Sep 11, 2025 · 1h 5m · mad
Goodbye Excel? AI Agents for Self-Driving Finance – Pigment CEO
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of The MAD Podcast, host Matt Turck interviews Pigment Co-CEO Eleonore Crespo about building AI-powered Enterprise Performance Management platforms and autonomous finance agents. Crespo details Pigment's multi-agent architecture, her founder journey from quantum physics to tech leadership, and strategies for driving enterprise AI adoption.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Matt holds 21.6% of the talking time here. How this is scored →
Matt as informed peer 3.6 Guest teaching 5.3 Guest disagreement 0.6 Matt pushing back 1.3
speaking balance: gold is Matt, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 36:30 Eleonore rejects the premise that AI will kill Excel
When Matt suggests AI might finally fulfill the long-standing SaaS promise of killing Excel, Eleonore firmly rejects the premise, predicting Excel will easily survive another decade.
Hardest push from Matt ▶ 10:13 Matt challenges Eleonore's PhD timeline compression theoryMatt refuses Eleonore's initial framing that AI will shorten PhD timelines, arguing that expectations will simply rise or autonomous AI discoverers will render PhDs obsolete.
Biggest teaching moment ▶ 1:01:31 Eleonore explains GSI quota realities vs boutique partnersEleonore delivers a masterclass on GSI economics, contrasting boutique partner speed with the immense $25M individual partner quota expectations at firms like Deloitte.
Matt holds his own ▶ 44:32 Matt probes token economics and gross margin impactMatt demonstrates high technical fluency by asking whether AI agent orchestration negatively impacts gross margins through high token intensity.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Matt as informed peer | Guest teaching | Guest disagreement | Matt pushing back | Why |
|---|---|---|---|---|---|---|
| Understanding Pigment and Enterprise Performance Management | 3 | 4 | 0 | 0 | Matt establishes the context as an early investor, citing Pigment's funding totals ($400M) and notable customers like Figma. Eleonore explains the core value proposition of Pigment using the GPS vs compass analogy. | |
| Eleonore Crespo's Journey: From Quantum Physics to Venture Capital | 2 | 5 | 0 | 0 | Matt asks standard background questions about Eleonore's transition from physics and Index Ventures to founding Pigment. Eleonore details how observing Google CFOs and Index founders shaped her understanding of business models. | |
| AI Acceleration in Physics, PhDs, and Scientific Discovery | 4 | 5 | 2 | 3 | Matt pushes back on Eleonore's assertion that AI will compress PhD timelines, suggesting the bar will simply rise or autonomous AI will replace PhDs entirely. Eleonore reframes the idea, emphasizing the ongoing necessity of human supervision in exploring the unknown. | |
| Founder Traits, Personal Culture, and Hiring for AI Fluency | 2 | 4 | 0 | 0 | Matt asks about personal traits and whether AI fluency is a mandatory interview criteria. Eleonore outlines Pigment's focus on hiring self-coaching, competitive candidates who display natural AI curiosity. | |
| Internal AI Operations, Security Guardrails, and Productivity Tools | 3 | 4 | 0 | 2 | Matt presses slightly on why Pigment chose to build proprietary internal growth tools rather than purchasing existing vendor solutions. Eleonore explains how security guardrails and specific process requirements dictated an internal build. | |
| Pigment's AI Agent Architecture: Analyst, Modeler, and Planner | 3 | 6 | 0 | 1 | Matt asks why Pigment launched three specialized agents (Analyst, Modeler, Planner) instead of a single unified agent. Eleonore provides a detailed architectural breakdown of why distinct task domains require separate agents supervised by an overarching layer. | |
| Multi-Agent Orchestration, Accuracy, and Human Supervision | 5 | 6 | 1 | 2 | Matt probes the mechanism for error prevention, asking if Pigment has eliminated hallucinations. Eleonore explains that because calculations execute deterministically on the Pigment engine rather than in the LLM layer, accuracy remains 100% auditable. | |
| Real-World AI Impact, Supercell Case Study, and the Future of Excel | 5 | 6 | 3 | 3 | Matt brings up the historical pattern of SaaS startups claiming to kill Excel and asks if AI will finally achieve it. Eleonore politely rejects the premise, predicting Excel will survive 10+ years due to enterprise adoption latency and its superiority as a data rendering UI. | |
| Self-Driving Finance and Autonomous Enterprise Planning | 4 | 6 | 0 | 1 | Matt extends the autonomous enterprise concept to real-time supply chain adjustments during geopolitical crises. Eleonore reveals ongoing work with a major global transportation customer on self-planning systems to eliminate human wishful thinking. | |
| Model Agnosticism, Partner Ecosystems, and Token Economics | 6 | 5 | 0 | 2 | Matt demonstrates strong technical familiarity with AI economics, asking directly about model provider partnerships and token intensity impact on gross margins. Eleonore explains that LLMs serve primarily as a translation layer, keeping token usage efficient. | |
| Customer Reaction and Enterprise Adoption of Pigment AI | 2 | 5 | 0 | 0 | Matt asks how customers react to Pigment's agent strategy. Eleonore contrasts existing customer enthusiasm—illustrating with a story about Supercell's CEO praising the software—against prospective customer hesitancy around workflow changes. | |
| Top-Down AI Push vs. Bottom-Up Employee Reality | 3 | 5 | 1 | 1 | Matt references a viral Twitter cartoon mocking top-down CEO mandates for AI. Eleonore agrees with the top-down pressure assessment and emphasizes the need for change management to address employee job fears. | |
| Reskilling, Career Evolution, and Vibe Planning | 4 | 6 | 2 | 3 | Matt draws an analogy to AI coding tools like Cursor and asks whether 'vibe planning' will emerge in finance. Eleonore rejects the premise, asserting that strict regulatory requirements and financial principles make deep subject matter expertise mandatory. | |
| Building a Global European Company and US Expansion | 5 | 5 | 1 | 2 | Matt highlights the VC narrative violation of scaling a US-dominated tech company from Europe without relocating the founder. Eleonore details how post-COVID remote selling and a US-based executive team enabled global expansion from Paris. | |
| Scaling Through Strategic Partnerships and GSIs | 5 | 7 | 0 | 1 | Matt brings up the difficulty of scaling software through System Integrators and GSIs. Eleonore educates on partner quota economics, explaining why startups must start with boutique firms before attempting to feed $25M GSI partner quotas. | |
| Long-Term Vision for Pigment | 2 | 5 | 0 | 0 | Matt asks about long-term success metrics over a 3-5 year horizon. Eleonore lays out a vision to build a $100B+ category definer that expands beyond EPM into broader enterprise software suites. |
Statements from this episode (21)
Crespo: Pigment becomes useful for companies at around 1,000 employees
“And really, I think pigment starts to be useful when you start to be, you know, approximately a thousand employees.”
Crespo: AI will compress 15-year scientific research timelines into two years
“I actually think that now, probably in two years, you can actually, actually achieve things that you would have done before in 15 years.”
Pigment rejects all job candidates who lack AI fluency and curiosity
“Indeed, for sure. I mean in, in the interview process, whether we are hiring for An executive, or whether we are hiring for a seller, or whether we are hiring for an engineer if they do not have the curiosity and haven't, do not have like a thoughtful idea aro…”
Pigment built custom AI lead-scoring and seller-routing tools internally
“We have built also our internal AI tools with the growth team that we have internally to help actually feed the right leads to the right seller at the same time, know exactly. When someone might be ready to buy, etc.”
Pigment launches three AI agents: Analyst, Modeler, and Planner
“So we have launched and we have announced our first three agents, analyst, modeler, and planner.”
Pigment plans to launch tens of AI agents including a consulting agent
“But before that, I think we're gonna launch tens of agents that are gonna do different things. We want to launch a consulting agent, for instance, that is gonna help during the implementation of pigment.”
Crespo: Enterprise AI finance systems will always require human supervision
“I do believe that in what we do, they will always be on top of that a supervising human.”
Coca-Cola uses Pigment's AI analyst agent for supply and demand analysis
“I could take the example of Coca-Cola today, Who is using today's analyst agent, which is the first that we launched to actually understand where there are problems between supply and demand.”
Crespo: Microsoft Excel will still be actively used in ten years
“I have a prediction that Excel will still be here in five years. And there is a good reason. You can keep that. In five years, Excel will still be there. And in 10 years, Excel will still be there.”
Pigment building autonomous planning system with global transportation giant
“Yes, so actually, we are working already with some customers on an autonomous planning system, and we are working right now very seriously with, I can say, the largest transportation company in the world, I think, as of today.”
Crespo: Pigment runs calculations natively rather than relying on LLM tokens
“It's probably not as token intensive as other businesses because we run our calculations ourselves, so it's more like you use the LLM to actually question the platform. But then the platform does the calculation itself.”
Pigment hires forward-deployed engineers for enterprise AI deployment
“At Pigment right now we are doing like what several AI companies are doing out there, which is hiring forward deployed engineers and people to really help with change management and deploying the models in the right way.”
Crespo: Developers cannot effectively use Cursor without understanding coding fundamentals
“The fundamentals of what coding is today, you are not able to use cursor. If you don't understand coding”
Crespo: Pigment reduces cohort analysis build times from months to minutes
“I think vibe planning is the sense that now you can build easily models on pigment. So you can do vibe planning on like, you know, like building super, super simple, like maybe, you know, like a cohort analysis that would have taken you months to build. Now yo…”
Building a US-facing startup from Europe was impossible before COVID-19
“I think that would not have been possible to be honest, before COVID.”
Crespo: The US accounts for 60% of Pigment's revenue
“U.S. Is our number one market. It's 60% of our revenue today.”
Crespo: Startups build partner velocity with boutique firms before GSIs
“What you see is, it's a little bit like when you build a business, is that you get velocity with the boutique firms first. It's exactly like you get velocity with your SMB customers before you get velocity with enterprise.”
Deloitte partners carry a $25 million annual quota, says Crespo
“For instance, I was with a Deloitte partner, and they were telling me, for them, their yearly so-called quota is twenty-five million. And for those who don't know, in enterprise software, the quota is more between one and two million if you're lucky, right?”
Enterprise startups must initially source all leads for Global System Integrators
“They will take time then to build the practice, they will take time to start feeding you with leads, so at the beginning, do not wait for them also to think they're going to source leads for you. At the beginning, you are going to be the only one to source lea…”