HMRC is using AI and data analytics to identify discrepancies faster and more accurately than ever before. Tax investigations are no longer driven by random checks alone, with enquiries now increasingly triggered by digital records, online activity, and automated risk analysis.

For individuals and businesses, this means HMRC enquiries are becoming more targeted, more data-led, and often far more detailed than in previous years.

With expansion of digital reporting requirements and continued rollout of Making Tax Digital, understanding how HMRC approaches investigations is becoming increasingly important.

The new face of HMRC tax investigations

Over the last 15 years, HMRC’s approach to tax investigations has changed beyond recognition.

Historically, enquiries were often opened through local compliance activity or random selection. Today, the vast majority begin with HMRC’s Connect system, which analyses large volumes of data from multiple sources to identify inconsistencies and potential indicators of non-compliance.

HMRC’s systems now review information from:

• Banks and financial institutions

• Employers and payroll records

• Companies House filings

• Property databases

• Online marketplaces and digital platforms

• Overseas financial reporting systems

Using AI and machine learning, HMRC can compare declared income against spending, asset ownership, property transactions, overseas transfers, and digital activity to identify anomalies.

For example, HMRC may identify:

• Rental income received through Airbnb that has not been declared

• Online marketplace sales through platforms such as Etsy, eBay, or Vinted that appear to exceed casual selling activity

• Inconsistencies between Companies House records and tax filings

• Overseas income or accounts identified through international reporting agreements

• PAYE discrepancies or director loan irregularities

In many cases, an enquiry is only the visible outcome of a much wider data-led review already undertaken by HMRC.

‘The days of random HMRC enquiries are effectively over, with almost every tax investigation now starting with data.’

How HMRC uses AI and big data

HMRC’s Connect system reportedly draws on more than 50 data sources, including international reporting frameworks such as the Common Reporting Standard (CRS) and the Foreign Account Tax Compliance Act (FATCA), which was introduced by the United States and requires financial institutions to report certain financial information relating to US taxpayers.

These systems allow HMRC to receive and compare information from financial institutions and overseas jurisdictions automatically.

HMRC’s algorithms are designed to identify patterns and inconsistencies that may indicate undeclared income or errors in reporting. This can include situations where declared income does not appear to align with an individual’s lifestyle, spending habits, or asset ownership.

While these systems have significantly increased HMRC’s ability to identify risk, automated analysis is not always perfect.

Transactions can be misclassified, digital records may lack context, and entirely legitimate activity can sometimes appear unusual when reviewed solely through data patterns. As a result, innocent discrepancies can still trigger enquiries or compliance checks.

This makes accurate record keeping and clear supporting documentation increasingly important.

Making Tax Digital and faster HMRC intervention

The continued rollout of Making Tax Digital (MTD) is expected to further strengthen HMRC’s data analytics capabilities.

By requiring taxpayers to maintain digital records and submit updates more frequently, MTD provides HMRC with a far more regular flow of financial information than the traditional annual Self-Assessment cycle.

This allows HMRC to identify discrepancies and risk indicators much earlier.

Rather than opening broad retrospective enquiries years after transactions occurred, HMRC may now intervene within months through prompts, nudge letters, or targeted compliance activity.

While this may improve efficiency, it also increases the importance of maintaining accurate and consistent records throughout the year.

What can trigger HMRC scrutiny?

HMRC’s systems are increasingly designed to identify:

• Lifestyle or spending patterns inconsistent with reported income

• Undeclared digital platform or online marketplace income

• Overseas accounts, transfers, or remittances

• Payroll inconsistencies or director loan issues

• Differences between VAT, Companies House, and Self-Assessment filings

• Irregularities identified through Making Tax Digital submissions

Even where discrepancies are innocent or explainable, enquiries can still become time consuming, disruptive, and stressful without the right support.

Preparing for a data-led HMRC enquiry

As HMRC’s use of AI and automated risk analysis continues to expand, proactive preparation is becoming increasingly important.

Effective preparation often includes:

Why early advice matters

Understanding how HMRC approaches investigations, what information it may already hold, and how to respond strategically can make a significant difference to both the outcome and the length of an enquiry.

Early specialist advice can often help taxpayers:

• Respond clearly and accurately to HMRC requests

• Avoid providing unnecessary or misleading information

• Resolve matters more efficiently

• Reduce disruption to business and day-to-day operations

• Minimise potential penalties where issues are identified

Need support with an HMRC enquiry?

If you have received correspondence from HMRC, are concerned about discrepancies in your records, or would like advice on managing tax compliance risks our Tax Disputes and Investigations team can help.

Our specialists support individuals and businesses with HMRC enquiries, voluntary disclosures, compliance checks, and complex tax investigations. We can review your position, manage communication with HMRC, and help bring matters to a close as efficiently as possible.

Fill out the form below and one of our specialists will be in touch.