Quality of Earnings Automation: What Advisory Firms Are Doing Now
Quality of earnings automation is changing how advisory firms approach FDD engagements in 2026. Quality of Earnings is one of the most judgment-intensive parts of financial due diligence. It is also one of the most time-consuming. That combination makes it an interesting test case for what AI can and cannot do in a professional services context.
What QoE actually involves
For those outside transaction services, a Quality of Earnings analysis is essentially an assessment of whether the earnings a business is presenting are repeatable. It involves identifying items in the P&L that are one-off, non-recurring, or that have been presented in a way that flatters the trading performance.
The output matters enormously. A QoE adjustment can change the valuation of a business by millions. Getting it wrong in either direction has real consequences.
The parts that can be automated
The early stages of a QoE analysis are more systematic than they appear. Before any professional judgement is applied, there is a significant amount of pattern recognition involved.
Finding cost items that appear in isolated periods. Identifying revenue that does not recur across the full dataset. Spotting margin movements that are not explained by the narrative in the information memorandum. Flagging intercompany items that may need to be stripped out.
These are tasks that a well-built AI system can surface reliably and quickly. Not perfectly, and not with the context that an experienced TS professional brings. But well enough to generate a working list that a senior person can then interrogate.
The firms doing this well are not asking AI to produce the QoE schedule. They are asking it to produce the candidates, and then applying judgement to decide what makes the cut.
The parts that cannot
The defensibility of a QoE adjustment depends on narrative as much as numbers. Why was this cost one-off? What does management say about it? Does that explanation hold up against the broader trading history?
That conversation requires a human. It requires someone who has sat across the table from a management team and knows when an explanation does not quite stack up.
AI does not have that context and is not going to have it anytime soon. The professional services firms that understand this are the ones using AI most effectively. They are not trying to replace the judgement. They are trying to free up the time so that judgement can be applied where it actually counts.
What is changing in practice
The practical change in forward-thinking firms is that the QoE workstream starts earlier. Data goes into the platform on day one and a first-cut list of potential adjustments is available before the kickoff call has happened.
That changes the quality of the early conversation with management. Questions are more specific. The team looks better prepared. And the time that used to go into building the initial list can go into testing and refining instead.
For a firm doing several deals at once, that shift in how the first week of a deal feels is significant.
The risk of getting this wrong
The firms most at risk are not the ones ignoring AI entirely. They are the ones adopting it without thinking carefully about where human review needs to sit in the process.
A QoE adjustment that goes into a report without proper scrutiny, because the team assumed the AI had done the hard work, is a serious professional risk. The technology is a tool. The accountability still sits with the advisory team.
Getting the workflow right matters as much as getting the technology right. PinpointAI automates the data preparation and normalisation phase of QofE work, so your team can focus on the analysis that actually drives the report.
