The 5 Phases of Design Thinking: An Overview
“Beware jargon. It usually hides ignorance and carries little knowledge.” — Bene Gesserit Coda
‘Design thinking’ is a design methodology not only relevant to user experience design, but potentially to the design of everything — from software to automobiles to everyday things. Design thinking provides a framework by which users can be empathised with, a problem can be clearly identified, and solutions can be explored and developed in a way which is structured and meets the needs of users. Design thinking is used by designers everywhere and is even the user experience design framework of choice in major corporations such as Google and Microsoft.
There are a few different implementations of design thinking, but traditionally it consists of five phases — Empathise, Define, Ideate, Prototype, Test. Different UX professionals will have different specialisations (and therefore different responsibilities during the UX design process), but generally speaking, the entire UX design lifecycle of a product can be represented in these five stages.
So, what does each step mean, and why is it important to the process?
Empathise
‘Empathise’ is the first step to design thinking, and arguably the most important when it comes to user experience design. It is the stage at which a designer seeks to understand their prospective user base through a series of activities surrounding researching and analysing their habits and expectations to begin to understand their wants, needs, and challenges.
This is the phase at which most user research will take place. UX researchers may start with a round of secondary research to provide a theoretical background to their project — this may take the form of browsing journal articles, academic papers, or even less formal sources like blog posts and magazine articles. Sometimes research doesn’t have to be about hard statistics and formal research, but instead insight can be gleaned from looking at opinions and trends in popular media.
Following this, primary research can be conducted to begin to get a sense of user opinions and habits. Methods such as conducting surveys and interviews are used to gather data on target user groups — this might mean seeking out participants with specific interests, who are based in a specific geographical location, or some other defining criteria. There is a lot of science to research design when it comes to primary research of this nature, so it is useful to have foundational understandings of statistics and psychology.
Define
The ‘define’ stage goes hand-in-hand with the previous one. At this stage, designers combine and evaluate all of their research and use it to attempt to observe specific challenges their users might face — it forces designers to examine the problem from a more human-centric perspective.
It’s at this stage that designers will begin to generate user personas and user empathy maps, examine user journeys, and so on. These activities are associated with more specifically identifying and representing the needs of larger groups of users, such as those with particular characteristics or challenges in common.
User personas allow designers to generate a theoretical user which represents a specific demographic or set of common characteristics, so that they are better able to think about the challenges those user groups might encounter. For example, users with disabilities may have difficulty using specific features of a product, like a wheelchair user trying to manually operate a heavy door.
Ideate
The ideation phase is very popular with creatives. It’s at this stage that designers use the information gleaned from the previous two phases to begin to generate ideas for solution designs. Ideation involves gathering together the entire design team to examine different viewpoints and opinions on the user goals, frustrations, and challenges identified in the define stage to inform the direction of the actual design.
It is at this stage that the design team engages in ideation activities such as brainstorming to generate ideas for a solution — at this stage, the focus is on generating as many ideas as possible rather than trying to come right out of the gate with quality ideas right away. In fact, a popular strategy when approaching a design is to start with the worst conceivable design — this is a great way to identify potential design pitfalls and ensure that they don’t make it into later ideas.
Ideation sessions should be fixed in length, include the whole design team in an appropriate environment, and should also consider (and, if viable, include) other stakeholders. The design team alone usually cannot perfectly capture all of the needs of a client or their users, so it helps to have stakeholders on hand to shoot down any completely non-viable ideas before they can make it to the prototyping phase.
Perhaps most importantly for ideation is the reservation of judgment. Designers should be unafraid to pitch their ideas, and if another designer suggests that an idea is non-viable, they should be ready to explain why. Again, this phase is about quantity over quality, and only through exploration of many ideas can an optimal design solution eventually be developed.
Prototype
Prototyping is about taking some of the best ideas generated in the ideation phase and turning them into something real and tangible. Prototyping allows design teams to present clients and stakeholders with an experience simulating parts or all of a product, so that they are able to provide more specific and valuable feedback.
Prototyping can mean a lot of things in the context of user experience design. There are many different common types of prototype, all of a varying level of fidelity — regardless of the technologies used or the complexity of the prototype, however, the most important aspect is to deliver something tactile that stakeholders can inspect and evaluate, which properly represents the direction the design is likely to take.
Examples of protoypes include paper prototypes, which are often paper simulations of otherwise digital experiences (such as interfaces for websites, mobile applications, and so on), or even high-fidelity digital prototypes such as very simple implementations of a proposed interface. Generally, designers will start by generating low-fidelity prototypes and, as the concept for a design becomes closer to the stakeholders’ expectations, gradually work toward high-fidelity prototypes closer to a representation of the final product.
Many design and development methodologies revolve around rapid prototyping — this means quickly generating relatively simple prototypes according to feedback, and steadily changing them with successive rounds of feedback until the stakeholders are satisfied that a design is optimal for purpose.
Test
Testing is difficult to really attribute to its own separate phase because it ties in so heavily with the previous phase — testing is an essential element of prototyping because it can and should inform the next step in an iterative prototyping process. That being said, oftentimes testing does not refer to running prototypes by stakeholders, but instead actually exposing a potential user to a late-stage prototype not only to collect direct feedback, but to observe how well users interface with the prototype. This information can then be used to inform the very final stages of prototyping and development.
This phase allows designers to see exactly where flaws in the design exist, and where there might still be some unaddressed user pain points — sometimes it is impossible for designers or clients to identify these issues independently, meaning that even very well developed designs can still have glaring issues for the end user.
Testing can take many forms — direct methods such as observation and interviewing are popular choices and can result in quality information, but can put pressure on a participant and hurt the quality of the data. Other methods allow for remote user testing, with many web services allowing access to a pool of anonymous users. These methods vary very greatly depending on what is being tested, however, and domain-specific methods may need to be used to ensure that usable information is gleaned from the results.
I’d love to hear your thoughts on this article! Feel free to reach out to me on LinkedIn.
If you’d like to hear my less wordsy musings, check out my Twitter!
Also, please visit me at my website!
