This isn’t really a pilot, it was a run through with an audience in front of various international commercial commissioning big wigs – doing these with an audience is fairly unusual, the hope is someone sees something in the idea to order a fully fledged pilot. So there’s little point going on about the set (it was some screens and some podiums set up on stage at King’s College’s Greenwood Theatre) but we do have a good idea about the format and concept. When I first heard about it I suspected it was a spoiler for the upcoming Wild Rover Interact-o-quiz format (trying to create a successful Interact-o-quiz does seem to be some sort of holy grail with TV people), really this does try to be something different but falls into a lot of traps.

• Applause Store said get there by 2pm. We got in about 3:15pm. Those were priority tickets, but on the reservation screen it suggested getting there for 2:45pm. I’ve been standing around for an hour and a half before we got in, and you could have asked for people to turn up for 2:30 and it would have been fine. Christ you’re useless, and it puts everyone in a bad mood going into the recording.
• Anyway we finally get in and we’re given a sheet with the Kings College Wi-Fi and the link to the web app we’ll be using for the show. We’re going to get to vote on every question.
• The web-app by and large worked well (at least from the users perspective) – we were asked to log in as a guest but they were clearly aiming for social integration.
• Vernon Kay is our host. We like watching our Vernon do things like this, you never know if you’re going to get complete professional or stroppy teenager from one moment to the next. I know he’s not universally popular round here, but I think you’d come away with at least some grudging respect if you saw him work a studio in person.
• The titles are lots of blocks floating around the streets of Britain. It does feel a little bit like Come And Have A Go If You Think You’re Smart Enough. The blocks represent members of the public. That’s the show’s “thing”.
• Three teams of two will try to predict how Britain votes in a series of live opinion polls. As the show is meant to be live, the recording ran straight through as live, so we were done in about an hour-and-a-bit. The contestants are told the demographic make-up of the people playing, which should have shown up on the screen but it didn’t work. Apparently there were 300 people playing along in the audience and 400 playing online, although the “players” tally on the app never seemed to go above 380-ish.
• Each of the questions in the first rounds have four options. The questions are by and large a bit obvious – Who should be Prime Minister out of Russell Brand, Stephen Fry, Joanna Lumley and [forgot fourth option]? Which is rudest out of not holding a door open for someone, queue-jumping, speaking with your mouth full or talking loudly on a mobile? That sort of thing.
• The question asked – and on whizzy (and rather jerky) 3D graphics because they’ve got computers and they know how to use them, there’s a thirty second vote window. During this window the contestants must decide between them what they think the most popular answer will be and lock it in, and Vernon will read relevant facts out, ask the audience to shout out how they’re voting, or go to the “live” feeds of regulars  arguing at home in a clear nod to Gogglebox. It’s fair to suggest the audience were happy enough to discuss quite loudly, although I’m not quite convinced this translates to people deciding they’re going to watch it.
• Voting over the contestant’s guesses are revealed. The percentages are then revealed on the whizzy 3D graphics, sometimes one answer at once, sometimes all of them. The problem with not doing them all at once is that it’s quite easy to guess what the fourth percentage is once three have been revealed, so no real tension there. The reveals are done as falling blocks looking like a 3×3 Jenga tower, each block representing 1%.
• The other problem is that the back end of the software isn’t quite ready for prime time. Having said there was no tension, it was quite a surprise to discover that with three percentages of 40ish, 30ish and 10ish revealed, the final answer got a whopping 80ish percent. The software usually worked, but it wasn’t a unique occurance. Kay was pretty professional in ad-libbing his way round to be fair.
• There are five questions in round one. You score the percentages of the answers you pick as points. The lowest scoring team is eliminated. In the event of a tie, Britain gets to vote on which couple gets to stay (I don’t think they were expecting this, the whizzy graphic on the screen said “which couple should go?” but the question on the app said “which couple should STAY?” – they didn’t both revealing this with falling blocks.
• Round two is four questions. This time though after the percentages are revealed they get the option to gamble the points by putting their answer through a filter – today kids, OAPs, people with tattoos, people who eat hummus and mothers (IIRC), starting with the team that scored the lowest on the question. So if they score 20% in a poll of everybody, they might think that if you just asked the people who ate hummus that subsection would have voted a particular answer higher.
• It’s an interesting idea and brings a welcome layer of added strategy to the proceedings. But it’s also I largely suspect a bit complex of an idea for the average viewer (it’s a show that already features MATHS), and with the super whizzy graphics it now starts to look a bit like Quizzlestick.
• It wasn’t really used much in the runthrough, people were scoring 30-40 points per question and didn’t feel the need to risk, even though if you’re behind you probably should.
• It also really messes up the final reveal of the game – you have the blocks tumbling down and you have an exciting result except that’s not the exciting result because you’ve still got to ask about the filters. If there’s one thing this show needs to work on, it’s the reveals.
• I was flagging by about question six or seven, I certainly felt I’d seen everything the show had to offer by that point. I wonder how a thirty second vote window would work with several hundred thousand people playing the app, when you include the digital delay into the proceedings. Perhaps you could increase the window, but the show already feels rather stodgy, and not in the good jam rolypoly way. In the app it also looked like you could earn points. I hope predicting how Britain voted is different to actually voting.
• The winning team goes through to the final called the Exit Poll (nice). Five questions are asked with three answer choices (How do you like your eggs? Which hurts more out of stubbing your toe, a paper cut or getting sand in your eye?) and they have to answer the first four. Their score is converted into cash and multplied by ten – 300 points = £3,000. Each of the first four questions they get right doubles the pot.
• The final question is a gamble. If they decide to go for it, if they pick the correct answer they redouble (five correct answers will be worth around £100,000), if they pick the middle answer they take away what they earn going into the question, if they pick the least popular answer they leave with nothing. These terms seem quite generous, I don’t get why it’s being offered as a gamble, it should be just how the last question works especially with no such penalties in the earlier questions.
• If you were to base it on loudness of audience reaction to questions it would be a hit. If you base it on my gut feeling of “I’m not sure people not captive will actually tune in to discover how Britain likes their eggs” it probably won’t.

And then I bumped into TV’s Michelle Martin and we went down the pub and had an interesting discussion on how difficult it is to get interactive quizzes to work on TV. Apparently it’s also surprisingly hard to find a pub in the Borough area, we discovered.

Here’s a warning from history:

The graphics are less whizzy, the prizes much less big, but it’s not far off the modern reality to be honest.