Assigning accounting attributes such as Suppliers, Expense Accounts, VAT Rates, Cost Centres, Cost Carriers and other dimensions to transactions, invoices and reimbursements follows different rules, patterns and practices in every organisation.
AI Pre-Accounting automatically pre-codes the majority of your transactions, invoices — and business expenses in the Reimbursements module — for you.
Here is how we do it:
Learning from your data
AI Pre-Accounting reviews the accounting attributes you assign when exporting expenses. Those historical mappings are the foundation it uses to learn how your organisation codes spend.
Building an AI model for your organisation 🤖
Once you've exported your first 150 transactions, 25 invoices or 25 business expenses (Reimbursements), we automatically train an organisation- and spend-type-specific AI model using your data. The model analyses how Expense Accounts, VAT rates and other dimensions were assigned previously and uses those patterns to predict the correct accounting details for future items.
Prediction with AI 🤖
For all future transactions, invoices and reimbursements the model attempts to predict VAT, Expense Account and other dimensions. Predictions are probability-based: when confidence for a single value falls below the minimum threshold, the field is left empty rather than guessed, reducing the risk of incorrect matches.
When there are multiple higher-probability options, AI Pre-Accounting surfaces up to five suggestions in the dropdowns, pinned to the top so accountants can quickly choose the correct value without browsing the full list. If no suggestion meets the minimum confidence threshold, the dropdown will not show suggestions and fields remain empty for manual input.
Our internal data shows AI Pre-Accounting assigns correct accounting attributes more frequently than static rule sets, so we recommend enabling the "AI Pre-Accounting overrides rules" option in Settings so the top prediction(s) take priority. The default is rule-first, so rules remain authoritative until you opt in.
Self-learning with AI 🤖:
The Agent continuously improves through a feedback loop. If it makes an incorrect prediction, you can correct it and the Agent will learn from that correction. Each time you export transactions, invoices or reimbursements, we (re)build an updated model that incorporates your previous corrections and the newest exported data.
If you delete a specific Expense Account, VAT rate or other value, we rebuild the model so those values are no longer predicted.
Which spend types are supported and which fields can be predicted
Card transactions
Fields that can be predicted: Expense Account, VAT Rate, Cost Center, Cost Carrier, Custom Dimensions (if configured), Supplier
Invoices
Fields that can be predicted: Expense Account, VAT Rate, Cost Center, Cost Carrier, Custom Dimensions (if configured), Booking Text
Reimbursements (business expenses only)
Fields that can be predicted: Expense Account, VAT Rate, Cost Center, Cost Carrier, Custom Dimensions (if configured)
Low-confidence predictions
When AI Pre-Accounting is less certain about a prediction, the field is highlighted in yellow. This means the value has been pre-filled but falls below the high-confidence threshold — you should review it before confirming.
Yellow predictions behave like any other pre-filled value: you can accept them as-is, edit them, or clear them. If you correct a yellow prediction, AI Pre-Accounting learns from that correction and will improve over time.
If confidence is too low even for a yellow prediction, the field is left empty and suggestions are shown in the dropdown instead.
When the Agent can’t predict your data entries (and what to do)
New or rarely used Expense Accounts or VAT rates
AI Pre-Accounting needs examples to learn. If a value appears very rarely in your data (for example fewer than ~10 transactions), it's unlikely to be predicted reliably — and it may be ignored.
Identical items linked to different accounts
If identical transactions or invoices are routinely coded to different accounts (e.g. the same supplier is sometimes charged to different Cost Centres), AI Pre-Accounting will struggle to infer a clear pattern and will tend to pick the most common value.
Note: AI Pre-Accounting models are (re-)trained using data that has been exported to your accounting system. Assigning accounting attributes without exporting the transaction, invoice or reimbursement will not improve future predictions. To get the best results, export consistently and keep your exported data accurate.
AI Line-Item Extraction (Early Access)
Note: This feature is currently in beta and not available to all customers. If you're interested in joining, please reach out to your Customer Success Manager.
What it does
For beta customers, a new AI system works alongside the predictions described above. Instead of predicting accounting attributes for the expense as a whole, it reads the invoice document directly, extracts individual line items, groups by VAT when possible and pre-fills the split table with accounting data — ready for review when you open the expense.
This happens automatically in the background as soon as an invoice is uploaded. While processing is in progress, you'll see a loading state on the accounting fields.
How it works
The AI processes invoices in two steps:
Extraction — the AI reads the invoice and identifies individual line items: descriptions, amounts, VAT rates, and other data directly from the document.
Accounting — a second AI step takes the extracted lines and maps them to your organisation's chart of accounts, assigning the correct expense accounts and VAT rates based on your accounting history.
Auto-apply
By default, the extracted line-item splits are applied automatically when the invoice is uploaded. If you prefer to review the invoice before any splits appear, you can switch this off:
Go to Settings → Automation → AI Pre-Accounting
Turn off the auto-apply toggle
When turned off, the AI still extracts line items in the background, but the results are only applied when you click the split button.
What the new AI covers
The following fields are pre-filled per line item:
Amounts (gross, net, VAT)
VAT rate
Expense account
Booking text
What's still handled by existing predictions and rules
Cost center, cost carrier, and custom dimensions are not yet covered by the new AI. These fields continue to be populated by ML predictions and supplier rules, exactly as described in the sections above.
If your organisation has pre-defined supplier splits (custom split rules configured for specific suppliers), those take priority over the AI-generated splits.
Tips
Check the first few invoices from each supplier to confirm the AI picks the right expense accounts. Accuracy improves as the system learns from corrections.
Use merge and bulk edit if the AI extracts more lines than you need — select the lines and merge them into fewer booking lines. Learn more about bulk actions
Feedback is welcome — let your Customer Success Manager know what works and what doesn't.