Our innovation journey and how we redesign data rooms for today's M&A professionals.
Every strategic transaction begins with a mountain of documents. Collecting, classifying, renaming, redacting, and indexing them usually takes months. Virtual Data Rooms (VDRs) were supposed to make this easier. Instead, they often feel like digital relics: slow, clunky, and priced as if servers still charged rent by the page. Entropia is redesigning this experience.
Our R&D journey: AI document processing
Since founding Entropia, we have been exploring how AI can work in conjunction with private data. Our focus quickly narrowed to unstructured data: millions of PDFs, PowerPoints, Scans, Word or Excel files, even Autocad drawings that companies struggle to exploit. From early 2024, we developed expertise in four areas: automated document understanding, semantic search across vast corpora, information extraction with generative AI, and chat-based agentic applications.
These capabilities were first deployed in projects with clients in legal, finance, consulting, real estate, and even health care. They took the form of a search engine, a document analysis grid, and later an agent-based platform able to answer questions directly from large internal libraries. Much of this work centred on due-diligence material with strict confidentiality standards. It became clear that what began as tools for research and knowledge management had an obvious home in the due diligence and transactions processes, and that a natural stage for such technology was the data room.
We are using this technology to reinvent the dataroom
Ask any Associate preparing a deal. Data rooms remain indispensable but exasperating: interfaces are often slow, pricing opaque, and critical tasks are often managed outside the platform, in Excel and email threads.
Conversations with data room users revealed a catalogue of recurring pain points:
- Time-consuming preparation: Collecting, verifying, renaming, classifying and redacting thousands of documents is still largely manual and painfully slow. On the sell-side, analysts and associates too often struggle for the basics. Transactions are highly time sensitive, and every week of delay risks stalling the deal.
- Outdated interfaces: Platforms remain too often clunky, slow or buggy, forcing reliance on support teams, and generating stress.
- Lack of smart functionalities truly leveraging AI: Users expect smarter features to cut the drudgery of data room preparation. Yet despite grand promises, most providers deliver little beyond incremental tweaks.
- Stressful access management: Granular permissions are complex to set up and easy to mishandle. Some users have to call support teams to make sure they are properly setup.
- Processes outside the platform: Critical tasks often spill into Excel sheets, email chains, leading to errors, duplication and inefficiency.
- Opaque pricing: Bills are unpredictable and frequently exceed initial estimates, producing invoice shock for clients. Legacy page-based billing models punish users with escalating costs. That's also the experience Arnaud had a few years ago when his former company Arcane was acquired.
- Data sovereignty: For European clients, incumbents have yet to provide convincing answers on data sovereignty. Most rely on US-based cloud providers, potentially leaving sensitive files exposed to extraterritorial claims. Amid sharpened geopolitical tensions, a growing number of european customers now declare they would prefer more sovereign alternatives.
Meanwhile, a new generation of bankers, lawyers and real-estate analysts now expect workplace tools to match the sleekness of the apps they use in daily life, and have little patience for sluggish enterprise software.
With deep experience in document processing, we recognised that their frustrations was an opportunity for us, and resolved to reinvent the data room as if it were designed today: fast, modern, AI-native and transparently priced.
Our product priorities
Blazing-fast experience
The first complaint from many users is the sheer slowness of existing data rooms. What should feel like a simple web app too often behaves like a relic from the early 2000s. This is less about hardware limits than accumulated technical debt: years of bolted-on features have left incumbents with sluggish systems.
By contrast, Entropia is rebuilding from scratch on a lean modern stack and leveraging best practices that Pierre, our CTO, once deployed at Google to handle services at planetary scale. The goal is straightforward: a data room that feels as fast and smooth as the consumer apps people now take for granted.
Focus on Sell-Side pain-points
We are directing our initial design at the sell-side, with the aim of stripping away the grind of document review, facilitating low‑value tasks that consume associates’ time. At the same time, we will avoid forcing users out of familiar workflows: existing Excel processes, such as Q&A tracking, will be integrated and enhanced rather than replaced.
Here are the first smart features that we are going to put in production:
- Automatic renaming: Stop wasting nights re‑labelling files. Our data room recognises content and applies consistent names across thousands of documents, learning from your naming habits.
- Duplicates and version detection: No more confusion over which draft is final. We spot duplicates and versions instantly, reducing errors.
- Index allocation: Instead of dragging files endlessly into the right folder, our data room reads the content and suggests the proper slot on the index.
- Redaction assistance: Sensitive details (names, addresses, financial figures) can be flagged and removed faster, while remaining under your control.
- Semantic search: No more keyword guessing. Ask questions in plain language and retrieve the right files, even if they are buried deep, or written in another language.
- Smarter Q&A management: Keep Excel in the loop but make it intelligent. The platform surfaces similar buyer questions, reduces duplication and helps manage responses seamlessly.
Once this is fully live and adopted, a second wave of features will turn to the buy-side. The challenge there is will be digestion rather than organisation. Associates, lawyers and consultants spend weeks combing through contracts and financials to build diligence matrices, flag risky clauses and extract KPIs. Our aim is to make this work faster and more insightful, with AI tools that surface unusual clauses, pull key metrics and generate structured grids.
Beyond security badges: a zero-trust architecture
Data security is one of the first reasons why you want to purchase a dataroom. From day one, we have handled financial, legal and even health records, which left no room for compromise. Pierre, our CTO, brought with him Google’s habit of treating security as a first principle and embedded a zero-trust design into our systems early on. Three pillars underpin that approach today:
-
SOC 2 Type II (audited): As this article is being written, we are undergoing audit for SOC 2 Type II, which evaluates how effectively we operate controls over time. It covers everything from access management to incident response, ensuring not just policies on paper but proof of execution.
💡 Read more: Entropia achieves SOC2 Type II compliance → -
ISO 27001 (via our cloud provider): The infrastructure that hosts both our dataroom and our AI models is certified under ISO 27001 (among other standards), meaning it follows globally recognised best practice for information security.
-
Zero Trust (by design): We go beyond compliance. Every request is verified, every system is assumed hostile until proven otherwise. This architecture builds systemic resilience rather than relying solely on checklists.
💡 Read more: Our zero-trust security model →
Beyond GDPR compliance: sovereignty for European customers
Data sovereignty is a growing concern as geopolitical uncertainty mounts. On the legal side: extra-territorial laws such as America’s CLOUD Act or FISA 702 give U.S. authorities the power to compel access to data held by American cloud providers, regardless of where the servers are located. That makes it impossible for them to guarantee European clients full protection, even when they claim that “data is hosted on european servers”.
By contrast, hosting exclusively in Europe and with a European provider allows us to offer something rare in the dataroom market: genuine insulation from foreign surveillance laws.
Transparent and flexible pricing
Last but not least comes pricing, where most data rooms remain oddly stuck in the 1990s. Back when rooms were physical and files sat in metal lockers, charging per page made sense: more pages meant more shelves, more space, more rent. Yet the industry has clung to this proxy long after the files went digital. The catch is obvious: no one knows in advance how many pages a deal will generate.
That opacity has proved costly. Many sellers are lured by low initial estimates, only to suffer invoice shock at closing. Arnaud, when selling his previous firm, ended up paying three times the sum initially quoted by the Datasite representative. To make matters worse, critical functions for M&A professionals are often sold as pricey add-ons, slowing the process if they are not purchased.
We see little value in perpetuating that model. While we can match market practice if clients insist, our preference is to offer a fixed monthly price tied to gigabytes, with unlimited access to all features. No surprises, no invoice shock.
—
For transaction professionals weary of outdated tools, Entropia offers a chance to see what a data room looks like when reimagined from scratch. Built on modern infrastructure, equipped with AI where it matters, shielded by European data sovereignty, and priced with clarity, it is designed for the realities of today’s transactions rather than the habits of yesterday’s vendors. If you spend your nights chasing versions, managing Q&A in spreadsheets or bracing for invoice shock, we invite you to give Entropia a try and see how much smoother a deal can feel when the data room finally works for you.
Pierre-Louis
