Five Key Components of a Scalable Integrated Tax System

Five Key Components of a Scalable Integrated Tax System

Tax revenue is the fuel that powers the government engine and makes resident services possible. It’s in the best interest of both the government and the public that tax collection be efficient and effective, and that interactions between residents and revenue departments are simple and painless. It’s no surprise, then, that many state departments of revenue (DORs) are exploring opportunities to modernize their legacy integrated tax system (ITS).

One challenge, though, is approaching ITS modernization in a way that will optimize long-term ROI. The pace of technology advancement has accelerated exponentially since many states deployed their ITS – often decades ago. Today’s tax systems need to be scalable – able to rapidly incorporate new technology, respond to new security threats, and adapt to policy changes.

To do so, we recommend focusing on five key pillars of a modern, scalable ITS: integrated analytics, security, policy simulation capabilities, self-service applications, and robotic process automation.

Integrated Analytics

Revenue departments should be employing integrated analytics to uncover insights related to core businesses processes – fraud detection, collection, customer service, and more. In addition to leveraging internal processes and data sources, scalable systems must be prepared to incorporate a variety of large external datasets from financial institutions, credit bureaus, other government agencies, third-party data aggregators, and open-source repositories. These datasets, however, will only yield value if agencies apply advanced methods, including artificial intelligence and machine learning (AI/ML) approaches like natural language processing.

For example, states are using AI/ML models integrated into Voyatek’s RevHub Remote Seller module to understand and prioritize online retailers that are subject to taxation due to the Wayfair decision. AI/ML identifies and classifies retailers into low effort interventions – like “nudge letters.” This has proven to be a source of tens of millions in new revenue for states using it. This creative use of AI/ML has increased voluntary compliance and freed up staff to focus on pursuing cases unlikely to resolve on their own.


The security landscape is constantly evolving, with cyber criminals using new tactics for identity theft and tax fraud every day. The IRS estimates that tax noncompliance and fraud cost the United States hundreds of billions of dollars per year, so it’s essential that revenue agencies focus on strengthening and enhancing security features to ensure compliance with current standards, protect taxpayers’ information, and minimize revenue loss.

Here, again, advanced analytics and AI/ML come into play. Machine learning algorithms can quickly scan large data sets to detect anomalies that indicate potential fraud. Voyatek’s Revhub Insider Threat Solution uses advanced techniques like cluster analysis and text mining to detect fraud suspicous activities that would normally go undetected by traditional Security Information and Event Management (SIEM) tools. Furthermore, advanced analytics can identify specific techniques and methods used by bad actors and leverage this knowledge to develop and adjust policies—improving defenses against future attempts.

Policy Simulation Capabilities

State tax agencies are crucial partners in the front-end development, modeling, simulation, and analysis of potential tax policy changes proposed by the Governor and Legislature. Yet many struggle to develop models that adequately identify the effects of federal and state tax law changes and meet the performance and timeliness expectations of lawmakers, fiscal analysts, and state residents. Legacy tax simulators fielded by states often rely on complicated and difficult to maintain spreadsheet modeling or an agency grown system that requires significant IT assistance to load data.

Given the important role state tax agencies play in both the front-end development and back-end implementation of tax policy, we recommend embedding a tax policy simulator as part the ITS.

RevHub Tax Law Simulator: RevHub offers modeling and analysis of proposed tax policy changes for potential revenue impacts as well as existing tax laws to anticipate changes in tax compliance campaigns and scenario planning to determine specific areas of impact.

For example, by embedding the simulator in the ITS, analysts can track linkages between taxpayers in one tax type to their activities in other tax types, particularly in cases where there are potential tax policy changes across more than one tax type that need to be simulated. In addition, analysts can estimate revenue impacts not just at an aggregate state level but also across taxpayers based on geographic (e.g., municipality, county, legislative district), demographic (single, married, head of household, number of dependents), and economic dimensions. Integration of Tax Policy Simulation tools also reduces the thousands of hours staff spend coding legislative changes into that tax system.

Self-Service Applications

According to a recent Forbes survey, 78% of American adults prefer to bank via a mobile map or website. It’s only logical to assume constituents would prefer a similar experience when interacting with revenue agencies. DORs should invest in solutions that minimize face-to-face interactions and empower taxpayers to access the services they need online. Ideally, these self-service solutions will allow a taxpayer to log into a portal and instantly view comprehensive information about their tax history, including when they made payments, the amounts, a record of any past issues, etc.

Very few states offer that capability today, but there’s no reason every state can’t. Revenue departments already have this data. With the right integration solution, like RevHub, and integration with resident-facing portals, they can consolidate, analyze, and display this information.

While no solution can eliminate the need to transact with taxpayers in-person or via phone and mail entirely, smart investments in the right technology can ensure that that only the cases with the highest need end up in a customer service queue. Virtual assistants or chatbots can direct taxpayers to the right information, while AI/ML programs can, over time, identify which cases are most likely to need human intervention.

Robotic Process Automation (RPA)

Currently, many DOR workflows rely heavily on manual processes. This includes tasks such as processing overdue tax payments, identifying past-due accounts that will yield the highest amount of collections at the lowest cost, and discerning taxpayers who are likely to “self-cure” and require no further contact. All of these procedures necessitate significant human intervention. However, RPA provides a viable solution. By employing AI bots to track employee activities and learn process steps, RPA automates end-to-end processes. This can lead to DORs automatically categorizing taxpayers for more efficient workflows. Ultimately, this results in faster tax processing and improved taxpayer experiences.

-Voyatek Leadership Team