Skip to main content

Chapter 5 — Product Strategy

Vinova Lab's product strategy is built on a single organising principle: every product we build should make the engine smarter, and a smarter engine should make every product better.

Products are not independent initiatives. They are both the expression of the engine's capabilities and the source of its growth.

Domain Selection​

Not every market is right for Vinova Lab.

The company focuses on domains that are document-intensive by nature — where a significant portion of operational work involves reading, classifying, extracting information from, or acting on documents.

A domain is a strong candidate when:

  • people spend meaningful time processing, searching, or managing documents as part of their core work;
  • documents in that domain have relationships to each other — a contract references a party, a bill references a supplier, a compliance document references a regulation;
  • errors or missed obligations in document management carry real operational, financial, or legal consequences;
  • the domain has recurring, predictable document types that can be learned and generalised;
  • existing solutions are either too generic (spreadsheets, email, shared drives) or too expensive and rigid (large enterprise software);
  • the audience is large enough to support a product with sustainable recurring revenue.

Product Discovery​

Products may originate from three main sources.

Market Observation — identifying document-intensive domains where operational complexity is high and existing solutions are insufficient.

Customer Engagements — discovering patterns and recurring document challenges through consulting projects. Every engagement should be evaluated not only for its immediate value but for what it teaches the engine.

Internal Innovation — developing ideas through research, experimentation, and founder insight. As the engine matures, new vertical opportunities may emerge from capabilities the engine has already developed rather than from external observation alone.

The Bottom-Up Engine Model​

Vinova Lab does not design the engine in the abstract and then build products on top of it.

The engine is built bottom-up — from real problems, in real domains, with real documents.

Each vertical product is a focused application of the engine to a specific domain. As the engine processes more documents in that domain, it develops deeper classification accuracy, richer relationship models, and more reliable extraction. Those improvements are then abstracted back into the engine platform and become available to every other product.

Product development and engine development are not separate activities. They are the same activity, approached from different angles.

Research Strategy​

Vinova Lab's research strategy evolves with the company's resources and maturity.

In the founding phase, research is founder-led. It involves evaluating the landscape of available models, tools, and frameworks — identifying what can be adopted, adapted, or combined to build a first working vertical without requiring significant capital.

As products generate revenue and training data, the focus shifts to fine-tuning open-source models with domain-specific knowledge derived from real operational use. Document types, extraction patterns, and classification logic become increasingly specialised to the domains Vinova Lab operates in.

Over time, as proprietary training data accumulates, Vinova Lab may develop or retrain models purpose-built for specific document intelligence tasks — achieving levels of domain precision that general-purpose models cannot match.

At every phase, products are designed from day zero to generate structural training signal. Personal data and sensitive information are never retained. The signal that feeds future model improvement is always anonymised, structural, and derived — not a copy of the original document.

Product Qualification​

An initiative should become a product only when it:

  • operates in a document-intensive domain with recurring, predictable document types;
  • solves a meaningful operational problem for a well-defined audience;
  • can be offered to multiple customers;
  • can scale operationally and technically;
  • has a plausible commercial model;
  • contributes document types and patterns that strengthen the engine;
  • generates structural training signal through normal use;
  • aligns with Vinova Lab's mission;
  • can be supported sustainably.

Not every reusable component is a product.

Not every successful customer project should become a product.

Product decisions must be based on evidence, domain fit, and commercial potential — not on technical possibility alone.

Current Portfolio​

rAInty is the first Vinova Lab product. It addresses the property management domain — a document-intensive vertical where landlords and property owners manage contracts, bills, correspondence, and compliance obligations across their portfolio.

Tender intelligence is the second vertical in development. It addresses the evaluation of public and private procurement bids — a process where organisations must analyse large volumes of technical documents, requirements, and compliance specifications to decide whether and how to respond. The document complexity, the recurring nature of the process, and the high cost of errors make it a strong fit for the engine.

Each product is built on the shared engine while retaining independent branding, positioning, pricing, customer experience, market strategy, and product roadmap.

Portfolio Growth​

The portfolio should grow steadily — led by evidence of domain fit and commercial viability, not by ambition alone.

New verticals will be added when a domain meets the selection criteria, when the engine has sufficient capability to address it, and when the company has the resources to support it sustainably.

The Vinova Lab Blueprint — Version 0.1 — Confidential Working Draft