Financial Controller Interview Questions

I run the month-end close against a detailed checklist that I maintain and update after every close cycle. The checklist is sequenced: revenue recognition and accruals first, then prepayments, fixed asset depreciation, intercompany eliminations, and finally payroll reconciliation. I assign owners to each task with deadlines that build in enough buffer for review before the final numbers go to the board pack. The biggest risk in a close is a late adjustment that forces a restatement of the P&L in the first week after close, so I push hard to get all material accruals estimated by day three of the following month rather than waiting for exact invoices. I also run a variance check on each line item against prior month and budget before sign-off: anything outside a tolerance band gets a written explanation from the relevant budget holder. After close, I do a brief retrospective to capture what caused delays and update the checklist for next month.

I start the budget process three to four months before the new financial year. The first step is agreeing the top-down assumptions with the executive team: revenue growth targets, headcount plan, and any known capital commitments. I then work with each department head to build a bottom-up model for their cost base, anchored to those top-down targets. I challenge line items where the department head cannot give me a clear business rationale. Once the budget is approved by the board, I produce monthly management accounts that compare actual results to budget and to the prior year, with a written narrative on the key variances. Quarterly I run a reforecast: I use the year-to-date actuals as a base, incorporate any changes to the business outlook, and update the full-year projection. The reforecast is the document I use to have candid conversations with department heads about whether they are going to land within their budget or need to find savings.

My first step is to determine whether the variance is a timing difference or a permanent deviation, because they require completely different responses. A timing difference, where a cost was expected in March but landed in April, needs to be noted but does not change the full-year view. A permanent variance, such as headcount being added ahead of plan or a large unbudgeted project, changes my full-year forecast and requires a conversation with the relevant budget holder. I decompose variances into their components: for revenue, I separate price from volume from mix effects. For costs, I distinguish between fixed and variable components and look at whether the variance scales with activity or is genuinely out of control. I present variance analysis in management accounts with a narrative column, not just a numbers column, so the executive team can make decisions rather than just read figures. I also track variances over time to identify patterns, because a recurring small variance is often a signal that the budget assumption was wrong from the start.

During our annual budget process, the sales director submitted a revenue plan that assumed 40% growth in a market where our actual win rate had been declining for two quarters. The plan was based on a pipeline that included several deals that had already been lost by the time the budget was submitted. I pulled the CRM data myself and compared the submitted pipeline to the actual deal log. I prepared a one-page summary showing the win rate trend, the pipeline composition by stage, and a sensitivity table showing what revenue would look like at three different win rate assumptions. I asked for thirty minutes with the CFO and sales director together rather than raising it in a broad meeting. The sales director was initially defensive, but when I walked through the pipeline deal by deal, he acknowledged that two of the largest opportunities were unlikely to close in the year. We agreed on a revised plan with a 25% growth assumption and built in a mid-year review. The 25% target was hit.

During a quarterly close I was reviewing the draft board pack and noticed that the reported gross margin for one of our business units was 3.2 percentage points higher than the previous quarter, without any obvious operational explanation. I flagged it immediately and traced the variance. The underlying cause was that an intercompany recharge had been coded to the wrong cost centre, which had inflated the gross margin for that unit while understating the costs in the shared services function. The error was worth around £180,000 and would have materially misstated the performance of both units. I corrected the coding, reran the reports, and updated the board pack before it went to print. I also put a recurring reconciliation check into our close process so that intercompany recharges are validated against the master recharge schedule before the P&L is finalised each month. The error was caught because I look at trends, not just the current period in isolation.

When I joined my last company, the month-end close was taking eleven working days. Most of that time was lost to chasing accruals from department heads by email and waiting for manual journal entries to be prepared and checked. I mapped the entire close process in a flowchart and identified three bottlenecks. First, accrual requests were going out too late: I moved the deadline forward by three days and sent automated reminders through the finance system. Second, recurring journals (depreciation, prepayment releases, payroll accruals) were being prepared manually each month from scratch: I templated them in the ERP so they could be posted with a single authorisation step. Third, the review stage was sequential: I moved it to a shared workbook so the finance manager and I could review simultaneously rather than passing the file back and forth. Within two close cycles the process was down to six working days. That freed up meaningful time for analysis and business partnering rather than close administration.

I think about controls at three levels: preventive, detective, and corrective. Preventive controls stop errors before they happen: segregation of duties so the same person cannot both raise a purchase order and approve it, authorisation limits that match individuals to appropriate transaction values, and system access controls that restrict which users can post to which accounts. Detective controls catch errors that have already occurred: monthly bank reconciliations, supplier statement reconciliations, three-way matching of purchase orders, goods receipts, and invoices, and regular review of the aged creditor and debtor reports for unusual items. Corrective controls address issues once detected: a clear escalation path for material variances, a documented audit trail for any journal posted above a threshold, and periodic internal audits of high-risk areas like expense claims and petty cash. I document the control framework and test it at least annually. Controls that exist on paper but are not tested give false comfort.

I build forecast models with a clear separation between assumptions, calculations, and outputs. The assumptions tab is the only place where inputs are entered: everything else is driven by formula. This makes it easy to run scenarios and means a reviewer can understand the model by reading just one tab. For a rolling forecast I use actuals year to date as a locked base and then project the remainder of the year using a combination of run-rate extrapolation for stable cost lines and specific project-by-project estimates for capex and one-off items. I version the model at the start of each forecast cycle so I can compare the current forecast to prior cycles and understand what changed and why. A forecast that cannot be explained in terms of specific assumption changes is not a forecast: it is a guess. I also build in a simple dashboard that flags any assumption that has moved more than a defined threshold from the prior cycle, which forces an explicit sign-off rather than silent drift.

I treat department heads as my primary customers for management information. That means the reports I give them need to be readable without a finance background: plain-language narratives, no unexplained accounting codes, and a clear "here is what this means for you" summary at the top. I schedule a thirty-minute monthly review with each major budget holder to go through their P&L together rather than just emailing a report. In those meetings I ask them to explain variances in their own words, which surfaces context I would not get from the numbers alone. When a department is running ahead of budget, I do not wait for the year-end to have that conversation: I flag it at the earliest sign and work with them on whether it is a timing issue or a real overspend that needs corrective action. I also involve department heads in building their own budgets from the start rather than handing them a top-down number to accept, which gives them ownership and makes them far more likely to manage within it.