CPD lab · FP&A
Variance commentary
Check an AI-written commentary on a month's management accounts, statement by statement, and catch the wrong figures and the misleading stories before it reaches the board.
Comes with FP&A and forecasting and Business partnering, and with the year. 29 labs open now.
The brief
You are the management accountant at Thistlebrook Bakeries Ltd, a craft bakery with its own shops and a wholesale arm. It is Monday 2 November 2026. An AI assistant has drafted the commentary for the October management accounts, and it goes to the board on Thursday.
The pack has the October management pack (budget against actual), the monthly trend for five cost lines, the budget holders' notes and the AI's draft commentary, split into 12 numbered statements. Every company, person and figure is invented, so you can paste it into any AI tool.
Check every statement against the pack and the notes, using your AI tool if you like. Give one line per statement, with a short reason.
Verdict, one per statement. CORRECT: the figures and the explanation agree with the pack and the notes. WRONG: a figure, a percentage or the direction (favourable or adverse) is wrong. If a statement has a wrong figure or direction, it is WRONG even if its explanation is also off. MISLEADING: the figures are right but the explanation gives a false picture, for example a one-off described as a trend, a timing difference described as a saving, or a reclassification described as a real overspend or saving.
corrected_figure: for WRONG, the right value of the figure that is wrong. For MISLEADING, the underlying variance for that line in £k once the one-off, timing difference or reclassification is taken out. Leave it blank for CORRECT. Write numbers only, with no £, k or % sign, to one decimal place. Favourable is positive and adverse is negative, so 18.0k adverse is -18.0. Percentages are of budget.
Then write a short review note: what the AI got wrong, what you checked, and what goes to a person before the pack goes out.
Objective: To be able to review AI-drafted variance commentary against the management accounts and budget holders' evidence, correcting wrong figures and misleading explanations, so that the commentary gives directors a true picture of performance.
What's in the pack
Every name, figure and email in the pack is invented. Use your own AI tool: Claude, ChatGPT, Copilot or Gemini.
management-pack-oct-2026.csv
October 2026: budget, actual and variance by line (£k; favourable positive)
monthly-trend.csv
Monthly actuals April to October 2026 for five cost lines (£k)
budget-holder-notes.txt
Budget holders' notes on October
ai-draft-commentary.txt
The AI-written draft commentary, 12 statements
The traps
The pack has 6 deliberate traps: cases where AI tools often get it wrong and a person must check. We don't list them here. Finding them is the job, and your feedback names each one you missed.
What you submit
- Your result, as a table with the columns statement_id, verdict, corrected_figure, reason.
- The prompt you used.
- A short review note, 150 to 400 words: what the AI got wrong, what you checked, and what goes to a person.
How it's marked
- Your result, checked line by line against the answer key60%
- Your review note, marked on traits40%
The review note's traits:
- Wrong figures found and corrected
- Every statement tested against the evidence
- The true picture of the month
- The right things go to the right people
- Management accounts kept out of personal accounts
Pass at 80. Under 80, read the feedback and try again. Then write your reflection and copy the entry for your CPD record.
Price
- FP&A and forecasting, £79. The role track and its labs, this one included. No end date.
- Business partnering, £79. The role track and its labs, this one included. No end date.
- Everything for a year, £199. Every course, role track and lab for 365 days.
- Monthly, £19.99 a month, cancel any time. Every lesson, plus 8 marked tasks a month, across every course, role track and lab.
Buying for a team, or want a lab built on your own process? CPD for employers.