Truckie took bathers to the taxman for a deduction but found a cold plunge

Australian Taxation Office (ATO) assistant commissioner Rob Thomson warned of strict criteria to meet deductions and scrutiny of work-related expenses this year.

"While a lunchtime dip might clear your head for work, swimwear for a truck driver is clearly not deductible," he said.

Another eyebrow-raising case was a mechanic who tried to claim an air fryer, as well as other claims for a microwave, vacuum cleaners, TV, gaming console and accessories.

Those claims were denied as "personal" in nature.

Mr Thomson said work-related expenses must be supported with receipts or invoices.

"If your deductions don't pass the 'pub test' it's highly unlikely your claim would meet the ATO's strict criteria," he said.

There are two ways to claim working from home expenses: the fixed rate method, which allows you to claim 70 cents for every hour you work from home and covers your additional running expenses and the actual cost method.

H&R Block director of tax communications Max Chapman said the ATO was laser-focused on work-from-home claims this year.

"If you can't provide your hours or expenses, don't expect them to go unnoticed," he said.

"We're urging taxpayers to be crystal clear on what's claimable to keep watertight records."

He said the ATO's data-matching systems were more advanced than ever.

"The most common mistake we see is people claiming $300 without receipts, assuming it's a free pass," he said.

"It's not. You must actually have incurred the expense."

READ MORE: How much you'll get from Labor's surprise tax cuts

There were also concerns about people "double dipping", Mr Chapman said. That involved, for example, taxpayers claiming the whole or a substantial part of a phone or internet bill as work-related and claiming mobile costs separately.

Another focus was on people who sought deductions for investment properties and holiday homes, including excessive interest expense claims.

In addition, the ATO was now receiving reports from many platforms in the sharing economy (including Uber) that it could use to reveal data mismatches.