Most designers know how to collect research. Surveys, interviews, usability sessions — the methods are well-documented. What gets skipped is what comes after: the hard work of turning a pile of observations into something a team can actually build from.
I have sat with 50 survey responses, 20 interviews, and 100 Reddit threads and felt completely stuck. Not because the data was bad. Because I did not yet know how to read it.
Synthesis is the skill most designers never talk about — and the one that determines whether research actually changes anything.
Collecting data is the easy part. The hard part is sitting with it long enough to hear what it is actually saying.
What 100+ Reddit threads taught meMore research does not mean more clarity. Without a synthesis process, more data creates more confusion.
The most important insights are rarely in what users say directly. They are in what keeps coming up across sources.
Most teams go from research straight to wireframes. Synthesis is the layer in between — and skipping it costs everything.
When I worked on Lighthouse AI, I ran 50 surveys, 20 interviews, and dug through over 100 Reddit threads. The data was rich — but also overwhelming, contradictory, and scattered.
Synthesis is not about organizing data. It is about finding the thread that runs through all of it.
Pull everything in first. Do not decide what matters yet.
Not what one person said loudly. What shows up quietly across ten different sources.
Vague posts and frustrated language reveal overwhelm — the real signal.
Stop at patterns, not quotes. The insight is the conclusion.
Every synthesis should end with a direction. If it does not change something, it was not synthesis.
When I went into the Reddit research for Lighthouse AI, I expected to validate what we already assumed. Instead, the data kept pointing somewhere else — users were not asking for better advice. They were asking to feel less alone in their uncertainty.
That single shift — from information product to emotional support product — changed the entire design direction.
Vague posts do not mean unclear thinking. They often mean the person does not yet have language for what they are feeling. That is exactly where the product needs to meet them.
The Reddit insight that changed Lighthouse AIResearch without synthesis is just documentation. The difference is what happens after the data is collected.
Collecting research is the easy part. Synthesis is judgment — deciding what the data is actually saying and what to do about it.
That skill comes from sitting with the mess long enough to see the shape of what it is trying to say.