Trajectories at the BBC – after two days*
Steve Benford describes Trajectories as offering a new way of thinking about the design of extended user experiences. I think the more ways we have to think about the design of experiences the better. Structured methods for interrogating the result of our designs forces us to move our focus from the design of objects to the design of experiences. I think this is the real potential in trajectories.
From interface design to experience design
Good design has never stopped at the object, it’s always considered use. But the language we use reveals our priorities. I think as I’ve matured as a user experience architect my concern has shifted from interfaces, through interactions towards considering experiences. I think this maps a move towards a richer set of considerations. When I was focused on interface, my concerns were almost always spatial – I was concerned with the object. Interaction design forces a broader set of concerns as you consider the object in use. But it wasn’t until I started to think about ‘experience design’ that I fully embraced the notion that each user experience has a temporal dimension as well.
One reason why it’s harder and harder to design interfaces is that so much more is dynamic. It used to be that devices did one thing well. Now we have software replicating what would traditionally have been hardware. Dynamic publishing, aggregations, redundancy in design – all these things mean that it is less likely that our designed experience will ever really exist.
I think that designed experiences (in the language of trajectories, ‘canonical trajectories’) really describe an envelope of affordance. Maybe this is a natural information architect approach to design. Architects give instructions. And I like to think that good experience design is about establishing the boundaries within which experiences are meaningful and satisfying. I think trajectories help us do this.
Canonical trajectories – the designed experience
Trajectories are described through sketching and stories. One of the benefits of the technique is that it forces you to describe your design as an experience. Canonical trajectories describe the story of the ideal user experience. Most trajectory sketches depict an experience as a line. Because we’re used to reading from left to right, most people end up using an x-axis to describe the flow of the experience over time. The y-axis can vary – space, immersion, attention, proximity – take your pick.
Beginning with the simplest representation (or story) of the experience is helpful. You can focus on the heart of the idea. I then think you might be able to build up other trajectories that are still designed, but differ in some way. This might be the way to discover the boundaries of the designed experience.
Steve uses participant trajectories to describe the actual experience of the user. I think the effectiveness of this concept varies depending on whether you’re using trajectories as a design tool for a new concept, a way of iterating on a current product or to evaluate a past project.
What I like about participant trajectories is that they reveal the fact that interactivity ensures that there will always be divergence from any single, canonical path. I tend to think that all experience are the result of co-creation – that users actuate a design and so the ‘designed object’ can never really be perceived, it’s always a co-created version that is experienced. That might be a bit philosophical, but I’m interested to see whether a shift from thinking about ‘designed experiences’ and ‘experienced trajectories’ can improve the design process. So I might prefer the label ‘experienced trajectories’…
This is probably where I’ve veered away from Steve’s model most – so you might want to check out the official idea of how canonical and participant trajectories relate to each other on Steve’s website.
Steve’s model encourages you to consider a third type of trajectory – historic trajectories. I think this is another area that will vary depending on the type of experience you’re designing. I think the idea of a ‘shareable’ thing being the result of all experiences is useful. Encouraging you to think whether this shareable thing is a memory or an artefact, an object in a dashboard, binder or archive, a tweet, status update or email is definitely helpful and a stage in the design process that could otherwise go overlooked. I’m keen to explore how this type of trajectory relates to different types of experiences.
Steve also has a number of features of experience that can be used to interrogate the shape and nature of a trajectory. These include transitions and encounters…
I found this one of the richest areas to augment and annotate your trajectory sketches. Transitions are moments within the experiences where ‘continuity’ is at risk. I’ve written before about the challenges of creating inherent motivation when designing learning content, and generally I think the moments of risk are where you’ll find some of the most interesting challenges. Steve has described several type of transition:
Beginnings – I think these are interesting as they force you to think about whether your ‘canonical trajectory’ reaches back far enough. How could you enhance the entire experience by reconsidering where and what you consider as the beginning?
Role – This is probably my favourite type of transition. I’ve written before about the different types of search behaviour that users exhibit as they move around the web. I like to think about the shifts in motivation and intention that an experience can support and how through orchestration we can nudge users towards specific desired roles and behaviours.
Interface – This describes changes in the way the user/participant is interacting. It encourages you to consider the tools the participant will be using. I think this will be increasingly important for the BBC as we start to think about cross-channel experiences and consider the responsiveness, persistence and pervasiveness of our experiences and information architecture.
Temporal episodes – allows you to focus and describe the breaks and re-engagement with experiences. Again, I think for the BBC the idea of cross channel experiences and the notion of affording an experience of one-BBC across multiple episodes is an interesting area.
Physical-virtual traversals – is a jargon-heavy label for concerns around the levels of immersion within an experience. Particularly suited to real-world experiences, this type of thinking will probably help with ideas about augmenting content through second-screen experiences or potentially the use of linked-data driven onward journey recommendations.
Access to physical resources – providing a chance to consider where scarcity might be an issue. I suspect this could be extended to considering access to all types of limited resource, not just physical. I think we need to ensure that all the interactions we encourage, generate value for the user. Often this value generation is based on transforming limited resources into user value. For example, moderating comments might not always be considered a physical resource, but if we’re not able to support ‘user commenting’ fully, experiences and products could be undermined.
Seams – describe weak points in technology. Considering the technical barriers that could undermine an experience is vital. Caching and off-line access to content is an obvious example of where considering seams can help ensure that participants remain inside our optimal designed experiences.
Endings – sometimes things end. Like this list.
Considering encounters enables you to spend time considering the social aspect of the experience you’re designing. Concepts like isolation might allow you to play with levels of immersion. Pacing might be required to ensure that cohorts remain in sync. I think that layering ‘encounters’ thinking on the top of ‘experienced trajectories’ that belong to individual participants will help to reveal the threats and opportunities presented by the inherent social nature of almost all experiences.
I like trajectories. I think there is some work to do in turning the theory into a practical set of tools and strategies to make it more useful. I like the way that trajectories encourages you to repeatedly retell the story of your designed experience. This iteration process will improve any idea, but formalising a process might help to ensure more reliability.
I’m going to start to use trajectories to think about some of the designs I work on, and I think the process I’ll adopt will be roughly like this:
- Sketch and tell the story of the optimal designed experience.
- Use the concepts of transitions and encounters to interrogate the designed experience and identify opportunity for refinement.
- Transitions and encounters should also reveal the possible divergent paths from the optimal designed experience. Describe the supported boundaries around this ideal designed experience to create an envelope into which each of these should fit.
This is just an initial idea of a process. But I think trajectories will be most useful when it’s a facilitated process which gives creative teams just enough structure to explore their ideas creatively.
One of my favourite quotes is:
How do I know what I think until I see what I say.
Because trajectories forces you to describe experiences, it has a similar effect in making an idea tangible. ‘Trajectorizing’ an idea forces you to play out multiple thought experiments. It reminded me a bit of De Bono’s Thinking Hats which encourage you to analyse and play different roles in creative discussion. Trajectorizing makes you layer on the temporal reality of experience over the ‘idea’ of the object, so it shifts us from simple interaction design to considering complex experiences. Trajectories encourage us to consider how people interact with things in the real world. Life is interactive – it has a spatial and temporal aspect. Trajectories are one way to force us to acknowledge this as we design experiences.
* I’ve written this post in a frenzied state of evangelism after a two day workshop with Professor Steve Benford at the Mixed Reality Lab at Nottingham University, so it’s likely I have got over-excited and confused about Steve’s model. The above was written from my notes and things I’ve been thinking abut recently. It’s likely that after playing some more, my ideas will change – in which case I’ll write about trajectories again. I urge you to visit Steve’s site for an official view of what trajectorizing means – I’ve tried to avoid using the ‘z’ version for personal reasons. As always, I’m more than happy to respond to questions and comments.