The Novasign company record

This page is the company record for an investor, analyst or corporate-development reader doing first-pass diligence, one place that answers what Novasign is, what is proven, how it is funded and what it is looking for.Every figure on it is argued in full elsewhere on this site, and the evidence behind each one is published in the results overview.

Novasign builds one product, Novasign Studio (a browser-based workbench in which a process engineer builds a model of a unit operation from designed experiments and then simulates the runs they were about to book), and licenses it alongside the modelling and implementation work that gets a first model running. The method is hybrid modelling, and what it is and how it works is argued on the hybrid modelling page. The commercial consequence is the subject of this one.

It states no valuation and no terms, and it is not an offer, a solicitation or an invitation to invest.

Two established approaches each fail in a way the other cannot fix. Classical design of experiments is statistically sound but treats the vessel as a black box, so it needs a run for every question and cannot extrapolate past the conditions it sampled. Purely data-driven machine learning needs far more runs than a development campaign produces, and its output is a model a regulator cannot inspect, which is a hard limit in an industry that has to justify a process, not merely operate it.

Hybrid modelling gives the model the part that is genuinely known (mass balances, stoichiometry, the streams entering and leaving the vessel) as structure rather than asking it to be learned. The network is left with a low-dimensional rate function. That is why a usable model can come out of a handful of well-designed runs, and why it stays interpretable.

Novasign’s differentiation inside that category is documented rather than asserted: peer-reviewed science behind the method, a partnership with an instrument vendor that puts the modelling next to the hardware generating the data, and four deployment options including on-premise and air-gapped, which is what a regulated manufacturer’s IT function actually requires.

The most-quoted of the five: 8 runs instead of 22, 64% less experimental effort.

Across programs the reduction runs 35–70% fewer experiments, and the band is the claim rather than its maximum. It depends on the analytical precision available in the customer’s own campaign. The results overview marks which figures are measured and which are estimates.

We are seeking Series A growth capital to accelerate the commercial expansion of Novasign Studio. With our technology validated in client projects, we are ready to scale — strengthening our visibility, expanding our customer reach and driving broader adoption.

Alongside our current investor, we aim to bring together venture capital firms, strategic investors with strong commercial networks, and investors from other industries who can help expand Studio into new sectors. We are looking for partners who share our ambition and can help turn proven technology into sustained growth.

That is a deliberate line rather than an oversight. Figures of that kind change every quarter, and a page carrying a stale one is worse than a page carrying none. It is the same reasoning that keeps certification target dates off the trust page. Financial and commercial detail belongs in a conversation with a mutual non-disclosure agreement in place.

This page is a record and an introduction. It is not an offer, a solicitation, or an invitation to invest.

Novasign’s practice is to reply with a call rather than a deck, because the first useful conversation is usually about the science and the deployment model rather than the slides. If what you actually want is to see the product work, a demo reaches a modelling specialist, which is a better use of the time than a corporate overview.