For most finance teams, IFRS 16 is the gift that keeps on taking. The standard was designed to bring operating leases onto the balance sheet, but the operational burden is relentless. Every new lease, every modification, every renewal option, and every lease incentive requires a fresh calculation. The discount rate must be determined, the lease term reassessed, and the right-of-use (ROU) asset and lease liability re-measured. Spreadsheets, while flexible, are error-prone. A single misplaced cell reference can misstate the balance sheet, leading to audit findings, restatements, or a qualified opinion. The cost of non-compliance is not just financial; it erodes stakeholder trust.
The friction is compounded by data fragmentation. Lease contracts live in PDFs, emails, and physical files. Extracting the key terms—commencement date, payment schedule, escalation clauses, purchase options—is manual, tedious, and inconsistent. Controllers spend hours, not minutes, on each lease. During month-end close, this becomes a bottleneck. The result: delayed reporting, overworked staff, and a higher risk of material misstatement.
This is where a purpose-built AI workflow changes the game. Instead of replacing your judgment, it handles the heavy lifting: reading the contract, identifying the relevant clauses, computing the incremental borrowing rate (IBR) proxy, and generating the amortization schedule and journal entries. The AI does not guess. It follows a structured prompt that forces it to extract data, apply the standard’s requirements, and present a transparent, auditable trail. You remain the reviewer, but you spend minutes on review, not hours on data entry. The goal is simple: turn a week of lease accounting work into an afternoon.
Why a Structured Prompt Matters
Most AI failures in accounting are not because the model is weak, but because the instruction is vague. If you ask an AI to “calculate IFRS 16,” it will produce a generic response that lacks your specific contract terms, your discount rate assumptions, or your chart of accounts mapping. The solution is an “anatomy of a prompt”—a structured template that breaks the task into components: the objective, the input files, the reference example, the success criteria, and the execution plan. This forces the AI to behave like a diligent staff accountant: read first, ask clarifying questions, then execute.
Below are two copy-paste-ready prompts. The first is for a single lease calculation. The second is for a portfolio-level review, where you need to identify modifications or reassessments. Both follow the same structured format. Replace the bracketed placeholders with your specifics. Do not skip the “read files” step; the AI needs the contract and your accounting policy to produce something usable.
First, read these files completely before responding:
[lease_contract.pdf] — the full signed lease agreement, including all addendums and side letters.
[accounting_policy.md] — our IFRS 16 policy, including our policy for determining the incremental borrowing rate (IBR) and lease term reassessment.
[chart_of_accounts.csv] — our GL account codes for ROU asset, lease liability, interest expense, and depreciation.
Here is a reference for what I want to achieve:
[Upload a completed IFRS 16 working paper from a prior lease as markdown, or describe the exact format of the amortization schedule you use.]
Here’s what makes this reference work:
The reference shows a clear split between the liability amortization (opening balance, interest at the effective rate, payment, closing balance) and the ROU asset depreciation (straight-line). The journal entries are mapped to specific account codes. The discount rate is stated prominently, and the lease term is justified with a note about renewal options.
Here’s what I need for my version / SUCCESS BRIEF:
Type of output + length: A complete working paper in markdown, with a summary table, the full amortization schedule for the lease term, and the first 12 months of journal entries.
Recipient’s reaction: My controller should be able to review the calculation and post the entries without asking me for any additional data or explanations.
Does NOT sound like: A generic textbook explanation of IFRS 16. Do not include theory or commentary. Just the numbers, the assumptions, and the entries.
Success means: The closing liability balance equals the present value of remaining payments using the specified discount rate, and the total expense recognized over the first 12 months equals the straight-line rent plus the interest expense differential.
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 is designed for a single, straightforward lease. The AI will read the contract, apply your policy, and produce a working paper. The clarifying questions step is critical: it forces the AI to ask about the IBR if it is not explicitly stated in the contract, or to confirm whether you want to use the practical expedient for short-term leases. Do not skip this. The execution plan ensures the AI does not jump to a final answer without showing its work.
However, the harder problem is not the initial recognition. It is the ongoing reassessment. Lease modifications—such as a rent reduction, an extension of the term, or a change in the leased area—require a remeasurement. The AI must identify the modification, calculate the new liability using the updated discount rate, and adjust the ROU asset accordingly. The second prompt addresses this portfolio-level review.
First, read these files completely before responding:
[lease_portfolio_index.xlsx] — a list of all leases with ID, counterparty, commencement date, and original term.
[lease_contracts_folder/] — all 25 lease agreements in PDF format, including any amendments or correspondence about rent changes.
[ibr_spreadsheet.xlsx] — our quarterly updated incremental borrowing rates by lease term and currency.
[modification_log.md] — notes from the operations team about any verbal agreements or planned changes to leased assets.
Here is a reference for what I want to achieve:
[Upload a prior quarter’s remeasurement memo, showing how a modification was identified, the new discount rate applied, and the resulting adjustment to the liability and ROU asset.]
Here’s what makes this reference work:
The reference shows a clear decision tree: first, identify the trigger event; second, determine if it is a modification (change in scope or consideration) or a reassessment (change in term or purchase option); third, recalculate using the rate at the modification date. The memo includes a comparison of old vs. new liability, and the journal entry to adjust the ROU asset.
Here’s what I need for my version / SUCCESS BRIEF:
Type of output + length: A risk-ranked list of leases (high, medium, low) that require remeasurement, with a one-paragraph explanation for each, plus a summary table of the estimated dollar impact on the liability.
Recipient’s reaction: My CFO should be able to see the potential exposure and prioritize which leases to review in detail before the external auditors ask.
Does NOT sound like: A generic audit checklist. Do not list every lease. Only flag the ones with a plausible trigger event.
Success means: I can reconcile the total potential adjustment to within 5% of the actual remeasurement, and I have a defensible rationale for why each flagged lease was identified.
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.
The second prompt is about triage. It does not replace the detailed calculation; it tells you where to spend your time. The AI will cross-reference the modification log with the contracts and the IBR spreadsheet. It will flag leases where the term has been verbally extended or where the rent has changed without a formal amendment. This is the highest-value use of AI in lease accounting—not just computing numbers, but identifying the risk hidden in unstructured documents.
One practical tip: when you run these prompts, do not accept the first output. Ask the AI to “show your assumptions” and “recalculate with a 0.5% higher IBR” to see sensitivity. This will catch errors in the discount rate selection. Also, set a rule for your team: the AI output is a draft, not a final posting. You must review the amortization schedule for at least one lease manually each quarter to maintain the audit trail. This keeps the human in the loop without losing the speed.
Next, try applying the first prompt to a complex lease with a purchase option or a termination penalty. The AI will need to determine the lease term based on the exercise of the option, which is a judgment call. Use the clarifying questions to force it to justify its reasoning. Over time, you will build a library of prompts for different scenarios—initial recognition, modification, impairment, and disclosure. Each prompt becomes a reusable asset for your team.
Published on 27 August 2026 on growwithgpt.com