If you’re in higher ed web development, you probably saw this article making the rounds criticizing university web sites. Melonie Fullick put this together along with the feedback of other Twitter users after trying to research some information from various sites. I, too, recently had some complaints doing some research on programs at institutions and finding it infuriating at times trying to get relatively simple information. I’ve talked with a couple folks about the article as well, and thought I’d give some additional commentary. Not necessarily counterpoint, or refutations, just an additional viewpoint as someone who spent years behind that curtain.
For those that don’t know me, I’ll also preface with the point that I spent six years in higher education running a website for a mid-sized DII school. I’ve since moved on, but still maintain contacts with the industry, and have a side company that deals largely in higher education interactive mapping.
I’m going to go point by point on some things and reply to some of the specific tweets she cites. I’ll reiterate, too, that I’m not writing this to be contrary to Melonie’s article. More as a complement to it, because I agree, higher ed sites suck. The thing is, there is a reason for it, though that’s not to say those reasons are an excuse. I think it’s worth examining and exploring (and acknowledging that there are a shit-ton of committed people fighting to make it better. You know who you are, you motherfucking Viking warriors of web design). Maybe in that exploration, we can generate some community discussion on the matter that folks can take back to their offices. Maybe, just maybe, we can light a fire under someone’s ass to help out our friends in the trenches.
…it’s a problem that’s ongoing, and one that generates much wailing and gnashing of teeth among regular users of the sites, including current students and faculty.
Starting at the start. This is really part of the heart of the matter, isn’t it? That phrase: “regular users.” This is like ground zero for problems that university sites have. It’s not only that they have trouble identifying regular users, it’s that what constitutes a regular users changes drastically between areas of the site (I think this is well reflected in the replies to Melonie’s tweet). I’ll be perfectly frank, too: I don’t think there’s a good solution to this problem partly because there is no such thing as a regular user for them. That’s not to say that there aren’t solutions, I just don’t think there’s enough money or manpower available to schools to implement the sort of site that could address it. It’s precisely why most sites end up with audience landing pages just below their homepage in their information architecture (more on this below). Without quoting a ton of the responses she got, this definitely goes to a lot of the complaints: “the things I want aren’t where I want them.” It’s virtually impossible to prevent this complaint universally. Any change a developer makes will draw complaints from someone, even if every ounce of data and research would indicate the change is beneficial. There will always be irregular users that think they are regular.
…the front page of the website might be badly designed, making it difficult for users to find any signs of how to navigate…
I won’t comment on “badly designed” in the sense of the actual look and feel – I think that’s really qualitative, and is constantly improving (though behind the curve). While there are a lot of poorly designed sites, there are also a lot of really good ones, too. So it’s a wash. I’d say finding signs of how to navigate most sites from the homepage is actually quite easy, in the sense that it’s easy to find nav elements, though I can absolutely see how it might be tough figuring out which navigation to use. For instance, most home pages have no fewer than five levels of site navigation (which is insane):
- Utility nav (logins, calendars, giving…)
- Main nav (programs, alumni, athletics…)
- Audience nav (sometimes folded into the main nav, sometimes not)
- Intra-body nav (usually an amalgamation of clickable elements related to news, events, and other low value targets)
- Footer nav
And that’s just to start. Many of these can be broken down into separate elements, spread out, and built upon. Some sites have as many as eight or nine navigational elements or even more on their homepage. I think this problem has improved over the years, but it’s definitely not gone. Melonie later adds in:
A number of people sounded off specifically about useless (and huge) drop-down menus and how they affect a reader’s experience on the site.
This is one major failure for universities and is symptomatic of the noise generate by the politics of running an institutional site. In an attempt to deal with overwhelming, convoluted IA, they just end up putting everything in a jumbo dropdown from the main navigation (hey look, you have a link on the homepage now!), without considering how utterly unreadable and complex it makes the site. We’ve known big lists are usability traps for 13 years, yet we still use them, just like carousels on the homepage, because they shut people up. It’s losing the battle in hopes of winning the war. This dovetails with another criticism:
Why are so few designed with accessibility for people with disabilities? Especially when they have a Disability Studies program???— Joanna L Pearce (@jlphistory) June 22, 2016
Consider this issue for a moment. Really consider it. Think about how a jumbo dropdown looks to a screen reader. Consider how it’s navigated by someone tabbing through with limited motor control. Imagine how maddening that is as a utility. It’s a means to an end, but it’s one that is potentially hurting users, especially if you haven’t taken proper consideration into account for skipping navigational elements. You hide mountains of garbage in jumbo dropdowns, and waste precious homepage real estate on carousels devoid of value. And that’s just a single aspect of accessibility that happens to tie points together for the sake of this article, without getting into badly marked up code, videos without captions, color contrast, and on and on. Instead of viewing accessibility as a key feature of good usability in design and development, it’s seen as an anchor that burdens work. All of this culminates in one particular comment that goes to something I’ve talked about before:
Probably says something that I'd rather google "site:<X>.edu <whatever>" than use most unis' site navigation.— Cheng H. Lee (@chenghlee) June 22, 2016
I’ve longed for that idea that you could build an entirely search based site. And I think it has incredible merit for a university since they’re bottomless pits of largely useless content. But it goes to my comment above that I think doing it right and being successful with it is simply beyond the means of universities (and even most large companies). The funny part is, many universities already use Google as their search backbone, but they mire it in the overhead of the rest of the site and spend so little time on good SEO practices that the usefulness is crippled (aside: most higher ed web devs I know are well versed in SEO, the issue is that they have dozens, if not hundreds of non-web people responsible for day to day page management that don’t know SEO from NFL). Imagine what you could do with a setup that didn’t do away with secondary navigational elements, but instead elevated search to a primary navigation option instead of relegating it to a toolbar input field somewhere out of the way. Create semantic, searchable collections of like content that can be filtered on (news, catalog, library, athletics, etc). Use your analytics to inform the process. Focus keyword matches on top searches and common elements. Write a scoring boost equation that devalues old content. This all takes work, but it’s a huge opportunity to make an incredible improvement.
“Buried” information is often of the most basic kind, including schedules and timetables, fees and financial information, campus maps (frequently available only in PDF form), transcript requests, and even the university’s mailing address… Key links, including contact information for the institution, should be on the university homepage and clearly visible. These could also include links to the library, to postings of available jobs, to faculties and departments, and to a contact directory and a site index.
This one is incredibly hard to deal with, and just like with the audiences, highlights a fundamental problem higher ed web developers deal with. The nature of the complaint is a basic definition of why it’s a problem. University web developers never want to bury important and frequently requested information. But the problem is – there’s so much of it that’s important and frequently requested. Everything can’t be on the surface, and much of it isn’t related to each other. Universities are simply too big, and have too much content to successfully deploy a relatively flat IA. Ask a dozen people what should be linked directly from the homepage, get a dozen different answers. And the people responsible for giving those answers in a work context aren’t the ones listening to the user experience research and testing. They just want shiny whizbang toys everywhere. Sexy over substance.
“My other big pet peeve at the moment is that HIPPOs just don’t care about usability and the user experience. The response I get is that our students are smart, they will figure it out.”
This was one response a friend gave me with regards to priorities in design and development. Given the fact that higher ed is fairly regularly targeted with articles like Melonie’s, I think it’s pretty safe to say that students won’t figure it out. The smartest web people in the world, working at some of the largest brands, along with industry experts and UX professionals all know this to be the case. Yet higher ed still drags this old mantra along like a worn out blankey.
I mentioned this point to Mark Greenfield during an exchange. This comic strip has literally haunted higher ed for six years this month, and it’s still just as true now as it was when it came up (not necessarily on the specific points, but in illustrating the divide users and schools perceive in site usability). It speaks to an innate inability to escape the nature of their infrastructure. On this subject, Curtiss Grymala poignantly mentioned something worth considering every time this comic is thrown out:
“Regarding the XKCD comic: That is not user-testing, and I really wish people would stop throwing it around as if it were. Just because Randall Munroe only wants to find addresses, phone numbers and maps on a university website does not mean that the majority of visitors to *my* site are looking for that information. They could be, but you can’t use a Web comic as your basis for that assumption; you have to use user-testing.”
Mark summed it up more succinctly:
“Usability matters, end of conversation.”
The form that usability takes doesn’t matter, as long as you do the research, get the answers, and build solutions around them.
Melonie mentioned that it seems like sites still act as a mirror for the internal problems and dysfunction of the organization of a school, even after redesigns. It’s true because that organization generates the committees that dictate direction, rather than acting as a partner with the marketers and developers. That problem is compounded by the sheer amount of content that has to be addressed in a redesign, and the woefully inadequate staffing for the effort. You would be amazed at the number of universities that only have one person responsible for managing their website. Imagine how overwhelming that would feel, being one person trying to steer a 30,000 page website and solve all the problems we’re talking about here. I know exactly what that feels like. Strategic problems entirely aside, that one single logistical hurdle is virtually insurmountable. One person can only generate so many man hours of labor a week, and that’s before considering the committees and meetings they waste time on during the week doing nothing but talking about problems they aren’t being allowed to fix.
Circular links. School> program> certificate description> learn more> program— Dr. NOPE (@reallyHibbs) June 22, 2016
No commentary for this. It’s terrible, and schools shouldn’t let it happen. Bad design, bad IA, and bad UI elements all contribute to this, and it needs to be addressed. These traps happen because it becomes extremely difficult and time consuming to define user flow four and five levels deep, so what ends up happening is secretaries tasked with editing the web site are creating pages that far down, and automated navigation tools provided by content management systems fight what they are doing. It’s another result of not having the staffing available to police content and navigation to prevent it from happening. The simple answer is paying attention to the taxonomy being used in link generation, but that’s much easier said that done at the scale of these sites.
Another issue is the way university websites usually pre-categorise content by its presumed audience, and organise it accordingly — i.e. by group such as “current students”, “staff”, and so on.
Another tough pill. This falls into the camp of being a necessary evil. As pointed out above, there’s a ton of ground universities need to cover. They have to serve all of those people at the same time. The only thing worse than rough organization, would be none at all. Could it be better? Yes. And some school’s have tried. The problem is that there aren’t a lot of alternative modalities available that can justify breaking the convention. Others from the original conversation complained about content arranged based on the school’s internal organization structure. There are only so many reasonably logical ways to tackle this issue, and they all have major pros and cons. At least audience based siloing helps to define some degree of distinct activities they’ll engage in. I do think there are some huge improvements that could be possible in this area, but they all involve significantly increased development to address them. Most schools won’t commit the time to that matter. I have sympathy for those alumni looking at graduate programs offered only online who are trying to figure out where they start to look at that. You have to have some kind of boundaries to work from though, just to maintain a degree of sanity in the process.
I would say that many university websites reflect the structure of the institution in general, with informational silos that make it difficult to find out what you need to know — unless you already have some background knowledge.
Not to mention knowing their internal lexicon. How many high school graduates really know what a registrar or provost or bursar is, after all? Heck, there’s even confusion over prospective vs. current students (well, I’m a senior in high school, so I’m a “current student,” right?). Yet we still use these labels even though they only add confusion to the site. This is an example of picking the hill you want to die on. For most university web developers, there are such bigger fish to fry that it’s not worth fighting the battle over vocabulary. And trust me, it is a battle. People in higher ed are utterly committed to their titles. Some are lucky enough to get in a card sort test here and there, but they are still the exception.
Site search is another major problem. Search effectiveness can be radically different from one website to the next. For some universities a search can turn up useless details from an archived calendar from 5 years ago, but no basic contact info for a staff person.
Universities treat their websites like a journal, responsible for storing all information put into it in perpetuity, regardless of usefulness or effectiveness. This is because it’s nearly impossible to enforce strict content lifecycle strategy and review, since web offices might be tasked with helping create things, but are rarely given authority to take things down. Internal turnover can also make it extremely hard to research content continuity and know where a page came from, why it’s there, and if it’s still needed. Most also don’t have robust search backends (relying instead on things like Google Custom Search). Even a “small” university site is 20,000 pages, with major schools easily approaching the one million page count and beyond. At the smallest end, checking pages every day, you’d have to review 55 pages a day, seven days a week to be sure they need to be there and filter them out of search. That takes a huge bite out of the time you have for other projects. Universities are simply digital hoarders. Like so many other issues, the manpower just isn’t there to deal with it, and folks outside the web team don’t grok the reason this is a problem. The only solution is a hard reset, rebuilding from scratch. Just like trying to get a hoarder to clean their house though, this is a terrifying prospect to university administration. After all, they might lose the minutes from the April, 2003 meeting of the Committee on Committees. I wish this was a joke. Most folks that are still reading this far know it isn’t.
Which brings us to the next serious flaw on many university websites: missing, outdated, incorrect, contradictory and/or ambiguous information… It’s possible that many of these problems are caused by confusion about responsibility for the maintenance of websites… “at many unis, faculty are responsible for maintaining [department] pages. What do we know about web design?”
Bingo, see my last paragraph just above. Hey, higher ed administrators, if your users can tell this, that’s a pretty damning observation on the way you manage sites. The principle of putting content experts in charge of content is outmoded, and has been well proven to not work. How do I know? I listen to all the griping your developers do on a daily basis. I hear the complaints about users that break pages because they don’t know better, users that violate branding standards, use fonts because they’re pretty, don’t know anything about accessibility, include third party tools without telling you, and that you don’t allow allow your web team to take away access when these violations occur.
Distributed authorship model = tragedy of the commons in reverse. Everyone thinks, "nbd if I put up one page" x100 = can't find jack on site— Aaron Rester (@aaronrester) July 6, 2016
And as a user, I’ve experienced first hand the content created by people that never should have been allowed near your CMS. The lack of respect for any real web governance, and empowerment from school administration to execute some kind of web policy, is going to cripple your ability to function in the long run. For every one non-web person capable of maintaining their part of the university site, you have to allow 20 the same access who are clueless. Faculty aren’t going to walk in one day and be an expert at content creation. It’s a recipe for failure. It’s never going to happen. Stop trying to make it happen.
But to ring the same bell over and over, it’s a matter of staffing. There is a demand for so much output, but an unwillingness to properly staff web offices to do it right, that no choice is left except to put people in charge of making things on the website who are utterly and completely unqualified to be making those changes. Mark later mentioned to me:
“I think one of the biggest problems we have is that higher ed does not embrace their digital experts. Not only does this result in a poor website, the number of talented web people leaving higher ed is a real issue, and I don’t blame them for leaving. The fact of the matter is that the web just isn’t important to senior administrators. If it was, resources would be made available. Too often leadership views the web as a cost center instead of the institution’s most important strategic asset.”
So what do they do? They tell the secretary of the math department to edit their pages, because their time is cheap, and they don’t consider the value lost when web developers have to clean up the messes they make. Efforts are made to train and educate, but it always leads to more hours of the day being wasted doing phone support for something that could have been resolved with a help ticket and five minutes of work from someone qualified.
Lastly, there’s the conflation of promotional and informational material and approaches.
This is where things get sticky, right? Because the fact of the matter is, a school’s website is a promotional and marketing tool, through and through. If something on the site isn’t designed for, or geared towards acquisition or retention, it really doesn’t belong on the front facing website. Meeting minutes don’t belong there. Research abstracts don’t belong there. Faculty websites don’t belong there (given all the ones I’ve seen, they don’t belong anywhere). My argument has been – and continues to be – informational resources for currently engaged audiences (faculty, staff, students) belong in a separate place. As we’ve already observed, these audiences goals are not at all the same as incoming/prospective students or alumni, etc. A lot of time and pressure can be taken of marketing and web staffs if responsibility for those audiences was shifted into a utility driven portal. Yeah, it might be ugly, but if it gets the job done, those folks are a captive audience. You can afford to sacrifice pretty for functional in that case.
It’s possible that in trying to cater to so many audiences at once, the universities fail to please anyone.
It’s also possible they’re just not taking into account user feedback or even providing channels for that feedback to happen.
In general, yes and no. I’m sure plenty of schools pay no attention to user testing. But a lot do. The problem is that for a lot of schools, even with that information in hand, what the HiPPOs want, the HiPPOs get. And this is an incredibly huge problem. In general, universities need to start positioning CIOs or CMOs (or CDOs or CWOs if they want to be really forward thinking) in such a way that they are allowed to truly direct the development and strategy of the school’s web presence, free of interference from (unrequested) committees, faculty, etc. These people will be smart enough to know when to involve the right people at the right times from outside their teams. Everything here has solutions. None of them are perfect, but all of them could be done better if there was trust place in an individual to drive the strategy and produce a cohesive product developed on best practices and adequate research and planning. It’s not that there isn’t an interest in web governance, there’s just no understanding of what it really takes to do right with the people that would have to be on board to empower it. The web folks totally get it.
So I encourage users of university sites to have patience and understanding. Let’s also not forget that when you ask for Twitter to give you gripes, you’re going to get gripes. Everyone knows the experience is frustrating for someone. But there’s so much institutional barbed wire involved, that while solutions seem obvious from the outside, there are few places that exhibit qualities of such a morbid quagmire as higher education does. Plus, plugging one hole frequently opens up a new one. Solving your problem just makes a new one for someone else. It’s an unenviable task to juggle all that and stay sane (higher ed web offices have pretty high turnover, usually with a lot of great people taking private sector jobs that are less stressful and pay better). But they’re working on it. HighEdWeb 2016 is in October, and over 700 web developers from schools all over the world will be there talking about how to solve these exact issues (disclaimer: I’m speaking at it and you should definitely come and it’ll be awesome and we’ll be best friends and rob a bank… wait… what?).
This is where higher ed is very different from the private sector in terms of web development. At a company, the web team likely reports to someone, and that someone answers to a CMO or marketing/management board. They work together to plan timelines and priorities, and align their resources with the things that need the most attention. And the buck stops there. Other employees can’t just come in and inject their demands in the process (normally). If other people want stuff done, they send in a request or ticket, and it gets planned accordingly. In higher ed, the web team has a director. That director probably answers to a VP of one part of the school, who in turn is part of the presidential board, which answers to the president. But then you also have all the other VPs that also get to throw votes in. And it doesn’t stop there. Because you probably have four or five colleges within the university. Each of those has a dean who wants stuff done for their departments – maybe a dozen or so. Then each department has a chair that oversees several programs. All of these people at each level probably have Ph.D.s (or are working on one) and most certainly feel like Very Important People who need to get their way to get their job done. Like that one department that feels the university branding is stifling their ability to recruit because they don’t know why but it is so shut up and let them be free, dammit. They also get paid more than the web people, so clearly their opinion matters more.
So this brings me to my last point. Last week, I went on a bit of a… rant. I feel bad when I hear all of the problems my friends deal with at schools, and I save up those nuggets of information until… well… this happens. I Storified it, thinking I’d do something with it eventually, and it seems like this is as good a place as any to stick it. It should give you an idea of why it’s so hard to develop good sites in higher education.