A few use cases

To appreciate the potential of digital twins in the finance function, businesses will need to think beyond efficiency.

We examine a six-step invoicing process at a professional services firm to understand the wide-ranging benefits of digital twins.

Scenario 1: Forward-looking solution (proactive approach)

Assuming that the finance head of a professional services company is dealing with the planning and budgeting of a project recently awarded by a key client. During one of the invoicing cycles, the project team has completed Step B, as explained in Figure 1, and is moving on to Step C.  

Here, a digital twin can help understand the chances of realizing the forecasted revenue value and its variances percentage. It will also highlight the progress of the process, allowing the finance team to get the probable stats by the end of the final step (step F). Additionally, a digital twin can also identify potential issues through self-learning analytics and suggest ways to fix them.

Scenario 2: Backward-looking solution (zero-knowledge solution)

As SMEs with technological and analytical capabilities are joining the CFO team, every team member should be able to use the finance systems with ease. Therefore, the systems must be interactive enough for the users to understand and comprehend the financial process. Further, integration of the finance system with a cross-department data and KPIs system is required to generate qualitative insights. 

For example, if the profit margin of a project that is coming to an end is lesser than what was expected, the question that arises is, “What should have been the best course of action to achieve what was planned?” Thus, a system with the capability to articulate what could have been done to achieve the target will ease the work of the finance head and improve organizational visibility and operability across departments. The system will also give the non-finance team members an idea of what went wrong and how they need to proceed.

 Normally, to get a holistic picture of the organizational financial status, coordination with members of different departments like marketing, sales, IT, finance, and analytics teams is vital. A digital twin-enabled system will provide access to data from different departments without the need to connect with any member.

Here are a few more areas where digital twins can work wonders for the CFO organization:

• Capital planning: Optimization of capital planning scenarios is critical to ensure resiliency. Most CFOs will agree that having granular and real-time access to the cash flow process can help improve forecasting, resulting in a more financially secure and resilient business.

• Cash management:  Digital twins can help improve the balance sheet by gathering telemetry from financial transactions like vendor and customer invoices and analyzing simulation data of various situations. Similar to a recommender system, digital twins can aid in resolving vendor or collection conflicts and highlight from where money will flow in or out. Additionally, it can help identify the amount of cash needed for immediate expenses, allowing the treasury office to forecast cash in hand.

• Revenue improvement: Digital twins can be applied in finance operations like billing as they can scrutinize the whole system and track down revenue leakage areas. Digital twins can also recommend methods for a workaround or substitute the leakage cost center.

• Cognitive contract automation: Digital twins can help better understand clients' preferences and goals, assess their risk of contract default, and advise the best course of action, all of which will improve customer experience. This will further help frame better contracts with the customer and align resources already involved with the client, in addition to expanding the knowledge repository. 

Creating a digital twin for your organization 

Simple prototypes, iterative model creation, and modest iterative trials can be used to initiate the development of digital twins.

The program should be in line with the data collection strategy, as unbiased data is an important aspect of its accuracy. The value, effectiveness, and quality of the services can be increased by capturing data at the atomic level, experimenting with various improvement scenarios, and leveraging machine learning algorithms.

Things to keep in mind before building your digital twin:

Data collection

Financial and operational data is the lifeblood of a corporation, yet no one can vouch for the quality of the data. Poor data quality can impact the revenue of the organization. It also increases the complexity of data ecosystems and leads to poor decision making.

   Here are some best practices to improve data quality:

• Set the KPIs and data quality standards across your organization.

• All data should be cleansed, deduplicated, and harmonized for consistency.

• Build a trust-based semantic model to notify and rectify data discrepancy.  

• Data should not reside in silos and should be accessible in real time through a single source event-driven data-sharing model.

• Treat data as a product and identify the metadata in a data collection to retain or remove.
   

Reduce data latency

    In the past, organizations have utilized various data platforms matched to separate business lines that contain data in separate and distinct contexts called silo data centers. To reduce siloed systems, data fabric can be leveraged to enable decision-makers to view the data more consistently across the organization. This will allow CFOs to better comprehend the customer lifecycle and create links between previously unconnected data. 

How data fabric acts as a real-time, single source of truth:

• Data integration: Gathering data from diverse sources and giving a comprehensive view of the data

• Data discovery: Analyzing data from diverse sources to identify patterns

• Data governance: Managing international and internal corporate data governance

• Data curation: Creating, organizing, and maintaining data sets so they can be accessed and used by people looking for information; involves collecting, structuring, indexing, and cataloging 

• Data orchestration: Enabling businesses to automate and streamline data-driven decision making by making data tools available for analysis
   

Strengthen modeling

The intelligence layer or modeling division of an organization must have a robust model production operations process. Starting with small models, digital twin systems will be able to assess conditions, rules, policies, standards, and regulations, and then predict and provide suggestions. The system can be further expanded into other departments within the organization or implement the same in offices located in different geographies. As the data quality and availability improve, the model will become more robust and accurate.
 

Digital twin visualization

All business lines can be monitored through a digital twin dashboard, which will identify optimization opportunities and the likelihood of process failures. The dashboard will also offer suggestions to help users run processes so that users without finance domain knowledge can also run financial processes.