Tuesday, June 2nd, 2009
It’s always hard to work out in the open, because it means exposing unfinished thoughts and expressions to public scrutiny, something which is, quite frankly intimidating. We are much more used to gathering up our information, hammering it into shape and producing, in the end, a nicely polished piece of work. But that’s not necessarily the best way to do things.
Not only do I not have the luxury of the time it would take me to properly go through that process, but I believe that the end product will be better if I, as my maths teacher used to say, show my workings. But it’s not without trepidation and caveats that I give you the first draft of my assessment of a small sample of third sector websites.
After the jump, you’ll find a description of what I did and what I found, and I’d very much like to know what questions it throws up for you. I know that it poses quite a few more questions than it answers for me, but I shan’t prejudice your reaction by stating them here.
To put it in context, this is only part of my assessment of the current state of play with third sector use of social tools. My survey is still open, and if you are in the third sector, please do take a look at it (or pass it on to anyone you know who might be able to fill it in). I’ll also be doing a literature review (so please send me links to anything you think I should be reading).
These are preliminary results - I can’t stress that enough. Please don’t start reporting these or writing blog post about them (not that I’m assuming you want to!), not only because in the course of refining the report things may change, but also because they are lifted out of the context of the rest of the report. But please do feel free to critique my work as extensively as you like, in the comments or on email if you want.
I want to do a little bit more research for this section, but I’d like that to be directed not just by my own judgement, but also by yours. What else do you want to know? What questions does it make you ask? I can’t promise to be able to find all the answers, but I will do my best. (And don’t worry so much about typos, but please do point out anywhere where my maths doesn’t add up!)
Right. *gulp* Have at it!
A broad sample of 56 websites was identified, covering a number of different types of civil society association, (see below), of different sizes and operating at a variety of levels, e.g. local, national, or global. The list was compiled in conjunction with Carnegie UK Trust to ensure that it was as representative as possible. Some organisations which were known to be using social media were added to the list to provide balance.
Types of civil society association:
Voluntary community organisation
Of the initial list, 32 organisations’ websites were assessed, including at least one from each category. Of those, three were either non-functional or “under construction”, reducing the number of websites with completed assessments to 29. The assessment sought to answer both quantitative and qualitative questions:
1. Does the site use any social functionality or provide syndication feeds?
2. What third party tools were mentioned, linked too or included on the site?
3. What media, e.g. audio or video, was used on the site?
4. Does the site mention or link to any virtual worlds?
5. Does the website say that it has been optimised for mobile, or provide information on mobile applications?
6. What tone-of-voice does the site use?
7. What can visitors do on the site?
8. Rate the look-and-feel of the site, e.g. design and layout
9. Rate the navigation and usability, e.g. Is it easy to get around?
10. Rate the ease of finding the blog, if the site has one, and the use of standard blog ‘furniture’, e.g. calendar, categories, archives
11. Rate the ease of finding RSS, if the site has it
12. Rate the ease of finding audio/video
13. How many links in from other websites does the main URL have in Icerocket?
14. Does the site link out? To how many other sites?
Qualitative answers were based on the author’s experience and judgement, and sites were rated on a 1 - 5 scale.
Many of the websites assessed suffered from poor visual design, i.e. the way that the site looks, its page layout and colour palette, with 21 (72.4%) site scoring average or worse, including 13 (44.8%) scoring poor or very poor. Many of the sites were confused and cluttered, lacking clear calls to action and little opportunity to interact either with the organisation itself or the related community of interest. Only 8 (27.6%) sites scored good or very good for visual design.
The navigation and usability of most sites was also of a poor standard, with 25 (86.2%) scoring average or worse and just 4 (13.8%) of sites reading a good standard. The main problem was that many of the sites had multiple navigation menus, sometimes as many as eight navigation areas on one page, which made it very difficult to find information and keep track of which pages had been visited. Standard navigation tools such as ‘wayfinding’ navigation, i.e. clear text-based menus at the bottom of the page, was used by only two sites and breadcrumb navigation, i.e. showing the user their path through the site at the top of the page, was also rarely used.
Two sites had particularly poor navigation, with one site actually turning out to be a mishmash of microsites, some of which had the same design and navigation items and some of which had totally different designs and no clear link back to the starting point. The other had menu items that changed, with no visible logic, depending on which page the visitor was on.
On larger sites, the information architecture tended to be overly complex and confusing, reflecting the association’s internal organisational structure rather than the needs of the visitor. Few sites showed any evidence of user-focused design, which concentrates on understanding the site through the eyes of the user and what they wish to achieve during their visit to the site. The majority of sites would benefit from even basic usability testing and by utilising specialised feedback-gathering services such as UserVoice or Get Satisfaction to allow users to share and discuss ideas for improvements to the site.
Many of the sites that used social media did so in an almost perfunctory manner, failing to adapt to the prevalent culture of social media and thus missing opportunities to fully engage with their community.
1. Social functionality and syndication
Does the site provide any social functionality, such as blogs, or syndication tools, such as RSS, that visitors can use without having to go to a third party site or application?
Overall, 18 (62.1%) sites provided some sort of social or syndication functionality, however limited, but 11 (37.9%) websites provided only static, non-interactive content.
The most popular tool was RSS, which was used by 15 (51.7%) sites assessed, although 2 of the sites did not provide an explicit link to their RSS feed on their website. Instead its presence was indicated by the FireFox browser, which provides the RSS logo in the location bar on every site that has a discoverable RSS feed.
One third of the websites that provided an RSS feed, (i.e. 5 (17.2%) all assessed websites), did not have a blog. Instead, they produced feeds of press releases, news articles or events listings. Conversely, 1 of the blogs had no RSS feed.
The majority of the RSS feeds, 10 (34.5%), were easy to find, being linked prominently either in the navigation or on a dedicated page. Only 4 of the feed links were difficult to find, or completely absent (indicated only by the Firefox browser’s automatic discovery functionality).
Only 3 (10.3%) used RSS to display information on their own site from external sources, e.g Twitter.
After RSS, the next most popular technologies were forums and blogs, both of which were used by 9 (31.0%) websites. Many of the forums required registration, but were otherwise open to the general public, and some were accessible only to the organisation’s members.
Of the 9 organisations that provided blogs, 5 had blogs that were hard to find from the main site, either with no prominent link or no link at all in the main navigation menus. Some blogs did not link back to the main site, and were thus completely separate from the organisations’ main web presence. Four were rated as very easy to find from the organisations’ main page.
However, 6 of the 9 blogs were rated average or worse for their use of standard blog functionality, such as categories, calendars, archives, links to recent comments, etc. The use of such functionality is important as it provides the visitor with key visual cues that what they are looking at is a blog, which then allows them to properly adjust their expectations, e.g. regarding the provenance and tone of the content they are about to read. Only 3 blogs were rated good, and none as very good.
Very few used widgets such as Flickr badges, or feeds from Twitter or Delicious. 3 (10.3%) websites provided “Share This” buttons which allow the visitor to share the page on services such as Delicious, Reddit, Digg, or Facebook.
Only 2 (6.9%) websites allowed people to comment on content outside of the context of a blog, although for 1 of those sites, comments were not displayed on the site but were essentially a content feedback mechanism.
None of the websites surveyed provided a wiki of any sort, although one site did allow visitors to view and download computer code from CVStrac, a popular version control system used by computer programmers.
2. Use of third party tools
Does the site have a link to or a feed from (e.g. a badge or RSS feed) any third party tools displayed on their website?
Of the websites surveyed, 15 (51.7%) did not link to, mention, or include content from any third party social tools. Of the 14 (48.3%) that did, the majority (11, 37.9%) used video sharing sites, with YouTube being the most popular (9, 31.0%), Vimeo accounting for 3 (10.3%) and 1 (3.4%) site using MySpace for video sharing. No other video sharing sites, such as Viddler, were used.
The micro-conversation tool Twitter was also popular, used by 9 (31.0%) sites. No other micro-conversation tools, such as Identica, Jaiku or Friendfeed, were used.
Only 6 (20.7%) sites used social networks, with Facebook used by all six, 3 (10.3%) also using MySpace and 2 (6.9%) also using Bebo. No other social networks or community building tools, such as LinkedIn, Ning or Webjam, were used.
Social bookmarking was used by 3 (10.3%) sites, with 2 (2.6%) using Delicious and 1 (3.4%) using Google Reader Shared Items.
Photo sharing sites were used by 2 (6.9%) sites, who exclusively favoured Flickr, and presentation sharing service Slideshare was used by 1 (3.4%) site.
3. Use of audio and video
Does the site provide any audio or video materials for visitors?
Nearly half the sites assessed (14, 48.3%)) did not provide any audio or video material at all, despite the fact that for some of them it would be an obvious opportunity given the nature of their work.
Of the 15 (51.7%) sites that did provide multimedia for their visitors, 13 provided video and 3 provided audio. 12 sites used third party providers to host video, with 10 of those sites embedding that video on their own site. One site provided streaming video via a Flash interface which was not downloadable or embeddable.
One site that had a YouTube channel did not embed any of those videos on its site, and another organisation supplied videos to a specialised, sector-focused ‘TV’ website, but did not provide any of those videos on its own site.
3 (10.3%) sites used audio, only 1 of which also offered video. Of those three sites, 2 (6.9%) offered audio downloads, 2 (6.9%) offered podcasts and one offered both. The site which offered only an audio download mislabelled its download as ‘podcast’, but did not provide the audio file enclosed within an RSS feed, thus denying the listener the opportunity to subscribe to the podcast in their RSS aggregator or via iTunes. This audio cannot, therefore, be accurately described as a podcast.
In terms of the ease of finding multimedia offerings, 11 sites scored average or worse, and 4 scored good or very good. Most sites embedded videos on the pages where they were relevant, but few offered ways to aggregate all video/audio content in one place, so it was often hard to know at first glance whether or not the site provided multimedia at all. Those sites that provided site maps did not list which pages included audio or video, and which did not.
4. Use of virtual worlds
Do any of the websites mention or link to any virtual worlds?
None of the websites surveyed made any mention of virtual worlds such as Second Life or Habbo Hotel. It was not possible to search all virtual worlds for the presence of these organisations, but it’s reasonable to assume from the lack of links to or discussion of virtual worlds that they are not being used.
5. Use of the mobile web
Do any of the websites mention if they are optimised for mobile, provide a graphics-lite version, or provide applications for mobiles?
It was not possible to properly test all websites on a variety mobile phones, so it was not possible to definitively say how well prepared the sites were for use on mobiles. None of the sites mentioned either a version of the site optimised for mobile or a graphics-lite version that would be usable when accessed via a mobile browser. It may be that some sites work reasonably well on mobiles, as a co-incidental result of the content management system used to create the website.
None of the sites made mention of nor link to any mobile applications, whether for iPhone, Nokia, Blackberry, Android, Palm or other devices. Whilst it’s not possible to search all app stores for applications from or related to these organisations, the lack of links to or discussion of mobile applications seems to indicate that they are not being developed.
How are these organisations speaking to their visitors, whether on their own site, their blog, or in third party tools?
The majority of organisations used a formal, corporate tone to their communications, with 18 (62.1%) of sites having no individual, informal voices in evidence. All the websites that did not use social media in any form fell into this group, including sites that only used traditional forums rather than blogs or Twitter.
Generally speaking, the more social tools a site used, the more individual voices were heard. Only 1 (3.4%) site was mostly done by individuals in their own voice, and that was an ad hoc community group with no obvious official governance structure.
However, the correlation between use of social tools and voice did not always hold true. There were several sites that, despite using several social tools, still maintained a very polished, broadcast-style voice, rather than the more intimate voice usually associated with social media.
7. What can visitors do on the site?
What sort of actions can visitors take?
Five (17.2%) of the sites assessed were passive sites that provided information, but did not allow users to act or interact in any way with the organisation or each other.
The most common action, enabled by 20 (83.3%) sites was to be able to give money in some way, whether as a one-off donation, subscription, through paid membership or other mechanism. Yet only one site provided a specific way to help recruit others to join, in this case by allowing visitors to buy a gift membership.
17 (70.8%) sites allowed people to volunteer, either by providing ways to get in touch with an organiser, or by directly volunteering for a specific task via the website. Only 4 (13.8%) of sites allowed this latter option, which seems like a missed opportunity.
11 (37.9%) sites allow people to sign up for events, whether via the web or by email, and an equal number provide ways for visitors to fundraise. Some of the events were fundraising events, but not all the fundraising was via events.
9 (31.0%) sites had shops where visitors could by equipment, merchandise, gifts and other physical objects, or information, publications, services or consultancy.
Very few (4, 13.8) sites had any facility for visitors to ask a question of the organisation directly, and most of those that did were not real-time.
8. Incoming and outgoing links
Does the site link to other sites? Is it linked to by blogs?
The blog search engine Icerocket was used to determine how many blogs linked to the site in question, thus providing a rough measure of the popularity of the site. Links are a valuable currency in the online world as they are one of the measures of importance that Google uses when determining a website’s position in search results.
It appears that there is a rough correlation between non-use social media and the number of links an organisation gets. Sites which do not use social media, or which only use forums, tended to have fewer links than those that used a variety of social tools. Of the 14 (48.3%) sites that had no social media (but may have had forums), 8 had no incoming links, 4 had up to 15 incoming links, and 2 had over 160 incoming links.
The number of incoming links varied widely amongst 15 (51.7%) sites that used one or more social tools (not including forums): only 1 had no incoming links, 3 had up to 15 incoming links, 5 had between 16 and 100 incoming links, and 6 had more than 100 incoming links. There were no clear correlations between tool used or tone of voice and number of incoming links.
It was very difficult to count the number of outgoing links on a site, and only 5 (17.2%) sites actually collected outgoing links on a single page or in a list on their blog. It was therefore not possible to draw any firm conclusions about trends in outwards linking and their relationship to the use of social media.