The $250M Question Pigment's Headline Never Asked: Why 'Killing Excel' Is the Wrong Battle

The $250M Question Pigment's Headline Never Asked: Why 'Killing Excel' Is the Wrong Battle

Last week, a funding announcement crossed my desk that I have now read eleven times. Pigment, a European business planning platform, raised $250 million to — and I am quoting the framing directly — "challenge Microsoft's Excel and PowerPoint." That is the entire thesis. Two hundred and fifty million dollars, one sentence, and a product comparison that has been the death rattle of enterprise software companies for two decades running.

What stopped me was not the amount. It was the silence. No round disclosed. No investors named. No valuation. No ARR. No customer count. No product screenshots. No quotes from a founder. I have audited token vesting schedules that leaked more operational detail than this press release. So before I analyze what Pigment is, let me be honest about what this announcement actually contains: almost nothing. And the nothing is the story.

Listening to the errors that the metrics ignore

I spent the better part of 2021 auditing over fifty failing NFT marketplace contracts during the floor crash — most of them had died not from market conditions but from gas inefficiency in their batch minting logic. That job taught me a discipline I have carried into every subsequent analysis: when a system fails quietly, the failure pattern is more informative than the marketing. Pigment's announcement is a system failing quietly at the level of disclosure. A $250M round is not a small event. Companies do not raise that capital in silence unless they have a reason to stay silent.

The reasons are usually one of three. Either the round is structured with a valuation that would invite unfavorable comparison to peers (Anaplan's take-private, Adaptive Insights' $1.5B acquisition by Workday, Pigment's own prior-round marks), or the investors are strategic rather than financial and don't want the signal public, or the ARR numbers are still modest relative to the capital raised and the company wants the headline to do the narrative work instead of the financials.

My money is on the third. Not because Pigment is failing — the company is clearly real, clearly hiring, clearly shipping — but because the "Excel killer" narrative is precisely the narrative you deploy when you cannot compete on the metrics that enterprise buyers actually evaluate. When the numbers are quiet, the story gets loud.

I have seen this pattern before. In 2018 I sat through a pitch from a well-funded EPM startup whose entire deck could be summarized as "we are Anaplan but with a better UI." That company is now a footnote. The pattern is not unique to Pigment, but the $250M price tag makes it worth examining carefully.

What Pigment actually is, mechanically

Stripping out the marketing, Pigment is a multidimensional business planning platform — an enterprise planning and performance management tool adjacent to the EPM category that Anaplan, Oracle, and SAP have occupied for years. Its interface is a grid, its mental model is spreadsheet-like, and its differentiator is that the underlying data model is not two-dimensional. Each cell exists in a lattice of dimensions, hierarchies, and dependencies. Change one assumption, and the affected cells recompute across the dependency graph.

This is where the engineering — and the real risk — lives. Every spreadsheet-replacement product ever built has died on the same hill: the incremental recomputation problem. When a user changes a single driver assumption in a model with millions of cells and dozens of dependency layers, the system must identify and recompute only the affected nodes. Do it wrong and you get the classic "Excel-killer" failure mode: the product is elegant until you ask it to do something a real financial analyst would actually need, at which point it either hangs, produces stale values, or silently corrupts the model.

I have watched three separate enterprise SaaS companies ship products into this exact trap. The engineering team optimizes the demo path — a small model, a shallow dependency graph, a slick animation — and then watches in horror as the first real customer imports a model with a thirty-layer driver chain and the system collapses. Pigment's survival depends entirely on whether they solved this before scaling. The $250M round strongly suggests they believe they have. The lack of any technical detail in the announcement suggests they would rather we not ask.

Beyond the recomputation engine, the architecture that matters is the connector ecosystem. A planning tool that can only ingest CSV files is a toy. A planning tool with bidirectional sync to Salesforce, NetSuite, Snowflake, and Workday is infrastructure. The difference is the difference between a $50M valuation and a $2B valuation, and it is almost never visible from the outside until a customer tries to build a real integration and discovers that the "native connector" is actually a scheduled CSV import running every fifteen minutes.

The quiet confidence of verified, not just claimed

Here is the technical truth that the press release buries: the reason Excel-killer products have failed for twenty years is not that they were poorly engineered. The reason is that Excel's greatest feature is not its grid. It is its zero learning curve.

An FP&A analyst does not learn Excel. They grow up inside it. They inherit models from a predecessor, they add a tab, they hard-code a lookup, and the model works — not well, not cleanly, but it works. The cost of switching is not the price of a license. The cost is the entire mental model of how work gets done, and any product that requires users to first learn a new formula language before they can do anything is asking for a down payment on a bet the user has not yet decided to make.

Anaplan solved this problem by ignoring it. Anaplan said: this is not Excel, this is a new discipline, and we will sell it to the planning team as a system of record, not as a spreadsheet. That worked for a decade because the alternative was genuinely worse. Then Microsoft shipped Copilot.

The reason "killing Excel" suddenly became a plausible thesis in 2024 and 2025 is not that spreadsheet alternatives got better. It is that natural language modeling collapsed the learning curve. When a user can say "build me a Q3 marketing budget broken down by channel, with a 20% CAC increase assumption," and the system generates the model structure, the barrier that killed every previous generation disappears. The user does not need to learn the product's language. The product speaks theirs.

This is not speculation. In 2025 I designed a verification protocol for AI agents transacting on-chain, and the single most important thing I learned was that trust in a system scales with how little the user needs to understand about the system's internals. The same principle applies here. A planning tool that requires a certified expert to configure it is Anaplan. A planning tool that requires a sentence is something new.

So the question I would ask Pigment's board, if I had the microphone, is not how many customers they have. It is whether they have shipped a natural language modeling layer, and if so, whether it actually works on models with real complexity. The press release does not say. The press release cannot say, because the press release is written for the headline, not for the buyer.

Protecting the ledger from the volatility of hype

Here is where the narrative starts to crack, and it is not because of anything Pigment has done wrong. It is because the positioning is structurally fragile.

"Challenge Microsoft's Excel and PowerPoint" is a beautiful piece of marketing because it borrows Microsoft's cognitive real estate at zero cost. Everyone knows Excel. Everyone has felt the pain of a broken spreadsheet. The implication is that Pigment is the same weight class, the same category, the same level of necessity. That is a very efficient narrative.

It is also a trap. By framing itself as an Excel alternative, Pigment accepts Excel as the reference point. It accepts that the buyer's evaluation criterion will be "how much better is this than Excel," which is a bar that Excel sets and Pigment has to clear, rather than a bar that Pigment sets and Excel has to reach. Every quarter that Microsoft ships a Copilot improvement, the bar moves in the wrong direction. Every quarter that Pigment fails to ship a feature that Excel already has, the comparison deepens.

The better positioning — the one I would recommend if anyone asked me, which nobody does — is not to be the Excel alternative. It is to be the layer that Excel cannot be. Excel is a personal productivity tool. Pigment, at its best, is an organizational planning system. These are not the same product category. Excel scales to a person. A planning system scales to a company. The narrative should be that Pigment does not replace Excel, it replaces the practices that Excel was never designed for: version chaos, assumption drift, the twelve-tab financial model maintained by one analyst who is one resignation away from taking the entire company's planning capability with them.

But that narrative is harder to sell. It requires the buyer to accept that they have a process problem, not a tool problem. It requires an enterprise sales motion, not a viral campaign. It requires patience. And patience is not what a $250M round buys you. A $250M round buys you growth, and growth requires a story that spreads. "Excel killer" spreads. "Organizational planning infrastructure" does not.

I have watched this exact dynamic play out in the L2 space. In 2023 I spent two weeks reverse-engineering the consensus mechanisms of three major Layer 2 sequencers, quantifying the actual percentage of centralized control nodes. The finding that got cited by institutional analysts was not the one about latency or throughput. It was the one about a 15% single-point-of-failure risk. The story that spreads is the story about a specific failure, not the story about a structural improvement. Pigment's marketing team knows this. That is why the announcement is what it is.

When the floor drops, the foundation speaks

The strategic question nobody at Pigment wants asked out loud is: what happens when Microsoft decides to compete seriously?

Microsoft does not need to kill Pigment. Microsoft needs to make Excel good enough that the marginal enterprise buyer no longer has a compelling reason to switch. Copilot for Excel already does natural language formula generation. Copilot for PowerPoint already generates decks from data. The infrastructure is in place. The only question is how aggressively Microsoft invests in the planning-specific capabilities that would make Pigment's differentiation moot.

Microsoft's incentive to do this is not obvious. Enterprise planning is a small market relative to Microsoft's core business. But Microsoft also does not like being the reference point in someone else's marketing. And Microsoft has a long history of shipping "good enough" features that eliminate adjacent categories: Teams killed Slack's enterprise momentum, Power BI killed a generation of BI startups, Power Automate killed a generation of workflow tools. The pattern is consistent.

If I were Pigment's CTO, I would be spending the next twelve months not on feature parity with Excel but on the things Microsoft structurally cannot build: deep vertical integrations (regulatory capital planning for banks, clinical trial modeling for pharma), certified implementation partnerships with the Big Four, and a data model flexible enough that Microsoft's one-size-fits-all approach cannot match it. Those are the moats that survive a platform shift. Everything else is a feature Microsoft will ship in a minor release.

The audit trail as a narrative of trust

The other thing Pigment's announcement does not address — and this is where my security background makes me bristle — is compliance posture. For a European company selling planning software into financial institutions, the compliance story is not a legal checkbox. It is a competitive weapon.

GDPR alignment is table stakes for European vendors. SOC 2 Type II is table stakes for selling into US enterprises. But the real differentiation in 2025 and 2026 is going to come from the EU AI Act. If Pigment's natural language modeling layer is built on top of a large language model, and if that model is trained or fine-tuned on customer data, the Act's transparency obligations kick in. If the system is used to inform financial decisions — and a planning tool absolutely is — the Act's risk classification for high-risk AI systems may apply. That triggers requirements for human oversight, technical documentation, and conformity assessment.

I have not seen Pigment's compliance roadmap. I have not seen any indication that they have thought about this at all. But I have sat in the room with three different European SaaS companies over the past eighteen months, and the ones that are getting ahead of the AI Act are the ones that will win the enterprise deals in 2026. The ones treating it as a legal afterthought will discover that their largest prospects have compliance teams with veto power.

The deeper issue is that a planning platform handles the most sensitive data any company possesses: forward-looking financial projections, strategic assumptions, headcount plans, acquisition scenarios. This is not data that enterprises want sitting on shared infrastructure with unclear data residency. Pigment's European heritage is a genuine asset here — "your financial data stays in Europe" is a real differentiator against Anaplan and Adaptive Insights. But it is only an asset if Pigment operationalizes it into contractual guarantees, data residency options, and audit capabilities that satisfy a bank's risk committee. Nothing in the announcement suggests they have.

Rooted in the past, secure for the future

I want to be fair to Pigment. The company is real, the problem it addresses is real, and the $250M is not a rounding error. There is a plausible world in which Pigment becomes the European answer to Anaplan, wins a meaningful share of the mid-market planning space, and exits via acquisition or IPO at a healthy multiple. That world does not require killing Excel. It requires being genuinely better than Excel for a specific set of use cases that Excel handles badly, and being genuinely easier than Anaplan for a specific set of customers who find Anaplan too heavy.

That is a good business. It is not a $250M-round business at a valuation that implies a category-defining outcome. But it is a good business.

The failure mode is the one I have seen repeatedly: the company raises at a valuation that assumes it will win the category, then spends the capital on sales and marketing to justify the valuation, then discovers that growth is linear in sales headcount rather than exponential, then faces a down round or a fire sale. I have watched this happen to at least four enterprise SaaS companies. The pattern is not subtle.

The way to avoid it is unglamorous. Ship the natural language modeling layer and make it actually work on real models. Build the certified implementation partner network before the Big Four decide to standardize on Anaplan. Get the AI Act compliance roadmap ahead of the regulatory deadline. Publish the NRR number, even if it is not flattering, because the enterprise buyers who matter will find out anyway. Stop framing the product as an Excel alternative and start framing it as the layer that lets a company's planning survive a single analyst resigning.

None of that is a press release. All of it is the actual work.

I will be watching Pigment's next twelve months closely, because the shape of this company's trajectory will tell us something important about where enterprise software is going. If a $250M round buys a real category shift, we will see it in the mechanics — in the depth of the connector ecosystem, in the sophistication of the AI modeling layer, in the compliance posture that satisfies a bank's risk committee. If it buys marketing instead, we will see that too, and the tell will be a next-round announcement that features a different headline and a smaller number.

The quiet confidence of verified, not just claimed applies here the same way it applies to every protocol I have audited. The code either recomputes the dependency graph correctly under load, or it does not. The API either syncs bidirectionally with the enterprise data warehouse, or it does not. The compliance team either signs off, or it does not. Everything else is the story we tell while we are waiting to find out.

Pigment's $250M is a bet that the story will turn out to be true. I have read the announcement eleven times now, and I still do not know the answer. That is not a criticism of the company. It is a criticism of the announcement — and it is the most important thing the announcement never said.