How AI Multiplies Fund Analysis Efficiency in Financial Crime Cases — Reshaping the Forensic Workflow

The Caseload Benchmark Nobody Talks About Until It's Broken

Law enforcement agencies have long worked from rough benchmarks for how many active cases a single investigator should reasonably carry. Property crime units generally aim for somewhere in the range of 12 to 15 cases per month per detective (Prummell, Allocation of Personnel: Investigations). For major crimes units, the widely cited industry standard is 6 to 8 cases per month, or roughly 72 to 96 per year — a figure Sheriff William Prummell narrowed further using case-assignment data from the Charlotte County Sheriff’s Office, arriving at an average of 6 cases per month, or 72 per year, for that agency specifically (EVAWI, Law Enforcement – Investigator Caseloads). A separate audit conducted in Portland, Oregon in 2007, which also reviewed data from nine other cities, found a median annual caseload of 54 cases per investigator, compared to Portland’s own 5-year average of 56 (same source).

These benchmarks aren’t just historical curiosities — the strain behind them has continued to build. A 2025 peer-reviewed study based on interviews with 45 detectives across five police forces in England and Wales found rising caseloads compounding with a growing volume of digital evidence and reduced supervisory support. The researchers describe detectives as caught in what they call a “paradox of professionalisation”: procedures originally designed to raise investigative standards are now widely experienced as administrative weight that pulls time away from actual inquiry, accelerating burnout rather than easing it.

These numbers exist because they work — up to a point. They assume each case, on average, takes a roughly predictable amount of investigative time. Financial crime and corruption cases break that assumption almost immediately.

Caseload vs. Workload: Two Different Problems Wearing the Same Name

Discussions about investigators being overwhelmed often combine two different questions.

The first is caseload: how many cases are assigned to an investigator, how those cases are distributed across a unit, when cases should be prioritized or administratively closed, and how supervisors identify investigators carrying an unsustainable number of active matters. This is primarily a management and resource-allocation question.

The second is workload: how much actual effort each case requires. That includes the amount of evidence that must be reviewed, the number of sources that must be compared, the number of manual steps involved, and the time required to move from raw evidence to an actionable investigative lead. In complex financial crime cases, workload also grows with the number of relationships investigators need to establish across people, transactions, accounts, communications, companies, assets, and timelines.

Caseload measures how many cases investigators handle; workload determines how much effort each case requires. As workload increases, sustainable caseload decreases.

For many types of casework, agencies can address caseload through assignment and prioritization without fundamentally changing the amount of work required for each case. Financial crime and corruption investigations can be different. When the analysis required per case increases, the number of cases a unit can realistically carry decreases.

Available investigative capacity ÷ Average workload per case = Sustainable caseload

If the average time required to analyze each case rises, sustainable caseload falls—even when the number of investigators remains unchanged.

Conversely, reducing the time spent on repetitive analysis can increase the practical capacity of the same investigative team. That does not mean investigators should simply be assigned more cases; it means the unit has more capacity to devote to cases that require human judgment and follow-up.

Why Financial Crime Cases Break the Curve

Financial crime investigations often require investigators to connect fragmented data across multiple sources before meaningful relationships and investigative leads emerge.

Standard caseload benchmarks work best when the investigative effort required per case is reasonably predictable. Financial crime and corruption investigations often challenge that assumption because a single case may span multiple evidence sources and data environments.

An investigation might involve bank transaction records, phone and messaging logs, digital payment records, digital evidence from mobile devices and computers, property and asset registries, business ownership filings, and other records. Each source may use a different format, come from a different system, and represent a different part of the timeline. The information does not necessarily arrive as a connected investigative picture.

The challenge, therefore, is not simply collecting more data. It is making sense of relationships across data that was never designed to be analyzed together. An investigator may need to determine whether a transaction is connected to a particular person, whether a company is related to a suspect, whether communications align with financial activity, or whether an asset is consistent with the known financial profile.

Those relationships can turn a seemingly manageable evidence set into a time-intensive analytical task. The more manual cross-referencing required in each case, the less representative a simple case-count benchmark becomes.

Solvability Screening Gets Harder When the Evidence Is Fragmented

One of the standard tools agencies use to manage caseload is solvability screening — assessing early whether a case has enough evidence to realistically be solved, so limited investigative time isn’t spent on cases with no real path forward.

This works well when solvability is visible early. In financial crime investigations, it often isn’t. Whether a case is solvable—whether there is actually a pattern connecting the transactions, the contacts, and the assets—may only become clear after substantial cross-referencing has already taken place. The screening step intended to save investigative time can therefore require meaningful analysis before a reliable judgment can be made.

This is compounded by an experience gap that shows up consistently in financial crime units: the investigators who are fastest at spotting these patterns tend to be the most experienced ones, and that pattern-recognition skill doesn’t transfer easily. A senior investigator’s instinct for “this fund flow looks off” isn’t something that can simply be handed to a newer team member – it’s built case by case, over years. That makes it difficult for agencies to consistently transfer analytical experience across investigators and teams.

Reshaping the Investigative Workflow

None of this means that traditional caseload-management practices—fair case assignment, prioritization, solvability screening, and supervisory oversight—are no longer useful. They remain important.

The issue is that these practices can become less effective when the effort required per case varies dramatically. For financial crime and corruption investigations, the more useful question is not only, “How many cases is each investigator carrying?” It is also, “What is driving the amount of time each case consumes?”

That question shifts the conversation from case distribution alone to investigative capacity. If a unit can reduce the time required to connect and interpret information within a case, its practical capacity can increase without changing its headcount.

This is where a broader category of technology—often referred to as investigative analytics—becomes relevant. Investigative analytics encompasses technologies that help investigators organize, connect, analyze, and interpret information across multiple sources; identify relationships, patterns, and anomalies; and surface potential investigative leads that may be difficult to find through manual review alone.

For financial crime investigations, this value is particularly clear in fund analysis. The challenge is not just the volume of transactions, but the time required to turn fragmented evidence into usable data, identify meaningful patterns, and handle sensitive case information securely. SalvationDATA Analysis Solution addresses these challenges through three core capabilities: Structure, Spot, and Secure.

Structure. Built-in multimodal AI models process bank statements, third-party payment records, spreadsheets, PDFs, images, audio, video, and other case evidence, automatically recognizing, standardizing, and organizing fragmented information into analysis-ready data.

Spot. AI-powered fund-analysis models identify special and unusual transactions, trace funds across multiple transfer hops, detect activity outside normal hours, and surface frequently recurring accounts and counterparties—reducing the need for manual cross-referencing.

Secure. SalvationDATA Solution operates fully on-premises and offline, without relying on external large language model services. Financial records, communications, and other sensitive evidence remain within the investigator’s own environment, helping reduce data-exposure risks associated with cloud-based AI analysis.

These capabilities can be applied across a wide range of financial crime cases, including:

• Money laundering
• Fraud and financial scams
• Bribery, corruption, and embezzlement
• Shell-company and nominee-ownership schemesused to disguise who actually controls an asset
• Tax evasion and tax fraud
• Cross-border fund transfersdesigned to move money beyond a single jurisdiction’s visibility
• Terrorist financing
• Organized crime and illicit proceeds
• Cybercrime and ransomware-related payments
• Virtual-currency-linked activity

Together, Structure, Spot, and Secure move AI beyond individual analytical tasks to reshape the investigative workflow. By reducing the time spent preparing and cross-referencing data, the solution allows investigators to focus on what AI cannot replace: evaluating findings, developing investigative hypotheses, and deciding what to investigate next.

The next step is already taking shape: AI is moving from a tool for analyzing data to a technology that helps investigators make sense of entire cases. SalvationDATA is bringing this approach to financial crime investigation with a new AI-powered solution, designed to help investigators uncover connections, accelerate analysis, and turn complex data into actionable investigative leads.