The way the UK reports unemployment may not reflect the "true scale of joblessness", says a think tank.
Unemployment rose by 34,000 in April to reach 1.3 million, according to the Office for National Statistics (ONS).
But the Resolution Foundation argues that the 23% drop in average hours worked between early March and late April is a better indicator of unemployment.
The ONS said it publishes a large selection of analysis on employment.
Official numbers on how many people are out of work and claiming unemployment benefits will be published on Thursday.
Resolution Foundation chief economist Mike Brewer said: "Britain is in the midst of an unprecedented economic shock that is profoundly affecting millions of people's jobs.
"Unemployment is forecast to hit 4 million for the first time ever. And yet our official data is failing to show the true extent of this jobs crisis."
On Tuesday, the government's budgetary watchdog, the Office for Budgetary Responsibility (OBR), projected that unemployment could reach 4 million people, if the UK's economic recovery is poor, up from 1.3 million in 2019 in its latest analysis.
Meanwhile, data for people claiming benefits soared to 2.3 million for April.
But these figures could include some people who are eligible to claim support while still employed. The ONS said: "Enhancements to Universal Credit, as part of the UK government's response to the coronavirus, mean that an increasing number of people became eligible for unemployment-related benefit support, although still employed."
The Resolution Foundation says this data is not a good reflection of the true picture either because it includes furloughed workers who initially made a claim when the crisis first struck.
The think tank says it estimates that "fewer than half (700,000) of the 1.6 million increase in the claimant count between March and May is related to people who are newly out-of-work, and not receiving furlough pay or self-employed grants from the government".
It urges the ONS to make more of its ability to count the number of workers who are employed and not temporarily without work, alongside the headline employment rate, as this would provide "a far more accurate picture of labour market activity".
The ONS said it agreed that data on hours worked was an important component in understanding the unemployment picture in the UK.
"However, our detailed Labour Force Survey estimates are based on interviews with tens of thousands of people and provide vital detail not available from any other source," it said in a statement.
"It is difficult to interpret claimant count figures, as we know these include some people in work."
"Measuring the labour market, like many areas of the economy, presents additional challenges at the moment, especially with so many people furloughed. That is why the ONS has introduced new surveys and data sources to provide the best possible indicators of the impact of the pandemic. "
Separately, the British Chambers of Commerce is warning that almost a third of UK businesses (28%) they surveyed expect to cut jobs in the next three months.
The figure compares to last year, when only 7% expected to do so.
Some 7,400 firms took part in the BCC survey, which found more than a quarter of the firms (28%) said they had already shrunk their workforces since the pandemic began.