Stupid Usability Question! Duplicating Navigation Elements
October 16, 2012 10:51 AM   Subscribe

Web navigation best practice: I'm managing a newly relaunched website, and my superiors want me to duplicate navigation options in a couple practices (clarification after the jump). Something in my Krug- and Nielsen-lovin' soul cringes at the thought. Am I overreacting? If not, can you help me assuage their concerns elegantly without compromising the information architecture?

Setting the scene: The site relaunched about a month ago, and we're still in the tweaking and transferring content stage. I am the only person actively contributing to the multi-thousand-page site; my background is mostly in content creation, though I have taken a few good Web basics courses.

We have the typical five-section informative website—About Us, News & Events, etc. Say we have a regular lecture series or a neat piece of art (not a new series or a new exhibit; this is a longstanding institution). The Boss wants Lecture Series in the secondary navigation (the drop-down list from the primary nav) of not only News & Events but also About Us because "more people will see it and come to the event." Other Boss wants Art Piece in the secondary nav of not only Outreach but also About Us, because "Art Piece is good for marketing purposes."

As I mentioned before the cut, I'm a conservative sort of Web manager who trusts my audiences to figure out that events are under News & Events. I balk at duplicating elements in multiple sections because THAT DOESN'T GO THERE and SLIPPERY SLOPE. I'm trying to compromise by using some right column highlights to call out Art Piece and Lecture Series, but that's viewed as not quite enough.

What do I do? (Bad answer: Whatever they tell you to, because they sign your paycheck. That might end up the best solution for this scenario, but I'm looking for the best practice answer for future reference.)

Note: We're using a content management system that lets us repeat menu elements without duplicating pages of content, so our SEO mojo shouldn't be affected(?).

Thanks, Mefi.
posted by kwaller to Computers & Internet (9 answers total) 3 users marked this as a favorite
These discussions can easily be solved by proper A/B testing and letting the data guide you. Not even experienced web designers should assume that they can read the minds of their visitors.
posted by Foci for Analysis at 11:17 AM on October 16, 2012 [4 favorites]

I agree with Foci -- all the UI I've ever learned focuses really strongly on testing. The way you think isn't necessarily the way other people think, so what seems super intuitive to you may be incredibly confusing to visitors to your website.

And this definitely falls into "proper A/B testing" but make sure you use a variety of folks at different levels of computer fluency. There's stuff that people do that I would never ever think of doing because, to them, it just seems easiest.
posted by itsamermaid at 11:34 AM on October 16, 2012

"...who trusts my audiences to figure out..."

As someone who has witnessed first-hand a lot of testing and eye tracking, I can honestly say that this is a terrible idea. I mean no disrespect to your instincts or the gaggle of "rookie" internet users out there, but you can't trust your audience for anything. If people walked the streets the way they navigate the web, we'd all be running into each other, standing on top of buildings instead of walking into them, or walking into pawn shops thinking they're the bank.

I know at least 3 intelligent (off the computer) people who put full URLs into the Firefox Google search box and click the result instead of just putting it into the browser bar. Repeatedly. For years they've done this. Seriously.

Therefore, I second "proper A/B testing" wholeheartedly.

That said, if you can't afford testing, I would probably put them in both places. Your goal is to get them to click your important links; putting them in a second place, in my experience, will increase clicks. Just be sure that you only do this for a few links or you're going to get cluttered fast. But, if this is REALLY an important link or two, no real harm in putting it in two sections on a site with such a small main navigation.
posted by bender b rodriguez at 11:39 AM on October 16, 2012

I'm not entirely with the "Oh just A/B test it" folks here. A/B testing is great for testing tangibles (do more people click button X if it is red than if it is blue) -- it is not great for testing intangibles (is the navigation more confusing when we tangle up the site hierarchy, increase visual noise by adding links, etc). Sure, the B test with the duplicated nav item will get more clickthroughs to the duplicated nav item; you hardly need to do an A/B test to know that that will happen. But that doesn't really tell you anything useful about whether you've improved or degraded the navigability of your site by duplicating that nav item.

I am not saying A/B testing is bad. But I see way too many of my clients kneejerk their way to A/B testing that doesn't test what they think it tests, or set up inadvertently-rigged focus groups that invariably come to the conclusion they were hoping for in the first place. It's very easy to say "just try it both ways and see which is better"; it is often very difficult to set up a fair test that actually answers that question.

When I receive requests like this -- which is often -- my usual response is to guide the client to think of it as a task-oriented workflow. The user came to the site looking for a particular piece of information. If there is a plausible reason for the user to, having seen that piece of information, to want to jump to the other piece of information, even if that other thing logically fits better in some other section, then a crosslink is appropriate. If you're simply duplicating link A in section B because section B is popular and you want link A to be more popular, then that crosslink is probably not appropriate, because there's no use case that justifies it. There's no hard-and-fast rule here, but that's a good rule of thumb to start with: if it's good for the user, it's good; if it's only good for the company, it's questionable.

One useful compromise can be to make the crosslinks visibly different from the "real" site navigation -- rather than simply duplicating a navigation link in two sections, put your duplicate in a block of "see also" or "related" styled links at the end of a page, for example. That meets the crosslink need without confusing the site hierarchy -- and you can make those crosslinks bigger and more eyecatching without making the navigation look like Las Vegas.

All of this depends on your site hierarchy making sense in the first place, of course. Which it sounds like yours may not. It sounds like "Outreach" is one of your top level nav links, which is... problematic. As a user I would have no idea what's in there. "News" and "Events" combined into a single page is also a potential problem (from the user's point of view, those are separate tasks: news is if I want to find out more about what you do, events is if I want to find something to go to tonight. News is past, events are future.) "About Us" is almost always a one-page blurb about the company; it's rarely a good container for real content (and it usually belongs buried in the footer, not as a primary nav link, because nobody cares about that stuff.) This is where well-run user testing could help you. (Present nav structure. Ask user to find content X. See what they click. Present alternate nav structure. Ask user to find content X. See what they click. Etc.)
posted by ook at 12:12 PM on October 16, 2012 [5 favorites]

When I receive requests like this -- which is often -- my usual response is to guide the client to think of it as a task-oriented workflow.

I should add that thinking primarily in terms of user workflows, rather than content taxonomies, is really useful at all stages of site design, not just in deflecting Bad Management Ideas. Though it's also a good way to deflect Bad Management Ideas.
posted by ook at 12:15 PM on October 16, 2012 [1 favorite]

This is a really good opportunity to do a quick card sort study - requires next to no investment and you'll get some direction pretty quickly.
posted by chesty_a_arthur at 12:18 PM on October 16, 2012

I gotta say that while I totally agree with you, it is also true that nobody in the world clicks anything called News. What this has really highlighted is another issue: if you do not have promo spots on your website, it's a problem. Something like: Upcoming Events (list and link to next three events) and Highlights: Art Project in the sidebar would be worth suggesting.
posted by DarlingBri at 12:39 PM on October 16, 2012

Your bosses appear to be saying that providing information about some periodic or transient feature or event of your organization is an objective of the site. I'd suggest that a link in the secondary navigation to some sub section of the site, wherever and however, doesn't satisfy that use case.

As others have said, the specific implementation that address this need should be user-tested, but perhaps you could be considering calling out such information on the front page. People only explore navigation menus when they're motivated to, either by enthusiasm or frustration, not because anyone else hopes they will.
posted by normy at 5:33 AM on October 17, 2012

Many thanks for all your help, folks. Not only did you give me some excellent guidance for this situation, but you've given me a lot to think about for my next site redesign. You've taught me much about user testing, and I can see myself revisiting this info frequently.

Thanks again!
posted by kwaller at 8:34 AM on October 17, 2012

« Older Rebutting minus the full frontal assault?   |   StackOverflow hasn't answered iOS question. Can... Newer »
This thread is closed to new comments.