined measurement, incrementality testing, and an aggressive rollout of AI tools. The conversation covers everything from marketing attribution and budget allocation to how Ramp's marketers use AI agents to reinvent their roles. The episode further addresses the future of marketing in the age of AI, making a bull case for marketers but also warning about rapid changes ahead.

Key Topics & Insights

1. Drew’s Role and Background

• [04:59] Drew leads the growth data function at Ramp, acting as a strategic partner to marketing.
  • “Everything from like measurement to strategy is kind of in my team's wheelhouse.” – Drew Pinta
• He started in economics at the Federal Reserve, realized the slow pace wasn’t for him, and found his fit in data science for growth teams.
  • “There's actually like a ton of parallels between growth and marketing and economics... You're trying to model a really complicated system.” – Drew Pinta [05:29]

2. Ramp’s Marketing-Data Partnership

• Drew supports marketing by providing measurement insights and helping allocate budget.
• His philosophy: Marketers focus on brand and creativity, while data supports strategic decisions.
  • “Our most important role is like, where should we put money?” – Drew Pinta [07:20]

3. Measuring Creative Marketing Bets

• The Glass Box Stunt: Ramp put an accountant in a glass box for 12 hours as a brand campaign.
  • They used “event studies” to observe the campaign’s effects by comparing forecasted vs. actual outcomes post-campaign.
  • Supplemental: Letting an employee get married inside the box for real [01:54]
• Leveraged AI (LLMs like Claude) to scan 10,000+ Gong sales calls and identify qualitative mentions of the campaign, podcast ads, etc.
  • “We pipe all of that through an LLM and just be like, tell me every time a lead or customer mentioned the box.” – Drew Pinta [12:03]
• Notable Quote [15:43]:
  • “The fastest way to kill creative marketing ideas is to try to measure them like direct response... Measurement should meet the marketing, not vice versa.” – Drew Pinta

4. Incrementality Testing & Marketing Attribution

• Drew advocates forming hypotheses for campaign impact and testing them, tailored to each channel.
  • Example: Use “geolift tests” for billboards (compare regions with/without ads), A/B splits for direct mail or digital.
  • “Incrementality testing” seeks to isolate and prove the causal effect of marketing spend—core for justifying larger budgets.
  • “If you're doing something like direct mail, it's pretty straightforward to hold out 10%... [But] for podcasts, it’s way harder...” – Drew Pinta [17:01]
• On Attribution:
  • Scanning Gong calls yielded a surprising insight: LinkedIn was mentioned 3x more by prospects than Meta, despite models suggesting the opposite.
    • “I wouldn't trust any sort of deterministic attribution model like MTA or Last Touch or anything like that, especially today with all the privacy stuff.” – Drew Pinta [24:56]
  • Attribution models are signals, not gospel—useful for in-channel optimization, not cross-channel budget allocation.

5. Marketing Budget Allocation & Channel Testing

• Emphasizes the necessity to test channels with real spend (even with limited data) rather than over-planning.
  • “Start spending and see what works based on tests... The only way to do that is to go spend and get some data points.” – Drew Pinta [26:20]
• Ramp uses a 70/30 split—70% on core/proven channels, 30% dedicated to future bets and experimentation.
  • “Having an experimental budget, having that worked out with finance is key. You need a budget... to take bets and not have it tank your overall numbers.” – Drew Pinta [28:36]
• Channel Maturity Framework:
  • Channels advance from nascent (baseline reporting, some signal in sales/customer input) to mature (validated through incrementality/causal testing).

6. AI Transformation – From Marketer to Builder

• Ramp’s marketers now use tools like Claude Code to automate dashboards, analyze calls, build agents, and even launch verticalized marketing campaigns.
  • “Marketers are shifting from running campaigns to building. They are becoming builders, enabled by cloud code...” – Drew Pinta [46:59]
• Agentic Workflows:
  • Example: The “Vertical Machine” agent automates research and execution for entering new industry verticals (e.g., construction, healthcare), producing messaging, landing pages, SEO, and sales material at the click of a button.
    • “[The agent] does a bunch of research... scans sales calls... goes into each channel and figures out how to market to this vertical... spins up hundreds of pages on our website... all with the click of a button.” – Drew Pinta [50:29]

Enablement & Culture

• Executive mandate: CEO encouraged staff to “drop the ball” on existing work to learn and experiment with new AI tools.
  • “Our CEO...said: I'm okay with you dropping the ball a little bit on your...I want you to invest time in this.” – Drew Pinta [53:45]
• Bottom-up approach: Identified “AI-pilled” team members, made them evangelists to demo and share workflows, spread ground-level adoption. [54:34]

7. The Future of Marketing & AI

• Bull Case:
  • LLMs output the median—marketing wins from being creative and unique, which humans are still best at.
    • “In marketing the median is death...the alpha is being creative and having taste and things like that that I don't think AI is great at today.” – Drew Pinta [40:03]
  • Marketers are better-positioned than engineers for this future, because creativity/taste still matter more than standardization.
• Bear Case:
  • 80% of current marketing tasks may get automated; everyone will need to become more agentic and reinvent their roles.
    • “80% of what we're doing today is going away and going to be automated...I would much rather get ahead of it and reinvent myself now than wait for it to be forced on me.” – Drew Pinta [44:53, 46:29]
  • Marketers must embrace this shift—build, not just operate.

8. Memorable Moments & Notable Quotes

• On Creative Stunts:
  • “Not everyone can like put a guy doing expenses in a park in a glass box for a day... We did measure that.” – Drew Pinta [07:59]
• On Measurement Culture:
  • “Measurement should meet the marketing, not vice versa.” – Drew Pinta [15:43]
• On Testing & Failure:
  • “You're going to have a lot of Ls [losses]. Most of them probably aren't going to work... but you got to think about them as actually valuable lessons.” – Drew Pinta [20:47]
• On Role Reinvention:
  • “I don't do 90% of what I was doing nine months ago.” – Drew Pinta [45:18]

Timestamps for Key Segments

• Ramp’s Data-Driven Creative Approach: [01:22]
• Drew’s Background & Marketing-Data Partnership: [04:59] – [07:20]
• Measuring Creative Stunts & Event Studies: [07:59] – [13:43]
• Podcast Attribution & Long-Tail Influence: [13:43] – [16:24]
• Killer Quote on Creative Measurement: [15:43]
• Incrementality Testing Fundamentals: [17:01] – [20:11]
• Channel Maturity & Budget Allocation: [28:18] – [31:25]
• Bullish & Bearish Future for Marketers in AI Era: [39:50] – [46:29]
• Agentic Workflows & AI Enablement at Ramp: [46:47] – [55:19]
• Cultural Enablers for AI Adoption: [53:31] – [55:19]
• Risks & Guardrails for AI Automation: [56:01]
• Notable Closing Chatter: [59:07] – [60:50]

Actionable Takeaways

• Test systematically: Every new channel or campaign should have a hypothesis, test design, and learning milestone—even if measurement isn’t perfect.
• Let measurement fit the marketing: Don’t force creative initiatives into direct-response KPIs; tailor measurement strategies.
• Balance core and new bets: Keep 70% of your budget in proven, incremental channels; dedicate 30% to experimentation for future growth.
• Embrace AI tools: Shift marketing teams (and yourself!) toward building and automating, not just operating the old playbooks.
• Culture matters: Leadership must empower teams to experiment and learn, even at the expense of some short-term results.

Final Thoughts

This episode is a masterclass in building a modern B2B marketing engine—one that fuses creative risk-taking with hardcore data rigor and experimental AI enablement. Marketers who want to stay relevant need to learn, experiment, and embrace change, just like Ramp’s team. The future is creative, agentic, and data-literate—don’t wait for the market to reinvent you.

Notable Quotes Cheat Sheet

• "The fastest way to kill creative marketing ideas is to try to measure them like direct response. Measurement should meet the marketing, not vice versa." — Drew Pinta [15:43]
• "Marketers are shifting from running campaigns to building. They are becoming builders, enabled by cloud code..." — Drew Pinta [46:59]
• “80% of what we're doing today is going away and going to be automated... reinvent yourself now.” — Drew Pinta [44:53, 46:29]
• "In marketing the median is death... you need to be unique and like the alpha is being creative and having taste..." — Drew Pinta [40:03]