The limitations of the double diamond
Summary: The Double Diamond is good at communicating some important aspects of design. But it has limitations and using it brings complications — being aware of what these are can help you use it more effectively.
I have a love/hate relationship with the Double Diamond. I love the way it communicates two key ideas about design — (1) that design is as much about defining and creating boundaries around problems as it is about creating solutions and (2) design benefits from a dialogue between thinking styles, in this case Divergent creation/discovery and Convergent analysis, synthesis and decision making. These can be relatively hard ideas to communicate and grasp. The Double Diamond does a good job of introducing them. And it can remind designers that these fundamentals can be things to consider when we run into problems during a project. I love these aspects of the Double Diamond.
But I hate it when the Double Diamond is overused and misused, because it undermines designers and design managers and sets false expectations for those we’re talking to. This article isn’t a polemic against the Double Diamond. It is an exploration of what it can be used for and its limitations.
Definitions: what is the double diamond?
Let’s clear up what I’m talking about when I say, “double diamond,” so that we’re all on the same page. I’m talking about this:
This is my re-drawing of the “Double Diamond,” a model/diagram used to describe the design process. Disciplines often adopt abstract models to communicate across disciplines. The Double Diamond was developed by the UK Design Council around 2003 as a way for the organisation to standardise the way it described the design process. Since then, it’s seen widespread adoption. The model was influenced by academic writing, as well as the practical experience of the team at the Design Council. The work of Banathy (1996) is often cited as an inspiration for the diamond (or kite) shape and the discussion of divergent and convergent thinking (Taylor 2021).
What the double diamond does well
In the introduction I mentioned the two things I think the Double Diamond does well. It makes the point that designers make progress when they don’t only focus on solutions (what and how questions) but also interrogate, articulate and manipulate “problems” (why questions).
Design can ‘problematise’ situations. Skilled designers construct a frame around a situation that transforms it into a problem or opportunity to experiment within. This framing creates boundaries and allows teams to settle on and describe their purpose. It creates alignment within teams. And by introducing constraints, we bring order and can attempt to simplify the complexity of problems so that intentional intervention and experimentation can be accomplished.
This might seem like needless intellectualising. In commercial settings non-designers might dismiss this element of design: “We’ve written the brief or the invitation to tender, that’s the problem. We’re paying you to solve it” clients might say. But identifying and describing the right problem or opportunity is hard. It takes just as much insight and creativity as creating the solution. Lots of the time the real value in design is interrogating the problem. As we see in the JTBD approach, sometimes the most valuable answer that design can provide is to ask the right question. Otherwise, we just end up making faster horses. [Henry Ford is said to have once remarked “If I had asked people what they wanted, they would have said faster horses.”]
So, the Double Diamond is good at introducing this idea to non-designers — that design is more than just solutionising, it’s also about “problematising” to reveal the biggest opportunities. Admittedly, the Double Diamond can run into a little trouble here. The central point of convergence in the Double Diamond can suggest a point of fixed meaning. Fixing things in complex situations is useful and necessary — it gives you a foundation or point of leverage against which to hypothesise and experiment. But the advantage of design is really the ability to fix and unfix or frame and reframe situations. As Hobbs and Fenn describe, design encourages the creation of a “problem:solution ecology” which sees the co-evolution of problems and solutions to deliver the most efficient and effective outcomes. The Double Diamond might make it look a little like problem framing and solutionising are separate activities. But they’re not. Skilled designers know that the real value is in facilitating the dialogue between the two diamonds or sides of the problem=solution equation. This dialogue is helped by the other big idea in the Double Diamond — divergence and convergence.
Great design balances the use of different mindsets and thinking styles. Great design is neurodivergent by default. The Double Diamond gives us a chance to talk about divergence and convergence.
Divergence introduces fresh possibilities. Convergence focuses. The more you diverge the more variables you can introduce to both problem definition and potential solutions — you create more possibilities. But divergence also introduces more variable, uncertainty and instability. When we go too broad or when we disregard the reality of the frame we’ve created, which we sometimes do in complex systems, this complexity makes our solution less stable and resilient. Convergence constrains, decides and creates fixed points of leverage. Skilled navigation between divergence and convergence helps us land in a goldilocks zone where our problems and solutions don’t just match, but they balance each other out because they’re just the right size.
I might be being generous to the Double Diamond to suggest that two squares at a 45-degree angle communicate all this… but it gives us a way into understanding and communicating these important aspects of design. However, the Double Diamond does come with limitations.
The limitation of the double diamond
Personally, I like abstraction. Abstraction involves combining specific instances into more general categories. The Double Diamond is so successful because the team at the Design Council were able to take their experiences of successful design projects, look for the features that made them successful and form an abstracted/generalised description. Within this abstraction the two central ideas of problem/solutions and divergence/convergence shine out with just the right amount of clarity and weight. But, there are issues:
It’s too abstract to be useful practically, during design projects — this has consequences…
Abstractions become less useful when you get to specific questions. Google’s definition of abstraction is “the quality of dealing with ideas rather than events.” And that’s the most perfect critique I can think of for why the Double Diamond falls short as soon as you get into process questions about design. It’s too abstract to inform or standardise practice.
The Double Diamond usually operates at two levels of abstraction. Firstly, it refers to “problem definition” and “solution design” as being two parts (two diamonds) of design. It communicates the difference between “designing the right thing vs. designing the thing right”. It also introduces four less abstract labels: Discover, Define, Develop, Deliver. These provide more specific descriptions of the type of thinking required during design. However, while the admittedly alliterative labels are made more memorable, there’s no description of when to switch between these goals. There’s an implication of a linear relationship between the stages. And as I’ve argued above, the most successful design projects see a dialogue of back and forth, rather than a linear progression. The Double Diamond misses any depiction of iteration — a key element of design.
So, it can be misleading…
This is the challenge when you choose a model, diagram or abstract shape to represent complex ideas. A model is a construct to help us understand. Design is complex, emergent, and unpredictable. It often requires multiple cycles of definition, (re)orientation and (re)framing to address the disorientation that occurs as opportunities and possibilities emerge. The Double Diamond does a poor job of representing and communicating this. Jones talks about the difference between “black box” and “glass box” processes — where the transparency of “glass box” methods aids understanding (J. C. Jones 1992). Too often, the Double Diamond becomes a black box masquerading as a glass box. All the art is obscured behind a façade of ordered science and simplification. Design is sometimes a little messy. The seemingly disordered mess is often the result of creative leaps of imagination that generate genuine innovation. The high levels of ambiguity and non-linearity experienced in design (especially in discovery and discontinuous innovation) render the straight-lined orderliness of the Double Diamond misleading. This makes it more difficult for designers and non-designers to maintain confidence during design and undermines trust and confidence of stakeholders when the reality of the design they see doesn’t match the geometric perfection of the Double Diamond.
I’m not sugesting we all start using Danien Newman’s Design Squiggle — but emotionally it’s probably closer to what it feels like to design. https://thedesignsquiggle.com/
There’s no specific mention of creativity or abductive reasoning
The Double Diamond boils thinking styles down to divergent and convergent. As I’ve said, this is a useful idea to communicate. But it misses out other forms of thinking and reasoning. In particular, it doesn’t help you talk about abductive reasoning, a strength of design when compared to other approaches to solving problems and generating value. The ability of design to address ill-structured or “wicked problems” which are high in novelty, ambiguity and complexity is a primary strength.
Where are the people, and how do you make decisions about impact and outcomes?
The Double Diamond also doesn’t really mention empathy or people. And if that’s not a problem for you because you’re not into anthropocentrism, it doesn’t really hint at how to generate insights and inspiration or how to weigh alternatives either. Design is a way to generate possibilities. Without providing a view on who and how they weigh alternatives, you could criticise it for being ethically ambivalent. Missing out people, needs or empathy from the depiction of design feels like an oversight… incidentally one that an alternative depiction, the d.school model addresses.
On ‘decisions’ it gives us convergence as a thinking style and hints at the key decision being the design brief which connects the diamonds (as I’ve argued, that’s something to watch out for in itself). And you might say that decision-making frameworks belong more to a “methods bank” rather than a model like the double diamond. But it would at least be nice to open up some consideration of how designers should manage the transition between divergence and convergence.
Donald Schon gives us some language to describe what the skilled back and forth looks like. He describes reflection-in-action (Schon 1984), when practitioners work backwards and forwards manipulating the problem framing/definition to achieve better outcomes. He also describes ‘reflection-on-action’, where we do something and then step out of the activity to evaluate the outcome. For example, in “Sprints” designers are encouraged to work quickly towards testing prototypes with users to generate insight, feedback and data and then reflect to take into subsequent iterations (Knapp, Zeratsky, and Kowitz 2016). Reflection-on-action can also encourage designers to consider the practicality of implementation. This ensures we not only design what’s desirable but that it’s also viable and feasible.
Finding points in design to make decisions is important. We might be constantly checking in and adjusting. Often, we’ll also need to consciously pause and return to the problem statement — which is often a human need, and check with real people whether we’re succeeding. These are important things to think about and talk about when communicating what design is, how it works and how it generates value.
It doesn’t speak to the importance of “making” and visualising
The Double Diamond makes no reference to what I consider the superpower of design — translation. Through sketches, descriptions, iterations and prototypes we transform ideas into sharable experiences. Admittedly, the Double Diamond has the ‘Develop’ stage which might cover this. But for me that isn’t an explicit enough label to point at this superpower.
Design results in tangible experiments that generate either insight, value or both. Design invents new possibilities, so creating artefacts and “embodying” ideas is indispensable. The decision of when to “visualise” an idea is missing from the Double Diamond. It’s an important stage for every discipline to participate in as early as possible. Bringing visualisation to the design process builds a deeper shared understanding of ideas and enables more detailed critique and faster progress to be made. I understand that models often resort of acronyms or alliteration for memorability… but I think the 4Ds in the Double Diamond do design a disservice by missing this important aspect. We can all do alliteration, but sometimes its at the expense of precision and meaning.
It doesn’t describe the contexts which enable good design — multidisciplinary, high levels of psychological safety…
Experienced designers know that certain environments are more conducive to better work. Not only is the Double Diamond abstract, it’s also a little sterile. Diamonds are shiny. But they’re also hard. Maybe design needs a shape with softer edges or something with a little more humanity.
As well as mentioning the people you design for, it’s useful to consider and mention the people you design with, when you talk about design. Design is a contact sport — it involves lots of communication and collaboration. The Design Council’s iteration on the Double Diamond, their Framework for Innovation depicts some of this surrounding context — which demonstrates that it might be possible to build more into a model to describe design and introduce more of these important elements.
What are the alternatives?
You might not find a single alternative to all of these issues — and each set of considerations might be more or less important to you depending on your context. But there are alternatives. I’ve written about alternatives to the Double Diamond in another article — and hopefully this affectionate critique has introduced you to the limitations of this venerable model and helped you consider other aspects of design you might need to communicate separately if you are using the Double Diamond.
