I’m writing this blog post in my lunch break, having spent the last 5 hours with a script on my computer refreshing a page every 5 minutes trying to get a 26-30 railcard, 10,000 of which went on sale this morning as the first stage of the national trial (another 10,000 are due to go on sale at a later date).
Unfortunately, the implementation of this has been anything but smooth. This is what I’ve seen on each of the what must be 100+ screen refreshes I’ve performed on the first page of their ordering system:
It’s clearly not just me – rather than a railcard order form, 503 errors have been the experience for most, judging by the uproar on Twitter and elsewhere (#railcard has been trending on Twitter for most of today). The Railcards Twitter feed acknowledges the issue, but says their site is ‘performing better’ in their most recent Tweet:
Unfortunately, it’s not, for me at least – as I type, I’m still getting a 503, and have now given up for the day realistically.
What frustrates me most about this is that it was an entirely avoidable problem. Rail Delivery Group, who manage the railcards scheme, must have foreseen the demand, especially when there was only going to be enough railcards available for 1/500 of the eligible population in this first trial release, and there are several technical solutions available for managing this demand.
Companies such as queue-it offer systems that RDG could have put in place to manage the amount of traffic reaching their servers at any one time. This is the solution used by London Transport Museum to manage capacity on launch days for their immensely popular (and excellent!) Hidden London tour tickets – it’s also easy to implement, the traffic is managed entirely outside of RDG’s servers until each user reaches the front of the ‘queue’, on what is essentially a one-in-one-out basis.
The other question is – why limit the trial at all? There’s already been a postcode-specific trial and up until recently it was publicised that a full national launch would be next – why not go the whole hog and simply launch the railcard fully – or alternatively offer, for example, a 2 week period where anyone could get one, if they really did want to go down the trial route?
My hope is that RDG now does go down that second route and open it up to everyone for a longer period of time. A majority of the age group targeted are in full-time work, so it’s unfair to expect them to endure the luck of the draw of a Glastonbury-ticket-esque scrum for a railcard. I won’t be the only one that’s now given up (at least until this evening). Unless they sort their systems out or go for the open trial option, it’s only going to repeat itself when the next batch of 10,000 are released too, generating yet more bad publicity.
UPDATE 18:45pm: Railcards have now sold out, with the following generic statement Tweeted out:
26-30 Railcards are now sold out. We apologise to those who had difficulties buying one today. Keep an eye on our social accounts or register your interest here for further updates: https://t.co/2OZv59WEUS
— National Railcards (@_Railcards) March 13, 2018
I plan on writing to Rail Delivery Group in the next couple of days, copying in the railway-focused press. I’d be interested in reading your experiences of trying to get a railcard in the comments below.