Mapping trial balance lines: suggestions and approvals
Where Achekam's mapping suggestions for a client's trial balance come from, and the point at which a suggestion becomes an approved mapping.
By Achekam · · 4 min read · For Accountants & audit firms in South Africa, Nigeria, Kenya
Mapping is the unglamorous step that decides whether a set of statements is right. Each account in a client's trial balance has to land on the correct statement line and note: "Sundry creditors" under trade and other payables, "Loan account - director" under related-party borrowings, "Suspense" nowhere until somebody finds out what it is. Get one wrong and every figure downstream is wrong in a way that adds up perfectly.
Software can propose most mappings. The control question is the one this article answers: where do the proposals come from, and at what point does a proposal become a mapping the firm stands behind?
Four sources of suggestions
Once a trial balance has been read into Achekam from an Excel workbook, a CSV file or a PDF with a text layer, Achekam suggests where each line belongs. The suggestions come from four sources:
- The file's own mapping column. If the client's export or the firm's working paper already carries a mapping column, Achekam reads it.
- Earlier mappings. Decisions made for this client before are offered again.
- AI. Where an AI key is set up, AI proposes a line for accounts it can interpret.
- Keyword rules. Without an AI key, keyword rules are used, and the suggestions are labelled as keyword-based so the reviewer knows what produced them.
Each source has its own weakness. A client's mapping column may reflect their bookkeeper's view, not the framework's. Earlier mappings repeat last year's mistakes as faithfully as last year's good calls. AI can be confidently wrong about an account with a vague name. Keyword rules match words, not meaning, so "Loan from director" and "Loan to director" can look alike.
Nothing approved until a person approves it
No suggestion becomes an approved mapping until a person approves it. You will find the trial balance and its suggestions under FINANCIAL INTAKE > Trial Balances, then the trial balance in question.
Approval is where judgement goes. Focus it where it matters:
- accounts with large or unusual balances;
- accounts whose names do not say what they are, such as suspense, sundry or clearing accounts;
- loan accounts, where the direction and the counterparty change the line;
- any account the client opened during the year.
AI proposals placed in the Studio draft
There is one place where a suggestion shows up before approval. When the Financials Studio builds a draft, AI mapping suggestions with a confidence of 0.5 or more go into the draft at once, and are listed as proposals a person can accept, move or set aside.
That helps a team see a near-complete draft early. It also means the first draft is not a picture of approved mappings. No rule was found that stops signing while AI proposals are still open; the safeguards are the review stages and the partner's signature. So working through the proposals list is part of review, not an optional tidy-up.
Moving accounts by hand in the workbench
Staff can see and change where each account sits in the Studio's workbench, under Canvas and then "Mapping". It shows each account's line and note and lets staff move accounts by hand.
One navigation point saves confusion: the menu entry "Trial Balance & Source Mapping" opens the Source Room, not this workbench. If a team member says they cannot find where to move an account, that is usually why.
Moving an account by hand is the right tool for a reclassification the team has decided on, such as moving a shareholder's loan from current to non-current after reading the loan agreement. Record the reason on the file, so the next reviewer does not move it back. Classification decisions like this one are presentation questions, and the IFRS Foundation's summary of IAS 1 covers that standard's requirements for the presentation, structure and minimum content of financial statements.
Earlier decisions remembered next time
When a firm brings a client's ledger into Achekam, from the client's exported CSV or Excel files or from the client's own books on Achekam, mapping the migrated ledger draws on what the firm has already decided. Achekam remembers earlier mapping decisions and asks only about new or changed accounts. Where it has guessed an account's class, the guess is labelled as such.
That turns mapping in the second year into a review of changes rather than a full repeat. It also makes the first year's mapping more valuable, because those decisions will be offered again.
| Situation | Where the suggestion comes from | What the reviewer checks |
|---|---|---|
| First year, client export has a mapping column | The file's column | Whether the client's view fits the framework |
| Second year | Earlier mappings | Whether anything about the account changed |
| Unclear account names, AI key set up | AI | The real nature of the account |
| No AI key | Keyword rules, labelled | Accounts where similar words mean different things |
| Account class guessed during migration | Labelled guess | The class, from the ledger detail |
A mapping review routine
- Sort the trial balance by size and approve the large balances first.
- Investigate every suspense, sundry and clearing account before mapping it.
- Check loan accounts for direction, counterparty and term.
- Clear the AI proposals list in the Studio draft.
- Record the reason for every hand move.
- In later years, focus on the new and changed accounts Achekam asks about.
Getting the mapping right once
A well-reviewed first-year mapping pays off every year after, because it becomes the earlier decision Achekam offers next time. The file the mapping starts from is covered in uploading a client's trial balance: what is read, and building it from a ledger export in turning a general ledger export into a trial balance.
General guidance only, not advice on your own circumstances. Tax and payroll rules change: check the current position with the authority or your adviser.
