Design Synthesis
A step-by-step guide to translate research into actionable insights
Raise your hand if you can identify with the next scenario:
You have just finished your design research, interviews, observations, and secondary research. You have lots of information, now what do you do with it? How can you identify what’s really important and what’s not?
I have been lost in a sea of information many times before, and I’ve seen many designers around me get stuck too. So I wrote this article as an effort to create a guide that helps designers give structure and clarity to a process that most of the time is done intuitively. This analysis phase I’m talking about is called design synthesis, and it’s a step that happens right after the design research.
The problems
The object of research in design is to inform designers so we can make strategic decisions on what or how to design. But data itself doesn’t tell us anything without an interpretation; an analysis must be made. It is during this analysis that many designers face one of two problems.
Sometimes the amount of data produced by research can be overwhelming. Not knowing where to start — or stop — can lead frustrated designers to analysis paralysis. Other times we find ourselves going straight from research to design because, as we learned from users, we came up with ideas to solve our design problem. Trusting our intuition may work on occasions, but more often than not, our designer ego can lead us to unwanted products and services. So this is where a structured process can come in handy.
The theory
Design synthesis is the process of translating research data into actionable knowledge and is a critical part of the design methodology. The goal of this process is finding relationships between different pieces of data to uncover meaning in the behaviors that were observed during the research phase. This understanding allows us to identify opportunities and constraints that will set the space in which we’ll generate (more accurate) solutions.
Jon Kolko, one of the designers that have been writing extensively about this subject defines design synthesis as “an abductive sensemaking process of manipulating, organizing, pruning and filtering data in an effort to produce information and knowledge” (Kolko, Interaction Design Synthesis: Translating Research into Insights, 2009). Let’s break down this definition.
Abductive reasoning is a logical way of inference on why a phenomenon is happening. It relies on hypotheses as arguments that best explain something based on available data and previous experience. Doctors use abductive reasoning to diagnose their patients; given a set of symptoms and their previous experience, they decide on a diagnosis that can best explain most of them.
In the same way, designers use abduction to give meaning to the data gathered in the design research. Using their prior experience and the learnings from the users to find good enough explanations on why a behavior is happening. Abductive reasoning is a form of trained intuition.
Sensemaking is the process of converting data to knowledge that all individuals go through in everyday life. This is how learning takes place and how we understand our world. As a learning process is subjectively dependent on emotions, previous knowledge, and prior experiences of the learner, and so is design synthesis.
The difference between individual sensemaking and design synthesis is that the first often occurs internally in one’s head, and the second needs to happen externally in a group of people. During design synthesis, designers make sense of data collaboratively, by working on groups we can gain from multiple perspectives and broader interpretations to build a shared understanding of the research.
The practice
In much simpler words, design synthesis is what happens when designers interpret data collaboratively using their previous experience and users’ perspectives to form hypotheses that explain why a behavior is occurring. To achieve this goal, data must be manipulated through organizing, pruning, and interpreting efforts. Here’s a brief step-by-step guide on how to go through this process:
Step 1: Organizing
Is the action of grouping data by affinity, putting related items near each other to identify patterns or outliers in the research. To organize, we move pieces of data freely thought space and create clusters of similar data. Once groups emerge, we label them to make explicit the content of each group — this is known as affinity diagrams — . Doing it this way can prevent us from assuming preconceive ideas about how the data should be grouped.
Organizing show relationships and uncover patterns.
In this step, a visual organization of the data is crucial to gain a complete picture of the research. Since the amount of data is too large to hold by memory, externalizing it on a physical or digital space creates a spatial structure for the information that helps designers find patterns and relationships easily. The most common way to do it is by using a wall and move pieces of data around to arrange them in clusters. But it can also be done using online platforms for collaborative work, like Miro.
Step 2: Pruning
Is the action of selectively removing or ignoring data, deciding which items are more important than others, and therefore prioritizing. Pruning is made on popularity and rareness, items of data that are repeated don’t need to be captured more than once, and pieces of data that only show up once may not be important.
Pruning helps us focus on what’s relevant.
This step forces designers to define relevance; that’s why it requires a constant look back at our current design problem to use it as a north star as we prune. Like new knowledge, everything may seem interesting. But not all data will be relevant to our goal, comparing every piece with our design problem will help us identify and prioritize the most significant. By pruning, we can avoid losing sight — and time — with information that can deviate us from our primary goal.
Step 3: Interpreting
Is the action of assigning meaning to data, creating hypotheses to explain why a behavior is happening. Hypotheses are set by making inferential leaps to best explain the why of something; these interpretations are informed by the data but are subjectively influenced by the context of the designers.
Interpreting assigns meaning and creates knowledge.
During this step, designers need to look at data using their own perspectives and also the users’ points of view. Asking why a behavior may be happening and trying to answer with connections between elements that create a credible explanation based on data and their own context. These abductive inferences — called insights — create new knowledge that will be used to provoke new ideas.
These three steps are an easy to follow guide to go through design synthesis. At the end of this process, designers should have defined opportunities and constraints, in the form of insights, to create a framework for design solutions.
Design synthesis is where we set the direction of our design efforts when done right, it can inspire innovative solutions — that’s why it is a critical part of the design process, and emphasis should be placed to have a more conscious approach to it.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
