Webcredible vs Better connected
…also known as Better Connected Critique Volume XXVI in a series of … um … I don’t think Roman numerals go that high.
Webcredible have been reviewing the ‘leading’ UK local council websites for usability — or at least those 20 sites which have been identified as the leading top 20 websites by the 2008 Better Connected report.
Webcredible defined 20 ‘usability criteria’ (such as ‘is there a clearly marked home link on every page?’ or ‘is it easy to find your local councillor?’) and awarded between 0 and 5 points to each depending on how well they did, giving each site a usability score somewhere between 0 and 100.
Obviously, a number of these criteria scores will be subjective, depending upon the reviewer(s). The more reviewers looking at any individual site, the less likely you are to find any obvious discrepancies. Unfortunately, there’s no indication of who was carrying out the testing, nor how many reviewers tested each site, so on that basis it’s difficult to see whether the methodology is really any better than Better Connected — but the emphasis is certainly different.
What the report tells us is that the usability (at least according to Webcredible’s measure) is improving:
Compared to last year’s results, this year there’s a promising trend towards improvement across the sector. Last year’s average score of 45.5 has increased to 56.6 this year.Local Council Websites: Good But No Cigar (Webcredible)
…good news for those Local Authority bods who have been trying to fend of criticism that their websites are getting worse every year.
What the Webcredible report does do exceedingly well is to provide nice, clear examples of what they consider to be good and bad practice. Now there’s one part of me thinks that it’s a little rude to single out particular sites for criticism, but on the plus side you can learn just as much from what not to do. Apologies if you happen to be one of those singled out…
And then again, I’m not always going to agree with Webcredible anyway…
It’s a fact that many users don’t know how to change the text size of a page using their browser controls. A good solution to this problem is to provide text resizing controls on the actual web page itself.Local Council Websites: Good But No Cigar (Webcredible)
So Council sites which provide text resizing controls will score better than ones which don’t. However, a Council which educates their users how to adjust their own browser text size settings is actually of more benefit to the users — because this will enable them to change their text size on any site they come across, rather than just on that particular council site.
Having said that, if you are going to provide text sizing controls, it is a good idea to make them visible and prominent — otherwise what’s the point?
Webcredible’s section of “homepage priorities” also asks the questions “are to A-Z of services explicitly above the fold” and “are key tasks listed above the fold?”. Two questions are jumping into my head here — firstly, where is the fold? Surely this depends on screen resolution and/or text size? If we’re going to be testing things and judging them on this basis, we need to be a lot clearer about the methodology. Secondly, who decides what the council “key tasks” are? Once a year, school admissions will be a “key task” for many residents in the borough, but outside of the standard application time, this isn’t going to be so important…
The A-Z of services should be promoted high up on the page to provide a quick route to support users’ information needs. By spelling out the full A-Z list users can focus their efforts immediately through their first clickLocal Council Websites: Good But No Cigar (Webcredible)
One of the key points that they are making here is that if the users get direct access to the A to Z on each page, rather than needing to first follow an “A to Z of Services” link, they are making it easier for the user. Yes, it’s only one less click, but each click saved provides a better user experience. The tricky bit is balancing that with the pressures on screen real estate…
…but the other criticism here is that there is no equivalent “search must be above the fold”. This is particularly surprising since when I last looked at user navigation preferences and strategies for my previous employer (back in my LA days), it clearly showed that users clearly preferred to use search or site navigation — with only about 10% of users preferring to use the A to Z of services list. Although of course if our A to Z of services had been ‘explicitly’ provided users may have preferred it…
The report then looks at carrying out key tasks — decided here as paying council tax online, finding your local councillor, and viewing planning applications. It breaks these down into a number of sub-tasks, for example when finding your local councillor it asks whether users are easily able to:
- Find out which ward they live in to aid discovery
- Locate their councillor through various means, such as name, ward, postcode etc.
- Get the full name, phone number, email and a photo for their councillor
- Obtain surgery location and times easily and quickly
I’ve looked at the ‘Councillors’ one specifically here, because it has some meaning for me. This was one of the web applications that I worked on, and so I was keen to see what I had done well, and not so well. The site wasn’t listed as one of Better Connected’s top 20, so Webcredible didn’t review it, so you’ll just have to listen to my opinion.
Users can locate their ward either by knowing on a map where they are, or by knowing their postcode. Probably not perfect, but should certainly score for the first point. You can then find the councillor/ward details through a councillor name, a postcode, or a ward name, so we should have scored for point two. The councillor page then includes the full name, telephone number and photo, so we should have scored there too. The ward surgery times might have let us down slightly, as if I recall correctly, these were only available from the ward page, rather than the view for each individual councillor. On the whole though, I’d guess we’d have done at least fairly well on that one.
Webcredible then cite Richmond Upon Thames as a site which has done well, providing the councillor information clearly — although for me they still appear to be missing something. From the screenshot on the webcredible screenshot it would appear that there is no ‘contact me’ form and a user will require access to an email client to contact their councillor…
Their next section then looks at Transactional Capabilities, asking how well mandatory fields are indicated, whether an error summary is clearly provided, do forms provide an expected response time and so on. My experience of working in a local authority is that a lot of different online applications will be provided differently (some will be provided by a third party supplier, some will be developed internally — and of these, different internal developers will probably have different practices). This, to me, would tend to indicate that how well a specific individual form or two is constructed is not necessarily a clear indication that all forms on the same site will exhibit the same practices.
…but that doesn’t mean that Webcredible can’t use this to highlight some examples of good and bad practice:
Salford City Council leaves it to users to discover which parts of the form are required and which are optional.Local Council Websites: Good But No Cigar (Webcredible)
Whoops! This is particularly whoops-y because the standard and well established usability guideline for this isn’t particularly complex — using an * as part of the form control’s label to indicate that this particular field is mandatory.
With the error checking, they highlight good practice — not just providing a red mark next to the fields in error, but providing a summary of the errors at the top of the page. That’ll be a relief to any .NET programmers, as there’s a form control which will do pretty much exactly that for you (assuming you put it in the right place).
Apparently, 8 out of the top 20 councils scored 0 from Webcredible in this area. Which means that the others did pretty well, because as the overall average was given as 2.1, this means the other 12 councils must have averaged … let’s see now … carry the two … and divide by nineteen … three and a half points per site.
It’s risky to expect your users to have to hunt around to uncover the errors for themselves when using your forms.Local Council Websites: Good But No Cigar (Webcredible)
Clear enough for ya? Similarly, as well as a clear error summary, you should clearly indicate the errors themselves — and what the problem is. Is a date too far in the future? Is it not a valid date (e.g. 30th February)? Is a field mandatory? Provide this information next to the field in error. The job of the council website is to make it easy for users to complete their tasks… it is not that users should have to ‘learn’ how to use it.
Managing the user expectation is, as the report points out, crucial. If you fill in an application form or submit an enquiry, and no-one tells you how long you should wait, you don’t know whether to expect a response immediately, within an hour, within a day or two — and so you might well end up contacting the council in question again to find out (BUZZ! Avoidable contact! You lose 10 points). Whereas if you’re clear about what the users can expect, you’re less likely to encounter this problem:
Tameside Council’s site scored full marks for providing the response time [to a complaint] as well as a reference number and time & date of complaint.Local Council Websites: Good But No Cigar (Webcredible)
…and this brings us to the final section of tonights ‘Good, But No Cigar’ (anyway, I thought it was supposed to be “Close, but no cigar”? — see Wiktionary entry. Tcch. If you’re going to use a well known phrase or saying as your document title, at least get it right…) round, entitled ‘Navigation and Orienteering’. ‘Orientation’, sorry.
This looks at what should be simple things:
- Does the site offer a site map, called ’site map’?
- Can you see where you are within a section?
- Can you easily get back to where you were?
- Do all navigation components appear above the fold?
- Is navigation clearly distinguished from the rest of the page?
- Does search understand common mistakes and abbreviations?
Okay, the last one I’m maybe not so sure about, never having written a search engine, but they are are all clearly common sense (even if we still don’t know how far down ‘the fold’ is).
However, I’ve got to pick on Webcredible here. You get a picture of Bracknell’s site map, to which the caption says:
Bracknell’s site map quickly gives users an overview of the site’s structure. Barnet scored 5 out of 5 for their easy to use site map!Local Council Websites: Good But No Cigar (Webcredible)
Sorry? Barnet scored 5 out of 5 for their site map? Did you mean Barnet, or did you mean Bracknell? And if you did mean Barnet, then presumably the implication is that Barnet’s site map scored more highly than Bracknell’s did, in which case … (sigh: pause for breath) … Why the bloody hell didn’t you show us Barnet’s site map instead?
The ‘where you are’ bit then picks on North Lincolnshire, pointing out that while some pages have a heading, it doesn’t tell you what part of the site you are in, and contrasting that with the ’site breadcrumb’ approach used by many which, if done well, will show exactly where you are (and to what level) within the site.
Webcredible then discuss the importance of users being able to go back to their previous page, without needing to rely on the browser’s back button. This is a tricky one. A breadcrumb menu can solve some of these problems, but doesn’t work if users have arrived at a page via searching. However, at best it will be difficult to implement an on-screen ‘back’ button. For a start, Councils are under pressure to conform to WCAG 1.0, which indicates that you can’t rely on javascript. How then, can you implement a back button without that? The user’s previous history is not known to the server, only the browser.
Of course, if you’ve arrived from your own search page, there is a possibility that you’ve captured the search via session variables and so could build a history in this manner. And Webcredible think this is important:
With 13 out of 20 sites scoring 3 out of 5 and below this is clearly an issue that needs to be addressed.Local Council Websites: Good But No Cigar (Webcredible)
If you can do it easily enough, then fair enough. However, since there’s a perfectly good feature — the back button — pretty much common to all browsers which already exhibits this functionality, I don’t think this is a guideline I’d lose too much sleep over. Instead of being an issue that ‘needs to be addressed’, it could always be one ‘which isn’t very important’… now it could be that Webcredible’s usability testing has shown this to be important, but without any indication or evidence of this, I’ve got to assume that the guideline has been picked as a matter of opinion…
With navigation, navigation that is clearly and visibly distinct from the page body seems to be highly rated — the use of lines and/or different colours is something done by many (if not most) Council sites already, and is obviously what they are looking for.
What this report is is not an alternative to Better Connected, it’s a usability study of the top 20 sites, as according to Better Connected. It’s not perfect — more information on the methodology and why particular criteria were used as opposed to others would have been helpful — but it is an excellent example of one of the things that myself and Dan have been saying Better Connected should do — providing examples of best practice in individual areas, rather than just giving something a numerical ranking (although they have succumbed to the ranking/scoring bug.)
Listen: let’s stop scoring sites like this, okay? We don’t want our site reduced to a single figure score by a bunch of rankers…
While it’s not perfect, it is a document that Local Authority webbies can take and can learn from, which is one of the weaknesses of Better Connected. Most importantly, Webcredible realise that their scoring isn’t the most important thing — the most important thing is the usability for your site users, which really means you ought to carry out your own testing…
The usability guidelines presented in this report represent just the start to achieving excellent usability and an outstanding user experience. The use of usability guidelines is essential, but they should always be used in conjunction with usability testing on a regular basis. Usability testing involves analysing typical site visitors completing typical tasks on your websiteLocal Council Websites: Good But No Cigar (Webcredible)
Webcredible: well done. While not as comprehensive as Better Connected, and whilst only focussing on the usability, this is a really useful document highlighting best usability practice across the sector. On the other hand, the methodology ought to be clearer,and with references to usability testing to explain why particular guidelines were chosen, so I’ll have to say you were close… but no cigar.
PaulG says:
July 4th, 2008 at 9:30 am
Totally agree with this comment:
“So Council sites which provide text resizing controls will score better than ones which don’t. However, a Council which educates their users how to adjust their own browser text size settings is actually of more benefit to the users — because this will enable them to change their text size on any site they come across, rather than just on that particular council site.”
In fact, I downloaded the report, spotted that bit you quoted and promptly dismissed the whole thing – I was so disgusted.
I like your comments about finding councillors too, I think you are really spot on there. People dont live in wards – they live in houses, which are in roads.
This is the kind of thing LAs really struggle to get right, they simply cannot bend their cmss to do anything like this.
I am working on a small council focussed cms, and made my mind up to only return to this report as part of a feature checklist, just about its only use.
The Better Connected report it is derived from is just a grey soup of common practices, never thinking out of the box, and usually missing the big sea changes going on around us on the real web.
I continue to question the value of regurgitating and showing us the mirror the results of technical and buying decisions that were made, what, two years ago?
Let me give you two examples:
This is the first year that BC has mentioned RSS ! I started using it on an LA site at the end of 2002!
Lets analyse that. A significant new technology arrives which allows the sharing of data between organisations websites, and allows users to subscribe to the news they wish to follow. Did that only scream joined up government to me then? Was I on my own? Apparently not. This year there are loads of LA sites with RSS feeds of their own. How did these disparate organisations come to this decision?
Not by reading Better Connected – that was for sure.
Up until LAST years report, points were deducted for NOT having a “text only version” of their website, then mysteriously they are deducted points for having one!
By my reckoning that makes about 8 years of telling 450 IT departments that they should ideally do something that anyone with an iota of accessibility knowledge knows is just plain morally wrong, expensive, error prone and stupid.
Wrongness, uncertainty, doubt and bad steers emanate from these reports – the flaws are built in, because the nature of the beast is the internet. I have to read for at least an hour a day just to keep up with the changes in my particular corner of the web!
Blingy text-size changers are the “text-only version”s of tomorrow. Loved by designers, loved by Councillors because the designers told them it made the site “more usable”.
They will be discredited because of the reasons you point out and, mainly, they subvert the w3c standards.
Period.
paul canning says:
July 4th, 2008 at 8:11 pm
Thanks for this. I only scanned it and – yes – also spotted some doozers.
I don’t think the user priorities are right. What’s on (kids/pools etc) i think would rank right up top. As you say, no methodology for how they decided this.
Most usability issues can be addressed by actually talking to customers. That’s why I have ‘cheaper usability methods’ in my ‘10 pt plan).
xp
SteveH says:
July 9th, 2008 at 5:05 pm
I disagree with the comments about text resizing. The purpose of a council website is to provide information and services not to educate visitors in using the web. In fact, when the information on how to change text size using the browser setting is provided, my experience is that it is often hidden away so that anyone who, by definition, is having difficulty using the site, is unlikely to find it.
Beyond that specific point, all we are really seeing here is that if you ask a lot of individuals what is important to them you’ll get a lot of different answers. That’s surely the whole point of using a range of users to test a range of scenarios as both Better Connected and Webcredible seem to have done.
Paul Englefield says:
July 9th, 2008 at 9:31 pm
I too enjoyed the report and appreciated the illustrations. I would have appreciated a stronger focus on information architecture and content style. Given the number, diversity, complexity and sensitivity of council services, the deep challenges seem to be more about finding, understanding and trusting information than about the details of interaction design.
Stuart says:
July 10th, 2008 at 4:16 pm
Re: Text resizing – I think a marriage of the two methods – text resizing controls and active education would be the best plan, I’m not sure how it would work – I’d need to give it some thought, but a message like “We’ve noticed you’ve used our text resizer to resize text – did you know you can do it in your browser for every single site – visit [link to my web my way article] to find out more”
However, the only problem I can think of is users of IE6 becoming frustrated when they can’t resize a site where the text is sized in pixels and blaming the council – not very likely, but worth some thought.
Stuart says:
July 10th, 2008 at 4:22 pm
Also, it’s probably worth mentioning that the largest need for text resizing controls is amongst the elderly population, who are generally less web-savvy than us young folk, so clear signposting to text resizing can only make things easier for them…
Pages tagged "orienteering" says:
July 12th, 2008 at 3:33 am
[...] bookmarks tagged orienteering Webcredible vs Better connected saved by 1 others trackingz bookmarked on 07/11/08 | [...]