As far as weddings go, it was a pretty popular one. That goes without saying. But two billion people? For the literally hundreds of media outlets that parroted that estimate of the global TV audience for the William and Kate show, that goes without checking.

Everyone from the tabloid Daily Mirror to the Guardian to the buttoned-up Financial Times were members of the two billion club. The Mirror’s giddy take was typical: “The global audience is estimated to have been two billion – almost one in every three people on Earth,” it announced.

Many billion miles away, in plucky New Zealand, the bug had caught. “Prince William and Kate Middleton will exchange vows in front of an estimated two billion people this Friday,” gushed Stuff.co.nz. TVNZ was at it, too. So was TV3, who confidently asserted after the fact that “their wedding was watched by two billion people”. Even the Listener mentioned the figure – though at least was careful not to state it as bald fact.

Radio New Zealand upped the ante. “Estimates put the global television and online audience for the wedding coverage at 2.4 billion people,” it trumpeted, with all the abandon of a cartwheeling verger, and without any link to these “estimates”. (By the way, according to the Always Reliable Daily Mail, that acrobatic show-stealer has a New Zealand connection. Woohoo!)

Just a second, though. Two billion? Two billion? As one analyst has said, it’s surely absurd to imagine two in seven people in the world watching – “most people around the world were asleep. If you think about it for a moment, it’s nonsense.”

One commenter at Stuff was equally unconvinced: “So what genius in our midst has estimated the wedding audience at ‘two billion’? Did he/she use the back of a small, medium or large envelope to do his/her calculations?”

Well, the most recent source, or at least broadcaster, of the magic two billion number appears to be Jeremy Hunt, Britain’s culture secretary, a man who has become something of a mockery magnet in recent months, not least over his management of the Murdoch efforts to gain complete ownership of BSkyB.

“Hunt even cited this figure of 2bn to a meeting of the cabinet of the British government on 6 April this year.” That’s Nick Harris, of the excellent British website Sporting Intelligence. He continued: “If a government minister can spout such nonsense to others who help him run the country, you do wonder what else they’re getting wrong.”

Why is it nonsense? Why is the two billion figure “not even good guesswork”? Over to Harris (writing before the wedding):

It’s unsubstantiated garbage, as are so many numbers that have been bandied around for global TV audiences for ‘major’ occasions down the years … Many of the supposedly “most watched” events in history have been attributed viewing figures that were simply made up. And these numbers are passed down as fact, unquestioned.

It was said that 750m watched the 1981 wedding of Prince Charles and Lady Diana. It’s even in the Guinness Book of Records. And it’s wrong. If every single person in the world with access to TV in 1981 had watched the Charles-Diana wedding then it might have been something like that. But of course they didn’t. The true figure was probably closer to 100m people.

It was said 2.4bn or 2.5bn watched the funeral of Diana. They didn’t. It might have been about 10th of that figure …

How do we know this? Because there are authoritative and respected measurement bodies who collate this data. They include BARB in the UK and Nielsen in the USA, and TAM in India and equivalent bodies elsewhere.

Kevin Alavy, a director at Future Sports + Entertainment, an arm of leading international analysts Initiative, has spent years compiling accurate, verifiable data about major events from all the world’s major TV markets. In an industry awash with hype, Alavy’s work has been refreshingly bullshit-free.

“Figures like 750m for the wedding of Charles and Diana just aren’t based on any hard data at all,” he tells me. “And it’s completely and utterly ludicrous to suggest that William and Kate’s wedding will get anything like 2bn watching on TV.

Harris points to an article in the Sun newspaper ahead of the wedding that valiantly attempted to break down the viewer tally, which included an estimated “140 million viewers in the US”. That figure alone is fantasy, says Harris; it would mean outrating “the most-watched TV show in American history (the 2011 Super Bowl, which attracted 111m viewers in the USA), by 29m people”.

He was right. Nielsen analysis of Friday’s telly stats, reports the New York Times, estimates that an “average of 22.7 million viewers at home in the United States watched the wedding live”.

Harris concludes:

And even if we add up the numbers supposedly going to watch in the eight countries cited by the Sun, three of them among the top nine* in the world in population – the US*, Russia*, South Africa, Australia, Canada, France, Brazil* and Mali – then you get a figure of 275.59m people.

So where are the other 1.724bn watching from?

Answer: they’re not. And neither are the 275m going to materialise. It’s made up! Out of nothing!

But as Kevin Alavy tells me: “The figure of 2bn is out there now, somehow, and probably it’s going to be the one that’s accepted, and believed, even though it’s made up on the basis of no credible evidence whatsoever.”

The truth is that Big Numbers make journalists’ eyes glaze over. I know this because mine do, too. And once such a figure spreads like wedding confetti, there is certainly no shortage of apparent “sources” for the data – it’s just that they all are cut from the same unreliable sheet of paper. These numbers lose any meaning beyond being bloody massive – and this is precisely the sort of generalist media condition that makes bankers smile, probably. In this case, two billion proved not just too good to check, but too big.

(Update: I’ve just learned that RNZ’s Mediawatch programme ran an item on a similar theme, including an amusing snippet from the TV3 coverage of the wedding, in which the global viewership is estimated to be “heaps, mate”.)