The long tail of internet of things things #thingscon

by Dries De Roeck on May 11, 2015

During #thingscon 2015 I talked about meaningful connected products. I wrote about some central concepts of meaning on this blog before, but consolidated some of that thinking in a new model : the long tail of internet of things things.


Note : this model is not validated in any way. I put it together in a rather arbitrary way in order to understand ‘meaning’ better within the connected product context.

Note 2 : I used an ongoing Studio Dott project (DIOTTO) to illustrate this model. The origin of the model resulted from previous research which was conducted at the University of Antwerp, department of Design Sciences. I plan to write an extended post on the case used and how it relates to ‘meaning’.


This is the ‘classic’ long tail power law graph. The graph shows that a small amount of popular products serve a substantial amount of people. As you move to the right of the graph, the popularity becomes lower – but however small the niche, a product can be found. An example of this can be found in music, Justin Bieber would be on the far left – an obscure black metal band would be on the right.


When you map connected products on the long tail, something interesting happens related to the type of use of this product. I was able to identify three categories. Firstly, products that have a predefined use. An example here is the Nest thermostat, it controls the temperate of your house. You can interact with it in a predefined way, the expectation of this product is very clear.


A second category are products with a personalised use. This means that they can adapt themselves to the person using them. This is often achieved using modules or building blocks. Little printer is (was) an example of this; it offers a list of publications which you can choose to print on regular intervals. As a user of such a product, you are free to do things within predefined boundaries.


The third category is when the use of a product is self defined. These type of products typically offer a platform for interpretation. There is no specific expected use, there are no paved pathways. You assign the function of this product in your own way. What these products do provide are cues for interaction, like the GoodNightLamp, it looks like a lamp … so maybe it is something domestic … but it doesn’t have to be that if you don’t want it to.


Using these three categories, the potential level of meaning rises exponentially as you move to the right of the graph. A connected thermostat has a rather distant ’emotional’ relationship with the person using it. Products that allow people to define the function themselves can go much deeper regarding meaning. For example, the GoodnightLamp is a product that focusses on social contact. The product is merely the mediator for emotions and thoughts that go much deeper than turning a lamp on and off.

This is what the model is about, in a nutshell. When I saw Claire Rowland‘s talk at thingscon later that day, it became clear to me that there probably a difference whether a connected product is a ‘tool’ or a ‘product’. Claire used a Belkin network controlled plug as an example of a tool. The use of the plug is not really predefined, it requires someone to define what it is used for. I do wonder if someone could get emotionally attached to a plug though. So perhaps I should distinguish between a tool and a product in a future iteration of the long tail of internet of things things.

Thingscon was an excellent place to present this model for the first time. If you’re interested in chatting about this, have reflections or thoughts – I’m very open for critical constructive discussions!


Interesting stuff. I think it is a useful framework for thinking about personalisation/appropriation/modification of technological (and other?) stuff. Your final comment on the tool-product distinction is indeed a crucial addition to the model, methinks.

I think you will also find a long tail when looking at tool/product adoption. Lots of people buy connected products and use them, while there is a small minority (the long tail) that makes the effort to use, appropriate, configure and/or program tools for their own purposes.

However, the product/tool distinction is still to much of a dichotomy – it would be more interesting to look at it as a continuum. In between products and tools, you can think of ‘hybrids’ with predefined uses, that can also be appropriated/modified to serve some kind of self-defined use.
I guess it would really get interesting if you can look at actual usage of such a tool/product hybrid. If such a product would exist (does it?), and you would map it to your long tail, you could expect to get a specific outcome:
– Most people would stick to the predefined use – the meaning of the product would then indeed remain ‘distant’.
– A minority of people would actively try to use the product as a tool, and modify it. In that case, the product would get a much richer, deeper meaning for those people.

In that case, you can start asking even more questions: do people actually succeed in ‘going beyond’ the immediately available, standardized ‘product use’, and move towards a more personal, self-defined use? Or do they stick to the (probably more prominent) standard use?
Can you design a product in a way that it guides the bulk of users towards a pre-defined use, but also encourages a minority of users to probe deeper, and create something personal?
If people succeed in creating something new with the product/tool hybrid, is it only a shallow modification of the readily available ‘predefined’ option (but a bit personalized), or are people actually able to create entirely new usages? How do these new usages differ (different context, reinterpreted affordances,…)?

One final question: how specific do you think this model (and its relation between self-defined use and deep meaning) is for IoT stuff?
It’s probably harder to map this on today’s ‘standard’ technology (with application/technology goals being very much predefined during design). However, outside the technological realm, you could probably also use this type of model on other stuff – like ‘analog’ objects. I can use cardboard plates to serve food on (like most people would), but I can also give it to my five-year-old, who will create some creative object-thing with it. In that case, my five-year-old is in the long tail of self-defined use, and also gives a deeper meaning to the cardboard plate.

by JayD on May 12, 2015 at 3:09 pm. #

[…] much driven by efficiency and functionality, whereas I believe we should strive to create products with emotion and meaning. – When using a term like connected products, I feel that I am […]

by Braided products and wickerwork. « Dream here on November 24, 2015 at 9:48 pm. #

Leave your comment


Required. Not published.

If you have one.