Alex Harries's diary

I've written - no, ranted - before about how embarrassingly poor the Alliance and Leicester website is, when their badly-designed personal banking website managed to cost me a lot of money in penalty charges.

Checking my balance tonight (ok, I'm waiting for my student loan to come in so I can pay off some debts and, err, maybe spend some of it on a posh widescreen telly. Maybe...) I noticed an advert-style link at the bottom of the banking website. Apart from anything else, this position is one of the worst places to tuck important information as advert blindness tends to set in for anything placed in this part of a web page.

I'm a cynical git by nature - when I'm not off upsetting people, being embarrasingly drunk, or just being plain embarrassing - and decided to see what wonderous information the A&L were going to provide its customers that one can't already find out by reading any number of existing - and respected - websites (for example here, here, here, here ... etc).

The first thing to annoy me when I clicked the link was that it opened a new window. Not a major issue for most people, maybe, but I consider it bad manners to start popping up new windows on your visitors' computers unless you have a very good reason to do so.

The second annoyance - and this is the point where I decided I had to write a ranty blog entry about this - was the completely gratuitous 2-step entry process - first I was shown a picture of a safe, with some text fading in and finally a "[click here to] enter" link appearing in the middle of the safe's combination dial thingy:

... followed by an equally gratuitous animation of the safe opening:

... and after a few seconds, you're presented with this completely Flash-based website:

Very pretty, lovely... Just one question - why has someone gone to all the trouble to pay - probably a lot of money - to have a Flash website developed to do this, when a plain, boring - but just as pretty - HTML website would have done *exactly* the same job, been a lot cheaper to develop and bug-test, loaded faster, provided a much higher level of accessibility and user friendliness, and would (eventually) have found it's way into the Google indexes??

However, I ought to balance this argument out a little - the information in the site isn't completely crap, although it's extremely basic and does little to help people make use of the myriad free anti-virus and anti-spyware tools available on the net. Oh, and there's also a link to the security centre on the A&L online banking log-in page:

(More web geekery...)

I occasionally visit Silktide.com's site score analyser to see how UPSU.net's looking to the rest of the world. Last night I popped on there and had a miniature panic attack when it said "Your website appears to be visited so rarely that there is not accurate popularity information available - i.e. it is not in the top 10 million or so websites...", giving us a marketing score of 4.5 out of 10.

(I notice that our traffic is hovering around the 2,000 - 2,500 visits-per-day level (not including search engine spiders, apparently), which is definitely not too sad, but "could do better" I guess).

After all the work we put into the site, that's quite upsetting, really, so I was chuffed to see what it had to say about the BBC's homepage: "Your website appears to be visited so rarely that there is not accurate popularity information available - i.e. it is not in the top 10 million or so websites..." (4.6 out of 10).

Perhaps there's something not quite right there then...

On the plus side, without wanting to blow my employer's trumpet (stop laughing!), our scores for "How satisfying your website is likely to be" (based on variety of content, etc.) is 9.9 out of a possible 10 - the BBC "only" scores 9.8, and our "how well built and designed" score is also 9.9 vs. the BBC's 9.5 - cooool...

Of course, these great numbers mean nothing unless we're actually doing what it is that we're here to do - to keep Portsmouth's students informed. Do you think we're doing alright, or do you think we could do better? And how? Comments box, below, is open for everyone to comment in - please feel free to have your say! :o) 

You can compare the two reports here (BBC) and here (UPSU). Or not - whatever... :o)

Little update: I've just tested a site a did a couple of years ago for Gearboxman.co.uk - a gearbox company oop Norf(ish) - in Silktide and the results are really impressive (from my perspective - I was involved in the design, not the marketing of it).

Given that back then I knew comparatively little about web design, I'm really happy with the results - even the DDA requirements are good, with design, accessibility and satisfaction levels coming in at 9/10, 9.3/10 and 10/10 respectively.

The best bit about all this is that the site is the number one hit on Google.co.uk and Google.com for "racing gearboxes" - not too shabby at all.

(Caution: shameless geekiness ahead): These are, apparently, the BBC's 15 Web Principles which underpin everything they do online. The list comes fresh-ish from the BBC via. TomSki.com, a personal blog written by a member of the BBC. They make a lot of sense (well, to pixel-pushing geeks like me, anyway...):

1. Build web products that meet audience needs: anticipate needs not yet fully articulated by audiences, then meet them with products that set new standards. (nicked from Google)
2. The very best websites do one thing really, really well: do less, but execute perfectly. (again, nicked from Google, with a tip of the hat to Jason Fried)
3. Do not attempt to do everything yourselves: link to other high-quality sites instead. Your users will thank you. Use other people's content and tools to enhance your site, and vice versa.
4. Fall forward, fast: make many small bets, iterate wildly, back successes, kill failures, fast.
5. Treat the entire web as a creative canvas: don't restrict your creativity to your own site.
6. The web is a conversation. Join in: Adopt a relaxed, conversational tone. Admit your mistakes.
7. Any website is only as good as its worst page: Ensure best practice editorial processes are adopted and adhered to.
8. Make sure all your content can be linked to, forever.
9. Remember your granny won't ever use “Second Life”: She may come online soon, with very different needs from early-adopters.
10. Maximise routes to content: Develop as many aggregations of content about people, places, topics, channels, networks & time as possible. Optimise your site to rank high in Google.
11. Consistent design and navigation needn't mean one-size-fits-all: Users should always know they're on one of your websites, even if they all look very different. Most importantly of all, they know they won't ever get lost.
12. Accessibility is not an optional extra: Sites designed that way from the ground up work better for all users
13. Let people paste your content on the walls of their virtual homes: Encourage users to take nuggets of content away with them, with links back to your site
14. Link to discussions on the web, don't host them: Only host web-based discussions where there is a clear rationale
    
15. Personalisation should be unobtrusive, elegant and transparent: After all, it's your users' data. Best respect it.

p.s. Re. copyright: the list (probably) belongs to the BBC, while the content is lifted from this page on www.tomski.com

(Kind-of off-topic in a pixel-orientated way, not to mention long-winded, rambling and - arguably - pointless ;o).

Checking my bank account today to see if a transfer I'd arranged between a couple of my A&L accounts had cleared, I was more than a little confused to spot no sign of the transfer, and a mysteriously overdrawn third account.

Looking further into this, it turned out that my transfer - which should have gone from account A to B, had in fact been set up to go from account C (which didn't have enough money in) to B. Naturally, the transfer failed and account C had a £34 charge applied for a failed payment.

Bugger!

By making another test payment, I figured out what had happened: the A&L "my accounts" homepage has a quick-pay form on the left of the page. By selecting an account to transfer money from, and then moving focus away from the select box, a bit of Ajax pulls in a list of the other accounts you can transfer money to into the select box, below. For some reason though, focus isn't then sent to the second box, so you have to tab about a bit to choose which account you want to send the money to.

Somehow, then, I'd managed to select the right account, move away from the "move money from" box, then instead of tabbing over to the "move money to" box, I'd gone back to the first box and changed the account number, which cocked the transfer up more than a little.

Worse still, when you confirm the amount you want to transfer, you're presented with a screen confirming your choice which displays the account names and numbers on top of each other (thanks to a DIV clearing problem) in Firefox, so it's only possible to work out where your money's going from and to by selecting and copying the text into a text document, for example. So, thanks to this, I didn't spot that I was trying to move money from an account without any in it, which at least explains the penalty charge.

Eventually finding a phone number for the A&L and, having spent ten minutes lost in a seemingly endless array of numerical options, finding out my balance, pin number hint, date of birth, and bra size, and divulging what felt like all of my personal details via. a telephone keypad, I made it through to Sheila*. Sheila was a particularly ratty-sounding Leeds woman who, I can only presume, took an immediate disliking to the way I pronounced my name (maybe it was the Southern accent?) and told me - flatly - that the fault was all my own and that she categorically wouldn't lift the penalty charge.

Uhhm, hang on a minute...

I then manage to get through to Gheeta** in Internet banking and, explaining the problem again, Gheeta told me she would lift the charges "only this once" - hooray! - and would I mind explaining the problem with the website to their technical team - no problem.

So I'm then transferred to the technical support team and spoke to Cletus** who admitted they knew about the problem, but that it had only been found in Firefox 1.5. My new best mate Cletus didn't seem to know if they were going to fix it, and wasn't even aware that it could be fixed ("it's a DIV float clearing problem - honestly, a quick Google will turn up a way of fixing it" quoth I, maybe unintentionally coming across as a bit condescending).

After all that, I'm still waiting (two hours later) for the refund to come back into my account, and Cletus, Gheeta and Sheila are probably off home to drink Ovaltine. I've learnt that I'm no better at remembering names than I ever was, and that there are - it seems - better-paid web designers who don't really seem to know or care about cross-browser continuity.

I think I might stick to high street banking in future... ;o)

* name changed to protect t...