The audit profession is undergoing a technology-driven transformation. While audits have always involved some degree of automation, intelligent automation takes this to the next level by combining robotic process automation (RPA) with artificial intelligence technologies. This creates bots that can not only automate repetitive tasks, but also read, understand, and extract insights from both structured and unstructured data.
As a result, intelligent automation has the potential to significantly enhance audit quality and efficiency. This article explores why audit is ripe for intelligent automation, exactly how it can be applied, and key considerations for successful implementation.
Why Audit is Suited for Intelligent Automation
Auditing has remained predominantly manual despite waves of automation transforming nearly every other business function. According to a recent report, auditors spend only 27% of their time on value-added work with the rest consumed by collecting, organizing, and verifying data. Intelligent automation promises to flip this ratio.
Here‘s why intelligent automation is well-suited to transform key audit activities:
Data Collection and Analysis
Collecting and making sense of financial and operational data is incredibly labor intensive in traditional auditing. Auditors must gather information from multiple systems and formats, normalize it, and analyze it to assess risk and controls.
With intelligent automation, bots can be trained to log into systems, scrape data in any format, structure it for analysis, and highlight anomalies – saving auditors significant time. For example, RPA can collect HR data from multiple systems while NLP can read contracts and pull out terms and conditions.
100% Testing
Given the impracticality of manually reviewing populations of transactions, auditors traditionally rely on random sampling. However, sampling comes with the risk of missing anomalies.
Intelligent automation enables auditors to cost-effectively analyze entire datasets rather than samples. Bots with embedded analytics can screen through millions of transactions in search of red flags in a fraction of the time. This provides complete coverage and reduces sampling risk.
Continuous Auditing
Traditional auditing is a periodic activity due to its manually intensive nature. However, the risk and control environment is continually evolving and material issues can emerge any time.
Intelligent automation allows auditors to track transactions and controls in near real-time. Bots can be configured to monitor systems, execute analytics, and notify supervisors when risks emerge. This enables auditors to switch from periodic, backward-looking audits to continuous, forward-looking assurance.
Enhanced Risk Assessment
Assessing risk exposure is foundational to determining audit strategy and priorities. However, synthesizing risks across functional areas using traditional methods is complex, inconsistent, and heuristic-driven.
Using intelligent automation, auditors can take a data-driven approach to risk assessment, leveraging predictive analytics on both financial and operational data. This allows them to develop a holistic, empirical view of risk rather than relying on gut feel.
Key Use Cases for Intelligent Automation in Auditing
Now that we‘ve covered why intelligent automation is well-suited for transforming audit, let‘s explore some of the highest impact use cases:
Account Reconciliations
Reconciling accounts across multiple systems is repetitive and well-defined – making it ideal for automation. Bots can log into bank portals, ERP systems, and other sources to extract data. They can then match transactions, flag misalignments, and even make adjustment entries without human involvement.
Master Data Analysis
Understanding customer, product, employee and other master data is critical for auditors. Bots enabled with AI can rapidly scan master data sets of any size, profile the data to build a unified view, identify duplication issues, and detect abnormal data patterns.
Contract Review
Reviewing contracts is essential but requires lawyer-like scrutiny when done manually. NLP-enabled bots can read and parse contracts in bulk to extract key terms and conditions, clauses, dates and more. This allows auditors to easily track contractual risks and obligations.
Journal Entry Analysis
Scrutinizing journal entries is a key antifraud procedure but analyzing millions of lines manually is impossible. Automation allows auditors to implement continuous journal entry testing easily. Bots can screen all entries in near real-time against rules-based (e.g. Benford‘s Law) and machine learning models to detect anomalies.
Report Generation
Consolidating data, findings, and conclusions across audit areas into reports is tremendously laborious. Automation bots can gather data extracts, perform relevant analyses, and auto-populate customizable report templates without human effort.
Key Steps for Effective Implementation
The benefits of intelligent automation in auditing are clear. However, implementation takes thoughtful planning. Here are best practices:
Focus First on Repetitive Tasks
The lowest hanging automation opportunities are repetitive, high volume tasks with structured data. Process mining techniques can identify these tasks by analyzing system logs and process documentation. Start here for quick wins before expanding into more complex workflows.
Take an End-to-End Process View
While task-level automation delivers efficiency, optimizing entire processes end-to-end has a multiplicative effect on quality and speed. Before automated solutions are designed,Detailed process maps should be created to fully understand information flows, decision points and interdependencies.
Design for Exceptions
Even in structured processes, exceptions occur which can break automation. When designing solutions, incorporate capabilities to gracefully handle edge cases like incorrect data formats or unavailable systems. Channel exceptions to appropriate team members to resolve.
Focus on Value-Added Work
As repetitive aspects of auditing are automated,auditors are freed to focus on critical thinking and delivering insights. However,this requires developing skills in areas like data visualization, root cause analysis and presentation. Reskilling strategies should run in parallel to automation.
Take an Agile Approach
Unlike traditional solutions, intelligent automation allows for rapid iterations and expansions. Starting with a minimal viable product, additional functionality can be added iteratively while monitoring quality and adoption. This maximizes flexibility and lowers risk.
Continually Monitor for Improvement
To sustain value, bot performance and outut must be continually monitored once deployed. Analytics can track KPIs like process cycle times, error rates and capacity utilization. Performance degradations point to areas for additional training data, exceptions handling or general enhancements.
The Future of Technology-Enabled Auditing
As promising as early intelligent automation efforts seem, they merely scratch the surface compared to the coming wave of digital transformation in auditing. Emerging capabilities like process mining, robotic desktop automation, natural language generation, and blockchain-enabled smart audits will profoundly reshape auditing over the next decade.
Incumbents that fail to develop technology-forward cultures and future-proof skill sets risk losing talent and relevance. However, those auditing leaders that proactively reimagine their operating models for a digital future will deliver unmatched risk insights and assurance to the businesses they serve. The choice is clear – evolve or dissolve.