Employee stock options (ESOs) are a cornerstone of compensation strategy for growth companies, but their accounting treatment under IFRS 2 creates persistent friction for finance teams. The core problem is valuation complexity: unlike traded options, ESOs lack observable market prices, forcing companies to use option pricing models—typically Black-Scholes or binomial models—that require subjective inputs such as expected volatility, expected life, dividend yield, and risk-free rate. A single percentage-point change in volatility can swing the fair value by 10-15%, directly impacting the profit and loss statement. Meanwhile, IFRS 2 mandates that these estimates be revisited at each reporting date, creating a recurring compliance burden that strains small finance teams.
The second layer of friction is disclosure compliance. IFRS 2 requires extensive narrative and quantitative disclosures: weighted-average exercise prices, contractual lives, valuation assumptions, and reconciliation of outstanding options. Producing these disclosures manually is error-prone, especially when options are granted in multiple tranches with varying vesting conditions. Controllers often spend 8-12 hours per quarter just preparing the IFRS 2 footnote, with the risk of material misstatement if assumptions are inconsistently applied.
ChatGPT solves this by acting as an IFRS 2 specialist that can ingest your option grant data, apply the correct valuation methodology, generate compliant disclosures, and flag assumption inconsistencies—all in minutes rather than hours. It doesn’t replace the judgment of a qualified accountant, but it eliminates the mechanical drudgery and reduces the risk of arithmetic errors. For CFOs and controllers, this means faster closes, fewer audit adjustments, and more time for strategic analysis of how option programs affect diluted EPS and shareholder value.
How to Use ChatGPT for IFRS 2 Compliance: Two Critical Workflows
The key to getting reliable output from ChatGPT for accounting tasks is structuring your prompt with the right context, constraints, and success criteria. Below are two ready-to-use prompt templates. The first handles the core valuation and journal entry calculation. The second tackles the disclosure note preparation. Copy these directly into ChatGPT, replacing bracket placeholders with your actual data.
First, read these files completely before responding:
[grant_terms.csv] — Contains grant date, number of options, exercise price, vesting schedule, expected life, risk-free rate, expected volatility, dividend yield, and forfeiture rate for each tranche.
[prior_period_assumptions.md] — Lists the assumptions used in the previous reporting period for the same option plan, including the source of each assumption (e.g., historical volatility calculated from daily returns over 3 years, risk-free rate from government bond yield curve).
[IFRS_2_excerpts.md] — Key paragraphs from IFRS 2 on grant-date fair value measurement, vesting conditions, and forfeiture adjustments.
Here is a reference for what I want to achieve:
A valuation output that matches the format and calculation methodology used in our external valuation report prepared by [Valuation Firm Name], which applies the Black-Scholes model with continuous dividend yield and uses the simplified method for expected life as permitted by IFRS 2.
Here’s what makes this reference work:
– All assumptions are documented with a clear source and rationale
– The valuation is broken out by vesting tranche (cliff vesting vs. graded vesting)
– The journal entry shows debit to share-based compensation expense and credit to share option reserve, with a separate line for the deferred tax asset
– Forfeitures are estimated at grant date and trued up each period
Here’s what I need for my version / SUCCESS BRIEF:
Type of output + length: A table with fair value per option, total fair value, expense per period, and journal entry for the current quarter. Maximum 2 pages.
Recipient’s reaction: The CFO should be able to approve the journal entry immediately without asking for clarification on methodology.
Does NOT sound like: A generic textbook explanation. Must reference the specific grant terms from my data file.
Success means: The total expense recognized matches within 1% of what our external valuation firm would calculate using the same inputs.
My context file contains my standards, constraints, audience. Read it fully before starting.
DO NOT start executing yet. Ask clarifying questions first.
Give me your execution plan (5 steps max) before you begin.
This first prompt works because it forces ChatGPT to act as a specialist who must verify assumptions before calculating. The key insight is the “reference” section: by telling the model what a good output looks like (matching your external valuation firm’s methodology), you constrain it to produce something that will pass audit scrutiny. The success criteria of “within 1% of external valuation” sets a measurable benchmark. When the model asks clarifying questions—which it will—it will typically ask for the exact risk-free rate source or whether to use graded vs. straight-line vesting, which are exactly the questions a senior accountant should be asking.
First, read these files completely before responding:
[option_schedule.xlsx] — Contains grant date, exercise price, number of options outstanding, number exercisable, weighted-average remaining contractual life, and weighted-average exercise price for each grant from 2023 to current period.
[valuation_assumptions.md] — Lists the Black-Scholes assumptions used for each grant: expected volatility (with calculation method), expected life, risk-free rate, dividend yield, and forfeiture rate.
[prior_year_disclosure.docx] — The IFRS 2 note from the previous year’s annual report, including the format and level of detail accepted by our auditors.
[peer_comparison.pdf] — IFRS 2 disclosures from three comparable public companies in our industry, showing best practices for narrative explanations of assumption changes.
Here is a reference for what I want to achieve:
A disclosure note that follows the structure of [Peer Company Name]’s 2025 annual report, which was praised by their audit committee for clarity and completeness.
Here’s what makes this reference work:
– The note starts with a brief narrative describing the purpose of the option plan and the types of grants issued
– A table shows the movement in number of options (opening, granted, forfeited, exercised, expired, closing)
– A second table shows the weighted-average exercise price and remaining contractual life for each grant
– Assumptions are listed in a separate table with a footnote explaining how each assumption was determined
– The note includes a sensitivity analysis showing the impact on fair value if volatility changes by +/- 10%
Here’s what I need for my version / SUCCESS BRIEF:
Type of output + length: A complete IFRS 2 disclosure note, approximately 3-4 paragraphs plus 2 tables and 1 sensitivity table. Maximum 3 pages.
Recipient’s reaction: The audit partner should say “this is clean” and not request any additional detail or revision.
Does NOT sound like: A boilerplate template. Must reference our specific grant dates, exercise prices, and vesting schedules.
Success means: The disclosure passes our internal review checklist for IFRS 2 compliance, which includes 12 specific checkpoints (e.g., reconciliation of outstanding options, assumption disclosure, sensitivity analysis).
My context file contains my standards, constraints, audience. Read it fully before starting.
DO NOT start executing yet. Ask clarifying questions first.
Give me your execution plan (5 steps max) before you begin.
This second prompt addresses the disclosure pain point directly. Notice the reference to a peer company’s disclosure that was “praised by their audit committee.” This gives ChatGPT a concrete benchmark for quality. The success criteria referencing a 12-point checklist is powerful because it forces the model to verify its own output against specific requirements. When you run this prompt, ask ChatGPT to generate the checklist as part of its execution plan—this gives you a built-in review tool. One practical tip: after ChatGPT produces the disclosure, ask it to “identify the three assumptions that have the most material impact on your disclosure and explain why an auditor would scrutinize them.” This cross-checks the model’s internal consistency.
For controllers implementing these workflows, start with a single option grant on a quiet afternoon. Upload your grant terms and assumptions, run the first prompt, and compare the output to your manual calculation. Once you trust the valuation output, run the disclosure prompt for the same grant. After two or three cycles, you will develop a feel for where ChatGPT needs more context—typically around forfeiture rate methodology or the interaction between vesting conditions and expected life. Over time, build a “context file” that contains your company’s accounting policies, materiality thresholds, and auditor preferences. Upload this context file at the start of each session to dramatically improve output quality.
The bottom line: ChatGPT will not replace your judgment on complex valuation questions like whether expected volatility should be based on historical data or implied volatility from traded derivatives. But it will eliminate the 80% of the work that is mechanical calculation and formatting. For a company with 5-10 option grants per year, the time savings from using these two prompts is approximately 15-20 hours per quarter. More importantly, the consistency of the output reduces the risk of a material audit adjustment—which is worth far more than the time saved.
Published on 27 July 2026 on growwithgpt.com