AI for accountancy firms: a practical guide for compilation practice

Thursday afternoon at the office. Tomorrow you are discussing the draft annual accounts of an installation company and the file has to be finished today. So you open the trial balance, run through the suspense accounts, lay the VAT categories next to the general ledger and jot down questions for the client as you go: what is that balance on the queried items doing there, and why are the car costs so much higher than last year? You can only assess once the research is done. And it is exactly that research that eats up the afternoon.

That is precisely where AI becomes practical for accountants. Not a chatbot with generic answers, but an AI assistant that is connected directly to the client records and answers your review questions with the entries attached. In this guide you will read why compilation work is the logical starting point, which tasks already lend themselves to it today and how your firm can get started with it responsibly.

Why compilation work is the logical starting point

With a compilation engagement under NBA Standard 4410, you compile the annual accounts or other financial information based on data the business owner provides. You apply your administrative and reporting expertise to this, but you do not perform an audit and you give no assurance: there is no auditor's report with certainty.

That nature makes compilation work especially suitable for learning to use AI on client data. The work consists for a large part of reconciling, investigating and querying: do the VAT categories match the general ledger, what is sitting in the suspense accounts, which items deviate and which questions do you put to the client. That is research and comparison work, and a connected AI assistant is strong at that. The assessment of what comes out of it remains your profession.

The barrier is also lower here than in assurance work. You experiment with work you already do on client data anyway, within the access your firm already has.

Deploy AI where the research sits

Answer client questions while the client is on the line

A client calls: how much is still outstanding with his debtors, what was the turnover so far, is there room for that investment? Normally you note the question, look it up and email the answer. With a connected assistant you ask the question in plain English and get an answer right away, with the underlying figures included. The AI answers based on the real entries in the records. For relationship management that is quiet profit: you sound better informed because you are.

Have the file prepared: reconciliations and outliers

Preparing a compilation file is a fixed series of checks. Do the VAT categories reconcile with the general ledger? Which suspense accounts still have a balance? How old are the outstanding debtors and creditors? Which general ledger accounts deviate noticeably this year from last year? That series can be worked through nicely as a conversation. You ask, the assistant supplies the deviations, and you decide what is a booking error, what becomes a question to the client and what is simply correct.

Run fixed quality checks across the records

The same approach works outside the annual accounts too: a fixed set of review questions per client, for example every quarter. Are there entries on the suspense account? Are there sales entries without a VAT code? This way you remove clutter from the records before the annual work begins, and you spread the investigative work across the year. To be fair though: this is a guided workflow per administration. You select a client, run the checks and move on to the next. It is not one press of a button across fifty client administrations at once.

One firm, three accounting packages

A firm rarely chooses the accounting package itself; the client brings it along. One client works in Exact Online, another in Twinfield and a third runs AFAS Profit. That is why the AI connector from iWebDevelopment exists for all three: Exact AI Connect for Exact Online, the Twinfield AI connector and the AFAS AI connector. The way of working is the same everywhere: connect, select an administration, ask questions in plain English. Your staff learn one method, regardless of the client's package.

Authorisation keeps applying per administration and per user. The AI only sees what the connected employee is allowed to see, nothing more.

Keep the roles clear

An AI assistant working on client data is a tool, not a colleague with signing authority. The assistant flags deviations and makes suggestions; whether something is genuinely wrong is for you to judge. The AI never carries out write actions, such as a correction entry, without being asked: there is always an explicit confirmation step beforehand. In addition, the firm can restrict write rights per role or set the connector to read-only entirely, with a daily limit as an extra brake. For an accountancy firm that is not a side issue but the core: the tool must fit into your quality system, not the other way around.

That quality system is on the agenda for 2026 anyway. The Further quality management regulations (NVKM) replace the old NVKS and follow the international ISQM approach: setting quality objectives, identifying and assessing quality risks, putting control measures in place and monitoring the system. Firms without a statutory audit licence, including the typical compilation practice, may continue working under the NVKS until 1 January 2027 at the latest through transitional law; 2026 is therefore the implementation year, and the NBA published a step-by-step plan for small firms in early 2026. If you are rebuilding your quality system this year, record right away how the firm uses AI: who works with it, which rights apply and how you handle the outcomes.

Start small and only scale up after evaluation

A firm-wide rollout on day one is not needed at all. Choose one or two client administrations, set the connector to read-only and let one experienced employee use it for file preparation. After a few weeks you evaluate: which review questions actually produced something, where was manual work still needed, and what does the employee think of it? If it works, you scale up to more administrations and more colleagues, and possibly to write actions with confirmation.

The trial period is set up with that in mind: 30 days free with all features, up to 3 administrations during the trial, after which you can cancel monthly. And for the supplier assessment in your file: the connector is ISO 27001 certified and independently pentested; financial data is only stored with the Analytics subscription, in a secure analysis environment within the EU that we delete as soon as you stop.

Sample prompts for the compilation practice

Copy these questions into your connected AI assistant and adapt them to the client in front of you. The trial balance question relates to a Twinfield administration.

Are there any entries on the suspense account in this administration?

Show the trial balance for period 6.

Which general ledger accounts deviate more than 25 percent from last year?

List the reconciliation checks for the file: VAT categories, suspense accounts, debtor/creditor ageing.

Want to try it yourself? Start for free with Exact AI Connect →