This is an interesting piece on the topic of designing services for ubiquitous, always-on environments, and I agree that this is an area of design that is important for us to better understand.

I was surprised to see you present service design and user experience design as dichotomous, or at least as divergent. This is not how I experience or practice either of these two areas of design - and our studio does operate under the 'banner' of both at times. A well-designed service will have an ideal 'customer experience' in mind.

I don't agree with the notion that 'user experience design' is imperialistic. It is holistic. And it is holistic in the same way that good design of any type is holistic - addressing a problem space in a broad, rather than a narrow frame. User experience design is distinctive in the way in which it frames the end result - the experience. But it would be wrong to say that user experience design is the only design discipline in which we are concerned with experiences; as it would be wrong to suggest that user experience design is the only holistic perspective.

I don't wish to downplay the importance of the theory and practice of information architecture in the design of complex services, but I don't agree with the characterisation of a service as a series of information exchanges. This *may* work as a framing for a digital or online service, but this is a very narrow characterisation of services generally. A service might be the production of a stage play; the service of customers at a restaurant; or the public transit system in London. To look at these through the lens of an 'information exchange' - although each does entail some exchange of information - would be to lose sight of the core offering of each. The exchange of information is one (very necessary) component of the service delivery, but by no means the most important.

If I think solely of the 'internet of things' and look at the services we might design as a result, the availability of information via the service is a shallow result. The value of the service will lie elsewhere, and in my opinion, not in a way that is best articulated as an 'information exchange'. And I mean this in the same sense that a dating web site could be described in terms of an information exchange, but this would be seriously missing the point.

In his presentation to the attendees at Interaction 10 in Savannah, Dan Hill showed us a number of concepts in which 'objects consume their own metadata'. In these concepts data and metadata were collected and relayed back to people physically present as a feedback loop. Again, however, this information feedback was not the point of the activity. Rather, the feedback became a way of exploring group dynamics and individual behaviour at an experiential and somewhat existential level as their own activity was aggregated and replayed.

This work raises the possibility of another framing of services as a series of interactions which aggregate over time into an experience, which in terms leads to or involves a change in behaviour. The practice of interaction design - designing for behaviour - can obviously provide benefit in this context.

Your point about visual thinking is well articulated. Folks such as Dan Roam and Dave Gray are helping to publicise the practice of visual thinking.

Your example is an interesting attempt at addressing the multi-channel aspect of a service. Another challenge we face is in articulating and visualising the transitions between channels; and the transitions between states within a channel. The fact that a transition occurs is one thing; the nature of that transition is another. The former is an exercise in communication; the latter in interaction design.

You can see the practice of visual thinking in any design studio, however - through the sketching and concept ideation techniques used every day. And you can see these techniques in use in many corporate brainstorming sessions facilitated by designers working in service design or 'design thinking' projects right around the world.

This isn't to invalidate your point but to highlight something I feel you either fell short of, or deliberately avoided: Designing the complex services you describe is best undertaken using a design process. This is, in effect, what you describe when you say: "we'll need teams that are multi-disciplinary and individuals who can help us think visually."

I also believe you need teams that are empathetic, can embrace ambiguity, and can collaboratively reach shared understanding and shared vision.

There is a clear role for design practices in the era you describe; there is a clear role for design practices in solving any complex problem space.



Thanks for your thoughtful response. You make some excellent points. On the difference (or lack thereof) between user experience and service design, my sense is that they are historically distinct disciplines that are in the process of converging. There are a few studios (such as yours) that have already integrated the perspectives and practices of both. But, that's the exception, not the rule. I can't speak for folks from the service design world, but I'm positive that many user experience professionals know very little about service design practices. And yet, those same UX designers have a lot to offer, particularly since much of the change that's reshaping service design is being driven by information technology. That's why we need to start catalyzing conversations and building bridges. Cheers!

Your point about information and its exchange touches on something I've been working out in my own head for quite some time (and in public, when others are kind enough to indulge me).
I get what you're saying, that focusing on the information exchange in these services & experiences is seriously missing the point ... but I only agree if we think of the information as the output or product of these things.

To me that's a bit like saying pottery isn't "about clay" because people use pots for lots of things besides storing and dispensing clay. Well no, in that sense it isn't about clay -- but understanding how to work with clay is central to making pottery.

That is -- these services are largely *made of* information, data, semantic frameworks, etc.
In the dating site, you might say that the point isn't information -- but information is precisely what people use to communicate. It's a place made entirely out of language -- so semantics are extremely important.
And with augmented-reality, ubicomp and the rest, this semantic layer is being intertwined with everything that used to be somewhat "information-free."

That, at least, is how I read Peter's commentary -- not that everything is an information input or delivery problem in the sense of "content" and its organization as conventionally conceived, but that the world is becoming infused with information as an interdependent element, and working with that element is a crucial part of the holistic approach you mention.

Thanks Andrew!

I agree with your characterization of my commentary :-)

Peter and Andrew already articulated the point of view me and Luca Rosati are trying to frame in our (upcoming) Pervasive IA book much better than I can possibly hope to do, so I'll leave that alone. :)

But I'd love to add one comment to what Steve says when he mentions the "multi-channel aspect of a service": most of what Peter describes is not actually multi-channel, but cross-channel, or cross-media, as the buzzword goes.

It might seem an exercise in semantics, but it's really an important distinction: while multi-channel works (basically) with one service being delivered through any number of channels in complete and discrete fashion, cross-media services can be experienced as a whole (if ever) only across a number of environments, media, and channels.

Incidentally, this is also why we believe IA plays an important (not the only) role, because while interactions might differ along the process, cognitive continuity relies mostly on how we experience information.


Thanks Andrea! I agree that it's valuable to distinguish the delivery of the same service via multiple channels (e.g., watch a movie on your TV or iPad) from the integration of services across channels (e.g., searching for a Redbox kiosk and reserving a movie for pickup via your iPhone). And, now that you've explained this to me three times, I may actually be able to keep them straight :-)


You seem to have understood my point about focus of the design activity. I'd hate to see ubicomp services narrowed to 'information problems', which is the fear I have when I read about services as 'information exchanges'. That exchange of information is needed to facilitate *some thing* - it's that thing on which we should be focusing our design activities.

At the same time, a deep understanding of information and the structure and exchange of information within ubiquitous, distributed environments is a critical skill that needs to permeate the design of these services.

I don't think we're in disagreement. I seem to be less sanguine about the successful design & delivery of such services when they're framed in an information-centric way.

PS: Andrea, thanks for the distinction between multi-channel and cross-channel.

Let me qualify this last sentence: "I seem to be less sanguine about the successful design & delivery of such services when they're framed in an information-centric way."

I should say: ... when they're *only* framed in an information-centric way.

I would be equally doubtful if the service was only viewed as a series of interactions, or as a series of 'moments'. It's the choice of a single frame of reference for something that is too complex to be understood from just one frame that concerns me.

So: each frame of reference is important; none is sufficient on its own.

Yes. Let's avoid the tyranny of the ONLY.


I really liked your post about service design, and agree with many of the comments regarding the distance between service design and customer experience. I think that what you have drawn is the gap between back-office and front office - the line of visibility as Shostack calls it.
I wrote a piece on my blog commenting this:

I like your experience map idea, specifically using visualisations as a tool for eliciting innovations. I think this is a key role that designers can play during the early stages of the design process. This gives the designer a triple role - workshop participant, workshop facilitator (through the visualisation tool) and documenter (by drawing ideas so as to form a common team understanding). This is an area that is becoming important for design, and one that design education is struggling to solve (but working on).

Thanks for a though provoking post.


Here's a link to Simon's interesting review of the article...

...I suspected that the bridge would be the most controversial element, in part because many folks have an overly narrow understanding of information architecture. This article is one small step towards addressing that problem. And, my intent is not to argue that IA is the only link between UX and SD, but rather to suggest, as Jeff Howard explains in Design for Service...

...that "it makes a lot of sense for information architecture to be part of the conversation when it comes to service design."

Nice Post Peter,

It is very profound and thought-provoking. I loved the fact that your post investigates about Service Design and User Experience Design and how Information Architecture acts as a bridge between them.

I work as Information Architect (or Principal Consultant) for a consultancy firm. Currently, I am creating a suite of application for a large financial service provider.

If you inquired about my job, you would know that I am creating services for my client. My client is paying us to create services and not User Experience. We feel that "experience" is an inherent component of a service and cannot be separated. It is like separating malleability from clay which is not possible. Even when you do not consciously design an experience for a service, you cannot separate it - it would be there anyway. As an IA or UX designer, my job is to ensure that the experience is always positive.

I feel that as an IA, my job is to find these intangible bits of information in my client's head and make them dance to my end user's tunes. If I did it successfully, UX wins else it loses.

I feel that IA is a bridge between Service Design and "Positive" or "Delightful" user experience instead of just UX. If it is not positive, it is not IA, it is just some random configuration.

Peter, a very nice, encompassing review article. It sets the stage for some very interesting conversations and more importantly, fusion practices in the future -- "back to the future" of design, if you will.

One issue remains unspoken: design's heavy debt to engineering, from whence it sprang (in part), and to which its methods pay enduring obeisance. To become truly holistic, design of whatever sort, whatever it's called, has to shuck its engineering constraints and deal with matters not subject to inspection under a microscope, on a network, or in a picture. Call it the noumenal, that which inspires. Awe. Fear. Hope. Tomorrow. Zen no-mind.

Combined with the phenomenal aspects of experience, that would be holistic.