Ravical 融资 17M美元种子到规模会计专业知识与AI

供资

Ravical 融资 17M美元种子到规模会计专业知识与AI

在 Google 上的首选来源中添加 Unite.AI

会计师事务所往往通过增加人员来为更多的客户服务来成长。 Ravical拥有1700万融资的种子资金，以追求不同的模式：将公司最优秀人员的专门知识转化为共享的AI工作流程，可以在客户群中应用。

MMC Ventures领导了这一回合，现有投资者Lakestar和Pitchdrive与新的投资者Shapers一起参与。 融资带来 Ravical的总资金为2500万美元。

称公司在欧洲已有100多家会计师事务所使用其平台。 它将使资本投向英国、比利时、荷兰和德国的产品开发、团队扩张和增长，同时继续在欧洲展开并进入美国。

使资源具有专门知识

Ravical的论点是，在会计行业中，AI的获取会变得很普遍，但公司在如何应用知识方面仍然会有所不同。 个人助理可以改善雇员的产出，而不向同事或其他客户提供这种改进。

该平台旨在将公司既定的方法转变为可重复的工作方式。 这把重点从提出孤立的答案转移到将专门知识应用于各种业务，领导层能够了解AI的使用地点及其贡献。

“当每个雇员都以自己的方式使用自己的助手时，个人会变得更好，但公司不会，”拉维察尔的首席执行官称 Joris Van Der Gucht在筹资公告中表示，公司的目的是了解有经验的工作人员是如何工作的，并使这些做法在客户组合中发挥作用。

这也是一项商业主张。 能够确定现有客户之间相关咨询工作的公司可以扩大服务，而不必依赖每个关系管理人独立地注意到每一个机会。 这是否成为收入取决于建议的质量以及公司交付由此而来的工作的能力。

共享上下文层连接客户端信息

Ravical的工作空间描述了一个跨越客户关系管理信息、电子邮件历史、计费记录和文件系统的统一的知识层。 其客户记忆将来自这些来源的信息与同事在工作期间学到的和平台收集的信息结合起来。

这种结构涉及一种熟悉的摩擦根源：顾问在回答一个问题之前，可能需要从几个断开的系统重建客户的状况。 共享上下文可以让不同工作流程获得相同的基本信息，而不是要求每个员工再次组装。

Ravical将平台定位在会计、税务、薪金、审计和咨询工作的各个环节。 其网站包括编写对客户请求的答复以及利用全组合信息确认额外服务需要的例子。 这些例子描述了预期的工作流程，而不是规定每项答复无需专业审查即可进行。

技术上的区别是一般模型的知识与公司本身的工作环境之间的区别。 一个模式可以解释会计概念；一个有用的全公司系统也需要有关的客户记录、业务约定的界限和公司处理这项任务的首选办法。

控制室以成本和价值连接领养

公司控制室让领导层了解各团队使用AI的情况、成本和创造的价值。 这解决了与改进个人任务不同的一个挑战：了解AI的采用是否为本组织带来了可衡量的好处。

仅仅节省时间不一定能增加能力或增加收入。 企业需要知道节省的时间在哪里，工作是否符合标准，工作人员是否能够利用由此产生的能力为客户服务。 中央的能见度可以帮助领导人比较各个团队的收养，并决定进一步投资的地点。

对Ravical来说，这让治理成为产品经济学的一部分。 一致的方法和监督旨在帮助公司扩大它们的专门知识，而管理层的观点则提供了一种方法来评估这些方法是否正在提供有益的工作。

客户保密形状为 AI 设计

会计和税务公司处理敏感的财务和个人信息，使数据分离成为任何共享AI平台的核心。 Ravical的安全网页称每个客户都有单独的上下文，任务在客户和约定之间隔离。

该公司也不使用客户输入，输出或上传文件来训练基础模型。 它描述了实时检索，而不是将这些信息纳入模型权重，以及基于欧盟的AI处理和默认的人类监督。

这些是公司的建筑和合同控制。 公司仍然需要确定哪些记录是相互联系的，谁可以使用这些记录，以及如何在成为专业建议之前审查产出。 只有适当的保密界限与这种知识相通，共享知识层才具有价值。

欧洲扩张种子回合和美国条目

拉维查尔由范德·古赫特，CTO Ken Bastiaensen和CPO Benjamin Vandermarliere于2025年创立。 它以根特和伦敦为基地，将会计软件经验与AI专业知识相结合。 Van Der Gucht先前共同创办了Silverfin公司，公告称为18个国家的1000多家会计师事务所服务。

种子圆背取一站台策略而非单任务工具。 正如公告所描述的，MMC Ventures的投资论调是，成功的公司将让最强的雇员的专门知识在更多的客户中发挥作用，而不是简单地取代工作人员。

Ravical报告的收养提供了商业利益的证据，但释放并没有提供经审计的客户收入收益或标准化的业绩基准测试。 下一阶段将测试其共享的工作流程是否能够适应不同的市场和公司做法，同时保持质量和控制。

Ravical认为会计的AI优势来自客户背景、机构专业知识和可重复执行。 对公司来说，相关的结果是，这种组合是否使公司能够开展更有益的工作，而不要求少数有经验的人承担每一项额外的工作。

本文由 AI宝库自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。

Funding

Ravical Raises $17M Seed to Scale Accounting Expertise With AI

Add Unite.AI to your preferred sources on Google

Accounting firms often grow by adding more people to serve more clients. Ravical has raised $17 million in seed funding to pursue a different model: turning the expertise of a firm’s best people into shared AI workflows that can be applied across its client base.

MMC Ventures led the round, with existing investors Lakestar and Pitchdrive participating alongside new investor Shapers. The financing brings Ravical’s total funding to $25 million.

The company says more than 100 accounting firms across Europe already use its platform. It will put the capital toward product development, team expansion and growth in the UK, Belgium, the Netherlands and Germany, while continuing its European rollout and entering the United States.

Making Expertise a Firm-Wide Resource

Ravical’s argument is that access to AI will become common across the accounting profession, but firms will still differ in how they apply their knowledge. An individual assistant can improve one employee’s output without making that improvement available to colleagues or other clients.

The platform is designed to turn a firm’s established methods into repeatable ways of working. That shifts the focus from producing isolated answers to applying expertise across engagements, with leadership able to see where AI is used and what it contributes.

“When every employee uses their own assistant in their own way, the individual gets better, but the firm doesn’t,” said Joris Van Der Gucht, Ravical’s CEO, in the funding announcement. The company’s aim is to capture how experienced staff work and make those practices usable across the client portfolio.

This is also a commercial proposition. A firm that can identify relevant advisory work across existing clients may be able to expand services without relying on each relationship manager to notice every opportunity independently. Whether that becomes revenue depends on the quality of the recommendations and the firm’s ability to deliver the resulting work.

A Shared Context Layer Connects Client Information

Ravical’s Workspace describes a unified knowledge layer spanning CRM information, email history, billing records and document systems. Its client memory combines information from those sources with what colleagues have learned and what the platform gathers during its work.

That architecture addresses a familiar source of friction: an adviser may need to reconstruct a client’s situation from several disconnected systems before answering a question. Shared context can make the same underlying information available to different workflows rather than requiring each employee to assemble it again.

Ravical positions the platform across accounting, tax, payroll, audit and advisory work. Its website includes examples of preparing responses to client requests and using portfolio-wide information to recognise additional service needs. These examples describe the intended workflow, rather than establishing that every response can proceed without professional review.

The technical distinction is between a general model’s knowledge and the firm’s own working context. A model may explain an accounting concept; a useful firm-wide system also needs the relevant client records, the engagement’s boundaries and the firm’s preferred approach to handling the task.

Control Room Connects Adoption With Cost and Value

The company’s Control Room gives leadership a view of AI usage across teams, costs and the value created. This addresses a separate challenge from improving an individual task: understanding whether AI adoption is producing a measurable benefit for the organisation.

Time savings alone do not necessarily increase capacity or revenue. A firm needs to know where the saved time occurs, whether the work meets its standards and whether staff can use the resulting capacity to serve clients. Central visibility can help leaders compare adoption across teams and decide where to invest further.

For Ravical, that makes governance part of the product’s economics. Consistent methods and oversight are intended to help firms scale their expertise, while the management view provides a way to assess whether those methods are delivering useful work.

Client Confidentiality Shapes the AI Design

Accounting and tax firms handle sensitive financial and personal information, making data separation central to any shared AI platform. Ravical’s security page says each client has a separate context, with tasks isolated between clients and engagements.

The company also says it does not use customer inputs, outputs or uploaded documents to train underlying models. It describes real-time retrieval instead of incorporating that information into model weights, along with EU-based AI processing and human oversight by default.

These are company-stated architectural and contractual controls. Firms still need to establish which records are connected, who can use them and how outputs are reviewed before they become professional advice. A shared knowledge layer is valuable only if the appropriate confidentiality boundaries travel with that knowledge.

A Seed Round for European Expansion and a US Entry

Ravical was founded in 2025 by Van Der Gucht, CTO Ken Bastiaensen and CPO Benjamin Vandermarliere. Based in Ghent and London, it combines accounting-software experience with AI expertise. Van Der Gucht previously co-founded Silverfin, which the announcement says serves more than 1,000 accounting firms across 18 countries.

The seed round backs a platform strategy rather than a single-task tool. MMC Ventures’ investment thesis, as described in the announcement, is that successful firms will make their strongest employees’ expertise count across more clients, rather than simply replacing staff.

Ravical’s reported adoption offers evidence of commercial interest, but the release does not provide audited customer revenue gains or a standardised performance benchmark. The next phase will test whether its shared workflows can adapt to different markets and firm practices while preserving quality and control.

With $25 million raised in total, Ravical is betting that accounting’s AI advantage will come from combining client context, institutional expertise and repeatable execution. For firms, the relevant outcome is whether that combination lets them deliver more useful work without asking a small group of experienced people to carry every additional engagement.

Auto-fetched and translated by AI Gem Box. All rights belong to the original author.