KNIME for Finance: Invoice and dunning process management

This is part of a series of articles to show you solutions to common finance tasks related to financial planning, accounting, tax calculations, and auditing, all implemented with the low-code KNIME Analytics Platform.

Yearly renewals are often at the heart of a business’ responsibility to maintain a healthy cash flow and sustain business growth. Despite its importance, it's often an area that lacks efficiency or procedure.

The goal of a dunning/renewal management application is to track accounts that have paid on time and, more importantly, to implement a procedure to address accounts that are past their allotted due date.

In this post we show how you can improve invoice and dunning process management by using a data science tool. The video already gives you an overview of the KNIME solution and you can find a written walk-through of the workflow below.

We set ourselves the task to build an invoice and dunning process management solution that will both improve the efficiency of the finance team and provide the best service to customers.

In discussion with finance experts, we identified three key objectives to ensure a streamlined dunning process.

1. Track communications with late accounts through follow-ups and reminders to stay organized and keep customers aware of their current status.
2. Create more granular metrics around each customer to plan better in the future and ensure a more seamless renewal process. 
3. Centralize data to give renewal account experts a single location to consume all the data and improve efficiency.

Let's look at how we put the solution together.

One of the most important aspects of the new solution is getting access to the locations where the data related to your accounts is stored. This step is critical to centralizing this procedure away from siloed locations.

Typically, you'll be accessing data through a CRM, internal databases, invoice-related applications, or any combination of the three. This type of more complex data access can be handled easily by data science tools, as they offer a wider range of data operations. A low-code tool gives you a visual, intuitive interface so you don't need programming expertise to use it. KNIME Analytics Platform is also open source, so free for you to download and use.

In our dunning management task, the most important information is the renewal date, followed by unique customer identifiers, associated account personnel, and additional data fields that would be beneficial to the analyst working with this application.

We also want our solution to keep track of communication around the accounts. For example, let’s say there was a message to be sent out to an account via our KNIME solution. Keeping track of communications could help the account team strategize next outreaches or use this information as historical data to plan for future renewals.

Our centralized data will now be the "single source of truth" and make sure we have all our critical data at our fingertips to use easily, at any time.

The data does not provide much use unless it’s put in a more user-friendly and consumable manner. 

Shareable data apps are great for providing an easy way to include advanced data science into business processes without needing to train business users in data science. They give business users access to just the right touch-points and can be easily extended or adapted.

We therefore want to build data app that gives the renewal account experts easy access to all the required information to manage their dunning process, such as: 

• Listing overdue invoices
• Tracking overdue account communications
• Sending new reminders to accounts

While there is always more that can be built into the application, this should provide a good base. If needed, it’s possible to incorporate additional customizations or functionality.

Now that we’re feeding the data into the application and displaying it,  the final step is to understand how people are going to be interacting with the application and provide the renewal account experts with an intuitive way to work with the data.

The data expert can build the workflow that the renewal accounts expert might not have been able to write themselves, but can understand because of the intuitive, visual, low-code environment. The immediate clarity of the approach enables the data expert and the renewal account expert to easily discuss improvements to the workflow.

Let’s walk through how I built the workflow. You can download the Dunning Management Process workflow here and try it out on your own dunning processes.

For the purposes of our example, I first generated a table with dummy data.

The column expressions node is left as a reminder that the “Today’s Date” column by default has a static value, and when ready for production, it should be fed through the expressions node to replace the values with the actual date of when the workflow is running. 

Moving on to the “Select Overdue Accounts Component”—this is where a majority of the applications functionality lives. The following screenshot is the internal content of the component and will be broken down into its separate sections.

What the workflow does:

It reads in a data table that tracks all comments and notes about every overdue customer. The Excel sheet is tightly tied to the outputs of the application. You can go into the “Account Info” component and see what is getting written to the Excel sheet. The refresh button widget allows for users to select an account, and once the button is selected, it will then pull additional info regarding that specific account.

The data is merged, and the output will designate a green or red color to each late account. If the color assigned to the account is green, this means that our account reps have made contact in the past seven days. If it has been longer than seven days, the color will be red, signifying that it is time to try and contact the account once again.

The Table View displays the list of accounts with its respective color code to help guide the application’s consumer. 

The data formatted so that the renewal account expert can easily interact with the application and draw insights from the data.

The workflow provides additional account information and input fields for the renewal account expert to add comments or new information. When the renewal account expert selects an account, they will be able to expand past renewal comments made in current or past renewal cycles. After updating and refreshing, the workflow updates the red colored account to give it the green status.

Note: The outreach part of the application is not in this workflow. The decision is up to the developer to decide whether they want to reach out internally with updated information or to send emails to the account directly through the application.

The renewal account experts will use the data app to manage their dunning processes. You can see the what this looks like in the screenshots of the interface below.

In the video you can see how a renewal account expert can use the data app. They are prompted to select an account, upload custom notes, and finally update the account with the inputted comments.

Note: Find more information about building data apps in the Data Apps Beginners Guide. 

The use of data science tools is already spreading into the finance department! Spreadsheets have served us all well, but when data outgrows the capabilities of conventional tools, a data science tool can serve us with more sophisticated and robust techniques.

Download KNIME Analytics Platform and try out the workflow in this article for your own use case. It's free to use and adapt to your own invoice and dunning processes.