Rillet Closed $100 Million in Under 48 Hours. Its Customers Are Ripping Out NetSuite.

News Breakdown · FiscEdge Academy

Rillet, a New York startup that sells accounting software, made it official on August 19: a $100 million Series C at a $1 billion valuation, led by ICONIQ. Sequoia and Andreessen Horowitz, both already on the cap table, came back in, joined by Bain Capital Ventures, Oak HC/FT, Battery Ventures, FirstMark, Scale Venture Partners and Creandum.

It is the company's third round in roughly 14 months. A $25 million Series A led by Sequoia was followed twelve weeks later by a $70 million Series B, and now this. Total funding is past $200 million, against a company that only left stealth two years ago. Co-founder and CEO Nicolas Kopp said on X that the round came together in less than 48 hours.

The valuation is the least interesting number here. The one that should make every SaaS founder sit up: Rillet says its new ARR doubled in the last three months, and it now has more than 600 customers, including publicly traded companies. These are not greenfield startups picking their first tool. They are enterprises tearing out Oracle Fusion, SAP, Workday, Microsoft Great Plains and NetSuite to run their books on a four-year-old startup instead.

The signal under the headline: the system of record is now contestable

For twenty years, the received wisdom in enterprise software was that the general ledger was untouchable. ERP was the last system anyone ripped out, because the switching cost was measured in eighteen-month implementations, six-figure consulting invoices, and the career risk of the CFO who signed off. NetSuite's moat was never the product. It was the pain of leaving.

AI collapses that moat in a mechanical way. The expensive part of an ERP migration was never the database; it was mapping one company's messy chart of accounts, revenue schedules and reconciliation rules onto a rigid system that expected clean inputs. That mapping is exactly what a model is good at. When migration cost drops from twelve months to weeks, the incumbent's defensibility drops with it, and a category that hadn't seen a serious new entrant in a decade becomes contestable overnight.

Rillet frames its roadmap as "accounting superintelligence": agents doing progressively harder finance work inside a live general ledger, not a chatbot bolted onto a static system. That framing matters less than the buying behavior it produces. Finance teams are the most conservative buyers in the enterprise. If they are switching, the thesis has already been proven somewhere less risky.

Why a round closed in 48 hours

Kopp's "less than 48 hours" is not a flex about investor enthusiasm. It is a data point about what a growth round is priced on in 2026.

Three rounds in 14 months means Rillet never ran a process. Existing investors with live data on the account had enough conviction to re-up, and a lead came in on top. When new ARR doubles in a quarter, diligence stops being discovery and becomes confirmation. The metrics do the pitching.

The uncomfortable corollary: if your numbers do not do that, no deck will substitute. The bar for a fast round is a growth curve an investor can verify in an afternoon, and the unit economics that hold up underneath it. Rillet raised easily because a repeat customer motion in the most switching-averse segment in software is about as clean a proof as enterprise SaaS produces.

What this changes for how you build

Three practical takeaways.

Look for categories where the moat is friction, not product. Anywhere retention rests on migration pain rather than better software, an AI-native entrant has a cheap wedge. Payroll, billing, revenue recognition, procurement, claims. Ask what the switching cost is made of. If the answer is "someone has to map a lot of unstructured data," that moat is rented, not owned.

Sell the removal, not the feature. Rillet's customers are not buying an AI feature. They are buying the deletion of a month-end close, a purchase order signed against time returned to a finance team. Founders who lead with model capability lose to founders who lead with the process that disappears. It is the discipline we push in our startup strategy track: the deliverable is the outcome, not the technology.

Assume your own category is being looked at the same way. If you run a SaaS company with a sticky-because-annoying-to-leave product, someone is modeling how cheap your customers' exit has become. The defense is not lock-in. It is being the one who rebuilds your product AI-native before a funded stranger does it for you, which is the premise of building software with AI from the ground up.

One caution on the number: a $1 billion valuation on a company this young prices in years of execution. Doubling new ARR off a small base is not the same as doubling it off a large one, and the ERP replacement cycle has a long tail of accounts that will not move. The round is a bet on direction, not a verdict.

If you remember one thing

The incumbents most exposed to AI are not the ones with weak products. They are the ones whose customers only stayed because leaving was expensive. When the cost of leaving collapses, twenty years of accumulated advantage becomes a list of accounts in someone else's pipeline.

---

We break down capital markets, valuations and go-to-market strategy in FiscEdge's financial modeling course and startup strategy course. Browse the full blog. Follow @fiscedge for daily Business & AI analysis.