The sales funnel for ordering tickets from the official website looks roughly like this:

  1. Search for tickets. Selecting "only available" actually means "show only those that we don't know are unavailable".
  2. Click on an event. Find out that
     a) the tickets are actually not available at all ("tickets  unavailable" or
     b) only the most expensive tickets are still available (£200+)
     c) there are still tickets you can afford
  3. Add tickets to your shopping list
  4. Proceed to check out and payment

What's worse, the shopping list really is a shopping basket, not a wishlist. You cannot just add tickets to all the events you would potentially like to go to and then only pay for one pair of those that are available. If you do that, and in step 4, it turns out that of the 20 tickets you selected, 8 are available, you can only proceed by paying for all 8 tickets or cancel altogether (and probably lose the chance to buy tickets, as they are released into the pool again).

This means that in practice, you keep going through all the steps, adding tickets to your shopping list, proceed to check out, find out they are unavailable, go back to shopping list, remove unavailable tickets, rinse and repeat. It's terrible.

I finally did end up buying tickets, by the way, and I do want to point out some factual errors in the article:

  1. There are email alerts for newly released tickets. They are in the newsletter that you can opt-in for when you create an account on the ticketing website.
  2. While there is no clearly marked option "search all available tickets for all sports on all dates in all venues", there are other ways to search than just by sport. You have to specify one of sport, venue or date, which leaves:
    a) searching by venue or group of venues. Most of the sports are in the "Olympic park venues" or the "London venues", so searching by venue group is a LOT more convenient (see http://cl.ly/image/213p2f0r022p).
    b) searching by date. You can simply select the first day of the Olympics and the last day and it will search all sports on all dates in all venues.

- The Olympic committees often have final say over design, but a lot of effort is invested. We all know the phrase of something looking like it was "designed by committee." Well, that's pretty much what happens here. You have a lot of individual people with great ideas, but many of they diametrically opposed, not to mention multiple levels of approval. Approval comes from multiple layers of the of the Olympic committees & the committees are often political appointees. They may have little-to-no experience with design (or ticketing), but may still offer "tweaks" and the like. Everyone has the best intentions, but, well...

- CAPTCHAs suck, but it or something similar is needed. Scalpers pound the site. I know fines have been raised, but all that really did was drive it underground, away from the everyday person. (Not to mention resulted in lots of empty seats.) Without it, almost every ticket would end up in the hands of scalpers.

- To that end, most of his "This is bad" section doesn't apply. The vast majority of events sell out. The empty seats aren't because tickets didn't sell, it's because they're sold & people didn't show. I hope the 2016 Olympics learns from this & allows them be re-sold.

- In 2008, we had a simple grid to show what events were & were not still available. I'm not sure why that was dropped. I disagree with only showing events that still have inventory. A common use case is people have their tickets, but want to hop online as the event nears to get more details. Having that data readily available is a good thing.

A lot of it is typical blog snark, so I may have missed some of other valid points.

However, the last thing I'd add is the that the technology behind ticketing is hard. Prior to Google, I'd worked at Yahoo. Those are the only two companies I know of that have both the technology & talent to build a credible ticketing system. Most people who people who tell me they could start a company that could crush Ticketmaster's technology and even have thoughts about the MySQL schema make me chuckle. They often just stare blankly when it's pointed out that tickets are not a fungile resource, not realizing what that means or scaling challenges it presents. And that's just the tip of the iceberg of the problems.

> London's Metropolitan Police said they had arrested 16 people since Friday for illegal reselling of Olympics tickets

And then everyone is shocked by the empty seats.

http://blogs.telegraph.co.uk/finance/timworstall/100019212/o...

(Ticketmaster = confusion and stumbling, we are used to that in UK.)

1) Searching by event makes much more sense than 'show me what's available'. It goes some way to help the tickets (mainly) go to the people who specifically want to see _that_ event.

At a guess, you just wanted to see a random event (and that's fine) and the UX sucked for you. But if you were dying to see the Fencing, you might be glad that Fencing tickets were 'randomly' offered to people who just wanted to see anything :).

2) Showing events that are unavailable, again, this makes more sense than the opposite. If the search didn't show these unavailable events, they would undoubtably be inundated with telephone calls from people asking to buy tickets to these events.

The site is getting a few hundred thousands users a day right now trying to ger a few thousand tickets mostly between 7pm and 10pm.

While I agree LOGOC could of got a lot of the copy and menus better, trying to display "real-time" availability and keeping ticket sales fair is very very hard and Ticketmaster are doing a fairly decent job.

The 2010 FIFA World Cup site was shoddier than the Olympics, with rapid time outs, and other instabilities. They also lacked any way of keeping bots out, and I suspect a few people made money by blasting the site with automated ticket requests.

Agree with the rest of your comments. The "currently unavailable" and "these are not currently on sale" are particularly disappointing.

Of course, the Olympics is a far more complex set of events, but I'd have to imagine it would have been easy to implement this year (6 years later).

Fairly allocating a severely undersupplied stock of tickets without allowing the price mechanism to arbitrate is very hard.

Thinking it's a technology problem, when it's an economics problem, doesn't make it any easier.

I've built a system for the planning and seating of VIP hospitality packages that has been used for Beijing & London as well as the last Rugby World cup. Admittedly my ticket app is only for a select group of clients & users, so I haven't needed to scale anywhere near to the same degree as LOCOG, but I have done so with other sites in the past and the scaling techniques are the same.

That said, with the benefit of hindsight, I think the ideal way to have done ticket resale, would have been for people to register interest in events, possibly taking payment information in advance. Any inventory released back to LOCOG could have then been randomly offered to people who'd registered interest. There would be some element of first-in first served, but you're always going to get that with oversubscribed events.

The problem would be a lot easier to solve if you change the culture of releasing all your tickets at 9am, inviting hundreds of thousands of people to hammer your site all at once, at that specific time.

Not fun can be an understatement.