You can turn category suggestions on and off, using the 3 dots at the top of the bank transactions page.

From the Bank rules page, toggle the “Match transactions to invoices and bills” setting off if you don’t want MYOB to automatically match transactions for you. 

No, category suggestions does not use generative AI.

Generative AI is the type that creates new content - like writing text, answering questions, or generating images - based on prompts and assumptions. Category suggestions uses predictive AI.

Predictive AI, which is focused on recognising patterns and making suggestions based on existing data. It looks at your past transaction matching behaviour (and if you're new, patterns from similar businesses) to suggest the most likely category for each bank transaction.

It doesn’t create anything new - it simply helps you make faster, more accurate decisions by surfacing suggestions that you can review and approve.

You stay in control the entire time.

Predictive learning tools take time to learn your patterns. The matches may not be perfect immediately – the more you approve correct suggested matches and edit incorrect matches, the smarter and more accurate it gets over time.

It’s great for regular transactions, but it’s still learning for less frequent things like transactions split across multiple account codes.

If you’re using AccountRight, you can toggle off auto-matching by rule and also auto-matching by invoices/bills.

If you’re using MYOB Business, you can toggle off auto-matching by invoices/bills only.

You can change the settings by clicking the three dots at the top of the bank transaction page.

Yes, automatically matched transactions are indicated by a black star icon . Manually matched transactions won’t have an icon. 

Only record matches can be automatched. This setting can be turned on and off in your banking settings. Go to Banking > Bank rules and deselect the option Automatically match bank transactions to MYOB bills and invoices.

AI category suggestions will never be automatched. 

Yes, you can unmatch by clicking the ‘x' that appears when you hover over a matched record in the ‘match’ column. Or expand the row to see the details and select 'unmatch’.  

If a transaction has been reconciled, you’ll need to undo the reconciliation before unmatching.

Yes,  if a category suggestion has not been approved it will show in purple and have a magnifying glass next to it, simply click anywhere in the box and choose an alternative category or expand the transaction to find other records. If a suggested match (black text) has already been approved select the ‘x' that appears when you hover over a matched record in the ‘match’ column to unmatch. Then select an alternative category or transaction record from the list. Alternatively, you can expand the transaction row, select 'unmatch’ and then select categories or other records from there. 

If a transaction has multiple suggested records that are correct, you can select multiple from the quick-select dropdown or you can expand the list.  

To split a transaction across multiple categories, expand the transaction row and use the ‘select categories’ tab to split the transaction across multiple categories. 

Yes, category suggestions will work with imported transactions. 

Your rules will take precedence over category suggestions.

If the rule is for a spend money or receive money. You won't see the suggested smart-matching category.  

If the rule is an invoice or bill rule, you won't see the category suggestion.

Yes, category suggestions will work with previous unreconciled transactions.  

 Yes, click the tick to accept the suggestion. Or, click the magnifying glass to search for the correct category or record. 

Smart-matching is being rolled out across all eligible files. If you don’t see it yet, it will be coming soon.

 Our goal is to take the heavy lifting off your plate - without taking away your control. Category suggestions uses AI to do the work of finding the match, so all you need to do is review the suggestion and then click accept. These changes are designed to improve usability, reduce errors and give you a better, more intuitive experience every time you log in.