There is a season in every audit firm where the workload curve becomes vertical. The first day of fieldwork triggers a cascade of requests, data wrangling, and manual procedures that consume the team's attention for the next four to twelve weeks. Some of this is unavoidable. Most of it is not. The firms that have moved meaningful work into the pre-fieldwork window are recovering 15 to 25 percent of engagement hours, and the recovery compounds because the time-saved in week one frees attention for substantive testing in weeks two through eight.
Three workstreams that move forward
The pre-fieldwork window can absorb three workstreams that historically lived inside fieldwork. The Prepared-By-Client request cycle, journal entry testing on full populations rather than samples, and external confirmations. Each has a different mechanism for moving forward. Each has a different impact on quality and timing. Each has a different relationship to AI versus rules-based automation. Deloitte's expansion of agentic AI capabilities in its Omnia audit platform, announced in 2025, points to where the Big Four are placing their own bets: enhanced documentation review, automated tie-out procedures, and intelligent agent-driven data gathering, all moved upstream of fieldwork [1].
The hours saved in week one of fieldwork compound. Every hour not spent on data gathering is an hour available for substantive testing the client actually pays for.
The PBC request cycle
The single largest pre-fieldwork opportunity. In a typical mid-market audit, the PBC list runs to 80 to 150 items. Most of those items are predictable: trial balances at specific dates, sub-ledger reconciliations, contract listings, fixed asset rollforwards. Most of them are also requested every year in roughly the same form. Automating the PBC cycle does not mean automating the audit. It means generating the request list from the prior year's engagement, routing the requests to the right client contacts with deadlines, tracking completion, and ingesting the responses into structured working papers. The audit team's first day of fieldwork starts with most of the PBC list already received, validated, and organized. The work saved is roughly 8 to 12 percent of total engagement hours on a mid-market job, captured before any substantive testing begins.
Journal entry testing at population scale
Sample-based journal entry testing is a compromise dictated by historical processing limits. When testing 100 percent of journal entries against risk attributes is computationally feasible, sampling becomes harder to defend. Deloitte's 2024 Finance Transformation Survey found that more than 80 percent of finance professionals spend most of their close cycle on manual reconciliations and data preparation rather than analysis [2]. The same constraint applies on the audit side. Moving journal entry testing to full-population analysis, run before fieldwork starts, gives the audit team a risk-ranked anomaly list to focus substantive testing on, which both improves audit quality and reduces total hours. It also produces a defensible audit trail that satisfies oversight bodies' increasing focus on data analytics in audit.
External confirmations, moved upstream
External confirmations have historically been a fieldwork activity because they require client authorization. Moving the authorization workflow upstream, with structured templates and digital signature, allows the confirmation requests to be sent the day fieldwork begins, sometimes earlier. For a typical mid-market engagement with 30 to 60 confirmations across banks, customers, lawyers, and other counterparties, the elapsed-time saving is two to three weeks. The hours saving is smaller, perhaps 2 to 4 percent of engagement hours, but the calendar saving is what matters: it shortens the engagement by half a cycle, which means the partner can sign earlier, the client gets the report earlier, and the firm's engagement throughput improves.
The hours saved are real. The calendar compression is what reshapes the firm's capacity.
What this means for audit-firm staffing
The hours recovered do not disappear. They reallocate. In firms that have made this work, the dominant pattern is a redistribution from manual reconciliation toward judgment-intensive substantive testing and review. Junior staff hours decline modestly. Senior staff and manager hours stay roughly flat but shift toward higher-value activities. The Wolters Kluwer 2024 Future Ready Lawyer Survey found the same pattern in legal: AI-driven efficiency is reshaping the mix of work rather than reducing total work, with 67 percent of corporate legal departments expecting AI to impact billable hour models [3]. The pattern is recognizable in audit because the underlying economics are the same: time recovered from routine work flows toward complex work, which clients pay for at higher rates because the value is more visible.
The governance layer that has to come with it
Pre-fieldwork automation produces working papers before the audit team has seen them. That is a quality concern unless the automation is governed properly. Three controls have to be in place. First, the rules used to generate PBC requests and to flag journal entry anomalies are documented and version-controlled, with sign-off by the engagement partner before the season begins. Second, the system produces an audit trail of every action taken, so that reviewers can reconstruct what the automation did. Third, the engagement team reviews the automation output before relying on it; the automation is treated as a junior team member whose work is reviewed, not as an oracle whose output is accepted. Belgian and EU audit firms preparing for the EU AI Act's high-risk system rules, which take full effect from August 2026, should be designing this governance now, not retrofitting it later [4].
A note on tool-selection
The temptation is to buy the most-automated platform available. The right question is which platform fits the firm's engagement methodology and which one its team will actually adopt. The Big Four have built proprietary platforms because their engagement volume justifies the investment. Mid-market firms are better served by composable tools that integrate with their existing methodology rather than replace it. The integration work is unglamorous and where the value gets captured or destroyed.
What changes for clients
Audit clients experience pre-fieldwork automation as faster, less disruptive engagements with more substantive findings. The PBC requests arrive earlier and more clearly. The audit team spends less time on data gathering and more time on questions that require management response. The final report arrives sooner. For Belgian mid-market clients specifically, this matters more than headline price: the disruption-cost of audit season is often larger than the audit fee itself, and reductions in disruption translate directly into operating quality for the months that follow.
