Do you use custom navigation in your courses?

Just saw this comment on Tom's blog and thought it would make a good topic for anyone building courses with custom players and on-slide navigation.


Is navigational back/next buttons best practice? Or is navigational hotspots created within the defined imaging of the module better practice? 

I’m very interested on the end user experience and I find the over use of back/next just not a creative solution for navigating through content.

It’s a fine line between personal preference and what is navigational best practice these days? Thoughts…?

10 Replies
Steve Flowers

A couple of years back we attempted to build an experience with the premise that we wouldn't use the back and next buttons. In theory, this was awesome. I think it came out nicely and used contextual cues for forward progression through a set of experience platforms and activities. 

Problem was... everyone was already used to the back and next buttons. The absence of these affordances was really distracting to folks. The "nav on tracks" enabled folks to go backwards and forwards. They didn't need to mentally track where they were (even though we did add clear location indicators and an easy nav jump to the menu). Space and place in a sea of content is important. Personally, I hate being a slave to the next button. But I'm not my users - and I will surrender to users every time in that battle

I still think back / next is an easy out and an over conditioned packaging default. But we listen to our users during usability testing. I think we'll still keep the concept of "no next" in our back pocket for future use within a slide based presentation that plays more like a game (with levels of mastery). That should defeat the benefit of the back / next navigation features.

We are experimenting with other packaging methods like a topic based presentation in a single page that weaves media inline with article based presentation. I like the promise of this for folks like me and for mobile platform deployment. This is a layered approach that employs progressive enhancement to stack media on top of task driven content. I think it's got promise, but I was confident that folks would prefer a contextual flow to the dead soul of a back / next pair. I've been wrong before

James Starr

I'm the culprit for raising all of these best practices conversations. I recently started at a new company and since I am now chest deep in onlione course creation, I've decided to try to implement all of the great ideas I've learned in the tutorials, posts, and blogs related to this forum.

However, most of my implementations haven't been well received.

For example, in my latest course:

  • I had things to select on the screen to interact with the course. I removed the next/back buttons from the bottom navigation to force the user to interact. Everyone thought that would be confusing to end users.
  • I created blurred images of government office official seals so that when you clicked on them, the actual clear image of the seal would then appear alogn with the name of the office (using simple hyperlinking of slides). Again, I removed the back/next buttons. Not only did they not like the navigation again, but they disliked the blurred image because they did not know if it was supposed to be blurred in the first place.

I don't necessarily agree with everyone but my hands are tied. I'm sure some people can appreciate that notion.

Natalia Mueller

James- I'm a big fan of ppt hyperlinks so I understand the need to get the enduser to interact with the screen directly. Are you familiar with the Articulate functionality to lock the controls on a single slide? I use it very sparingly, but that's typically what I do so the hyperlinks are used and the course progresses correctly. Then the player buttons are still there for the rest of the course.

The first time I added hyperlinks in one of my courses, I got similar feedback from management. They thought the endusers would be confused. I disagree, but that's another topic.  The compromise I came up with was give them visual instructions (Icon, Click here, etc) as well as telling them in the audio. Management liked the interaction, as long as it had a true purpose and very clear instructions. 

Rebecca Fleisch Cordeiro


So, I guess I'd like to ask a question about this. Two of Jacob Neilsen's tenets around usability involve making a match between the system and the real world and memorablity. Standard previous and next buttons match to the real world, since we're used to seeing them on other electronic equipment, even going back to tape recorders and on through current hardware. Memorablity refers to re-establishing proficiency when returning to a design after a period of time. Which again (at least in my mind) speaks to users seeing buttons in a learning design and remembering them from other situations when they needed to go back or move forward. This could be using software or hardware.

How does one reconcile "navigational hotspots created within the defined imaging of the module" with those usability tenets? Would this mean back and next would/could look somewhat different in different designs?

Interesting discussion. Looking forward to responses.

Steve Flowers

Hi Rebecca,

I think those usability tenets apply to real world elements with the same mental model AND within Jacob's definition of user experience within information systems. I wouldn't bind my decision-making models to the Jacob Nielsen tenets. They are useful tools and I have and will use them, just keep in mind that it's one view and a relatively narrow context. Even so, let's give it a go. > In the case of most e-learning on the market, the model manifests as a conveyer belt that presents one small chunk at a time. This linear model is most like a video experience model and relates to the common interfaces, such as VCR buttons, you referenced above. 

I don't think this is necessarily a bad thing in some cases. Linear works when linear works. But as a default, this narrow design patterning limits the potential of the medium and paints us into a content-centric corner. We should be careful not to condition our users to expect a simple linear presentation of content in every case (oops, too late). Simple to use is a trade-off used to counter complexity. Sometimes complexity is necessary. Both simplicity and complexity should be used judiciously and almost always in balance. Eliminating complexity dumbs down an experience and conditions folks for a spoon feeding. This isn't healthy and it's patronizing to users that just want to get things done and aren't afraid to spend a little time thinking.

The navigation elements provide access cues for the affordances (what folks can actually do within the system). Good cues clearly reference a mental model that is already possessed or can be reasonably assimilated by the user. So in the case of a conveyer belt or linear model of presentation, the back / next become the primary and sole affordances at a very simple level. Looking at other mediums, we see similarities and differences in the affordances offered.

  • Web browser. Lets you pick your destination using multiple access methods. Lets you go back and forward through your selected history using fairly standard navigation controls. This works for this medium. 
  • Video games. You use the game mechanics to progress. Levels are normally saved and a "place save" affordance is usually provided that gets you close to where you left off. But there's no back button to rewind through the echoes of the experience at a granular level.
  • Software applications. You choose the tools and functions of the system. There is no back button.
  • Real work. You choose how you behave in this complex system. There is no back button. The next button is invisible.

I think we can agree that e-learning doesn't necessarily fit into these molds (if it should fit directly into any mold at all). The closest mold I can see e-learning *should* fit into is real work. That said, I don't think there is, or should be, a rule that says a back and next button should always be included.

There are many ways to think about building an experience. One metaphor is a static, linear corridor of content with next and back buttons to ratchet one move forward or one move backward through a segment of content. Another is thinking of an experience like floors in a building. Progression through the floors of the experience could be controlled through a variety of mechanisms (unlocking or open) that mirror the skill progression of the participant. Using this metaphor, you see two distinct sets of navigation opportunities. One on the inside of the activity and one at the *meta* level of the experience. Within the activity you might see a simulation, a short presentation, a set of slides or a single set of directions that say "Get away from the computer, go explore the real world". 

Use a standard next cue when the affordance is important. Use the standard back cue when the affordance is important. But I wouldn't be afraid to use other exploration metaphors, cues, and affordances provided that the mental model is available or can reasonably be assimilated. 

What's warmer and more contextual (1) or (2)?

(1) Select Next to continue.                             { Next }

(2) Alright! Next we'll hear from Joe about how he fixed a wicked problem with the roller assembly. Ready?

They both use recognizable affordance cues. (1) is blind and informal. It serves convenience and IS more consistent. (2) can be integrated into context within the presentation of the experience. This method can *also* employ recognizable visual cues to emphasize a call to action (trigger the mental model and encourage the user to select an action without confusion).

Here's an article that describes some of what we're hedging about in this discussion. The author describes an experience that is idiomatic. In other words, once he figured out how it worked operating the experience wasn't a problem. In the article he also references a really great book About Face 3: The Essentials of Interaction Design. The author of this book divides useful interfaces into three categories.

  • Instinctive - this category is largely about attention processing and the things we pay attention to.
  • Intuitive - this category deals with mental comparisons between functions of new things and things we already know.
  • Idiomatic - this category deals with a level of "figuring it out" that is easy to do when you've done it once.

The standard use of the back and next buttons isn't a terrible practice. This brings a globally recognizable consistency to navigation. But restricting navigation to these elements also trains folks not to expect higher level intuitive interface elements and doesn't help develop the skills or practice of "just try it, you'll figure it out easily". 

Each user test we run, we test a navigation affordance cue that isn't familiar. In some audience groups we have no problems folks will make the leap across the intuitive chasm and most will discover that once tried, it's not really that unnatural at all. In others... we see folks that have been conditioned to wait for explicit instructions on exactly what to do with a relatively simple medium. In this group, we see people stare at the screen in confusion even when clear instructions are presented and a clearly marked control is inviting them to take action.

Thankfully, this is the minority of my audience. I design for the majority of folks. The 2% at the bottom.... I'm not so worried about confusing a few conditioned or less capable folks (not talking about accessibility - this group is unwilling to venture outside of a very narrow mental model) as I am about providing a relevant, activity-centered, and sufficiently challenging experience to the other 98%. When the majority provides feedback, we listen very carefully.


Phil Mayor

Steve as usual talks a lot of sense, a lot of the courses we have developed use custom skins with back and forward buttons mainly because this is what people exepect, we have also built in a lot of custom flash navigations to allow custom branching (we lock out the players controls at this point) often this gives three options with returning to a main menu when all three are viewed the navigation is unlocked again.

I much prefer the idea of dropping a user into an environment that is very intuitiive with no visible navigation, and have probably spent the last year developing different menu structures and controls, and then doing user testing to see how they respond.

I often think when we develop we dont realsie the majority of our users will be able to navigate our courses and will not need Next buttons.  The work metaphor Steve talks about is very powerful, at work we are often dropped into situations and have to work our way out.  This is what I am trying to achieve by removing the standard navigation.

I often look at OS designs for cues for user interface design, the really powerful one is iOS if you have ever given one of these devices to a 3 yr old just watch and be amazed, they pick it up immediatly.  I do like Windows Metro (but do have reservations about this one)

Just my ten pence worth

Rebecca Fleisch Cordeiro

Phil and Steve, thank you so much for taking so much time contributing to this discussion. Phil, I'm just now going to link to the articles, so thanks for providing them. I''m always looking to learn more/read more about learning. Now, to respond to a couple of your points:

Re: Use a standard next cue when the affordance is important. Use the standard back cue when the affordance is important. But I wouldn't be afraid to use other exploration metaphors, cues, and affordances provided that the mental model is available or can reasonably be assimilated.
But restricting navigation to these elements also trains folks not to expect higher level intuitive interface elements and doesn't help develop the skills or practice of "just try it, you'll figure it out easily".

Sorry if I gave the impression that I believe we should RESTRICT navigation to these standard buttons in an elearning interface. No way did I mean to imply that. Certainly they're of no (or little) relevance in an experience that isn't linear and is designed to allow the user to explore in any direction. When the design IS linear, I see no reason not to provide them, as you say, along with other navigation elements. And, as others have described in various posts, providing a bit of instruction about the navigation, say through a tab in Presenter, can educate Learners to the interface navigation.

I agree Mr. Nielsen's model isn't the only one. I am partial because it speaks strongly to my many years of experience teaching computers to adult Learners. When I'm able to make analogies to the real world, those who have been at first confused by what seems to be a very foreign experience have had many "aha" moments. (For example, I've compared minimizing applications to the taskbar and off the desktop to taking something off a physical desktop and throwing it on the floor, because there just isn't enough room for everything. And when one is ready to go back to it, they restore it to the physical desktop, just as they can restore from the taskbar.) As Learners become more confident and discover the system is something they can relate to, their stress level decreases and the cognitive energy that was put toward de-stressing is now available for learning.

Steve, doing the user testing is of course the most important piece. Not just designing something that we prefer, but something that end users, our Learners, find intuitive/sensible. Regarding Metro, I'm liking it, but only time will tell. And regarding 3-year olds, yes, they can go to town! Happily for them, they don't have to unlearn many preconceptions that the world has "taught" them.

It's great to have this discussion and keep all these perspectives in mind when designing training. Thanks!

Rebecca Fleisch Cordeiro

Hi again,

First, Steve, thanks for this link Here's an article . It certainly speaks to a situation we've all been in with poorly designed instructions, and images that are not intuitive.

Second, I'm not sure how I came across this Tom Kuhlman blog post from May 2011: More than a dozen ways to navigate an elearning course, (, but it totally speaks to this discussion. In most of the examples, standard (or very close to standard) navigation buttons are included along with other ways to interact with the lesson. In cases where the navigation design is very different, navigation instructions are provided. Probably wouldn't have come across it without this discussion, so thanks to all, including you, David, for introducing it in the first place.

Rebecca Fleisch Cordeiro

Wow. What timing. This on navigation from Tom Kuhlman's post "How Google's Designers Can Help You Build Better eLearning"


Build Consistent Navigation

People generally expect that things work the way they expect them to. Navigation is one of those types of things. 

  • Don’t deviate from what people expect. Navigation should be consistent. Generally people look over the screen using a Z or F pattern. It makes sense that the back and forward navigation is at the bottom right corner since it isn’t critical information and one of the last things we need to see on that screen.
  • If you do deviate from expected norms, have a reason to do so. Don’t just haphazardly move things around the screen. I’ve seen buttons turned sideways and in different places just because there was no room on the slide or the designer wanted it to look different. This is a frustrating experience. When people start focusing on the navigation and not the instructional content, you’ve probably failed.