Today’s corporate decision-making processes often require employees to navigate the minefield of complex regulations, from international taxation to government legislation. While the competitive advantage often lies with organizations that are able to move fast in this decision-making process, is it possible to digitize consultations with tax and legal experts in order to seamlessly guide decision-makers towards consistent and compliant decisions?
Recently, a group within Deloitte has turned to Camunda Platform’s DMN Decision Engine to digitize the decision-making process. We spoke with Martin Krejci, Director at Deloitte’s Garage for Tax and Legal, about how his team uses Camunda Platform and about their go-to-market plan.
Camunda: Tell us a little about how your team at Deloitte uses Camunda Platform.
Martin Krejci: Deloitte is a large global company and last year our team was awarded funding by the Deloitte Innovation Board in Germany to kick-start our project — the DecisionService – to digitize tax and legal consultations. In terms of the DecisionService, our team acts like a startup within the business.
Our vision is to scale existing tax and legal expertise to simplify even the most complex decision-making processes and facilitate frequently recurring decisions. The corporate decision-making process is becoming ever more challenging, as employees have to comply with increasingly complex regulations. Seeking recommendations for action often involves a great deal of effort and expertise, which requires consultation with subject matter experts.
However, we can leverage Camunda Platform and Decision Modeling and Notation (DMN) to create pre-defined decision models for any questions, issues, levels of complexity or fields. The DecisionService guides users through an optimized series of questions accessed through a chatbot-style User Interface (UI). Behind the UI there is a DMN model built in close consultation with tax and legal experts, so users will access the same information and will come to the same conclusions as if they had called in a lawyer or tax consultant. One big advantage of the DecisionService is that it can be used for any kind of decision-making processes, and is not limited to a specific domain such as tax and legal. With the broad range of expertise within Deloitte globally, we have experts for just about every subject area in question.
The goal is to ensure that end users will have 24-hour access to up-to-date and reliable information to improve quality, consistency and compliance when making decisions. For example, supporting users to analyze whether revenues from existing or planned business models could be subject to a digital tax in a foreign country.
Camunda: How do you leverage Camunda Platform to build the DecisionService?
Krejci: We use a number of components provided by Camunda Platform.
In DecisionService it’s possible to use the same question in many different decision paths, which can lead to different results. DMN is a highly flexible notation, which proves useful when teams want to avoid not only duplications but also the maintenance of multiple decision paths. By leveraging DMN, we can easily describe and model repeatable decisions to create interchangeable decision models. This significantly reduces development and maintenance time.
However, DMN can be challenging to read, especially for process owners. So Deloitte also leverages BPMN as we build new decision models. Using graphical BPMN notation, our software engineers can easily collaborate with subject-matter experts to model the same pathway of questions they would use in an in-person consultation. Working together using BPMN also has the advantage that it enables process owners to refine the wording of each question and answer, helping end users to understand complex terminology in an accessible, straight-forward manner.
Camunda: We know it’s still early in the process, but what kind of benefits have you seen thus far?
Krejci: The decision model approach has shown numerous benefits for our internal clients, and we anticipate similar results when we go to market in June. Camunda Platform enables us to create both simple and complex decision models, which are provided in our DecisionService. Additionally, it helps immensely with regulatory compliance. AI algorithms can obscure the steps taken to arrive at a specific decision, and are often not 100% accurate, whereas decision models are entirely transparent in showing the actions to be taken to achieve a particular outcome – and they are 100% correct.
Camunda: What’s next for Deloitte’s Tax & Legal Garage and Camunda Platform?
Krejci: The idea for the DecisionService arose in the context of a larger development project, where a solution was needed to support users with complex business issues. We have used our experience gained from this project in the development of the DecisionService. Currently, our DecisionService is market-ready in a first version and we have decision models available. We are now thinking about other solutions where DMN serves as the foundation – identifying all the solutions that can have a decision model inside.
A huge thanks to Martin for sharing his experience with DMN within Camunda Platform. We’ll keep you updated in future blog posts on how Martin’s team, Deloitte and its clients use and benefit from Camunda Platform.