Research & Synthesis

Research generates raw material. Synthesis transforms that material into actionable insight. I have found that teams who invest heavily in synthesis—not just in collecting feedback—make better product decisions and waste less time building the wrong things.

Synthesis Overview

Synthesis is the process of transforming raw customer feedback, interview notes, support tickets, and behavioral data into structured insights that can guide product decisions.

Who Does Synthesis

In a small trio, the PM typically owns synthesis. As organizations grow, dedicated UX researchers take on more of this work—but the PM remains deeply involved in interpreting findings and connecting them to business strategy.

The key is that synthesis isn’t purely delegated. Even with a UX researcher doing the heavy lifting, the PM should be close enough to the raw material to develop intuition about customer needs. Synthesis at arm’s length loses nuance.

When Synthesis Happens

Synthesis begins immediately after research activities—not weeks later when context has faded.

Two streams emerge from every research session:

Stream	Destination	Timeline	Purpose
Immediate actionables	Ticket system	Same day/next day	Urgent bugs, major frustrations affecting current work
Broader insights	Insight management system	Within the week	Themes to inform roadmap, enrich future tickets

For immediate actionables, create tickets right away. Include why it matters—the customer impact, not just the technical issue. These enter triage for discussion in the current or next cycle.

For broader insights, capture them in an insight management system (like Dovetail) where they can be grouped into themes. These inform:

Quarterly and yearly roadmap discussions
The Research phase of new roadmap projects
Context added to tickets during backlog grooming

Synthesis Is a Full-Time Job

In practice, synthesizing feedback is where PMs or UX researchers spend a large portion of their time. This isn’t overhead—it’s core work. The value of customer research is unlocked through synthesis; without it, you have noise, not signal.

Other team members review the synthesis and add perspectives on what might have been missed—just as PMs add perspective to design or technology choices. But the synthesis itself requires dedicated, sustained attention.

Common Synthesis Methods

Different synthesis methods serve different purposes. I have found that journey mapping is the workhorse—the method you’ll return to most often—while other approaches serve as inputs or supplements.

User Journey Mapping: The Primary Method

User journey mapping is the central organizing principle for synthesis. All feedback—from interviews, support tickets, analytics, and insight management systems—should be filtered through your customers’ various journeys.

Why journeys work:

Customers don’t experience your product as features; they experience it as tasks they’re trying to accomplish
Journeys provide context—the same pain point means different things at different stages
Journeys reveal gaps between steps, not just problems within steps
Journeys make opportunities visible by showing where friction concentrates

Key insight: Customers have many stories, not one linear journey.

A single user persona might have distinct journeys for:

Initial setup and onboarding
Daily operational tasks
Troubleshooting and recovery
Expansion to new use cases

Each journey deserves its own map. Feedback that seems scattered often coheres when you ask: “Which journey is this customer in?”

Supporting Methods

Affinity Mapping / AI-Assisted Grouping

Tools like Dovetail use AI to group raw feedback by topic or concern. This serves as a starting point—a pre-filter that helps you find relevant material to add to journey maps.

I have found that AI grouping extends the reach of what you can process, but you’ll need to review its thinking and build your synthesis beyond it. The machine finds patterns; the human judges which patterns matter.

Health Tagging

When feedback enters your system, tag it with health indicators:

Severity (blocking, painful, annoying, minor)
Frequency (how often does this come up?)
Segment (which customer types report this?)

These tags help you filter and prioritize during later synthesis sessions.

Quantitative Context

Tools like PostHog, Datadog, and similar analytics platforms provide macro-level data. Use them alongside qualitative synthesis:

Does the pain point appear in funnel drop-offs?
How many users encounter this step?
What’s the success rate for this journey?

Quantitative data tells you where to look; qualitative synthesis tells you why it matters.

Market and Competitive Research

Your own research on market trends, competitor products, and industry shifts provides context that customer feedback alone can’t. Synthesis should incorporate external signals, not just internal customer voice.

The output of synthesis isn’t a report—it’s enriched journey maps that reveal where the biggest opportunities live.

Synthesis Best Practices

Preserve Original Context

When you map feedback to journeys, preserve the connection to original sources:

Link to the interview recording or transcript
Note the customer’s name and date
Include enough context that someone can go deeper later

Synthesis abstracts and generalizes—but the abstraction should never become untraceable. When someone asks “where did this insight come from?”, you should be able to show them.

Looking for Patterns—But Not Demanding Them

A common question: “How many customers need to report something before it’s a pattern?”

I have found that the threshold is lower than people think. An interface informed by one customer’s concerns is still better than one informed by zero customers’ pains.

When one customer identifies an issue:

It tells you where to look
You can probe the area in subsequent interviews
You can set up tracking to measure scope quantitatively

The first report is signal, not proof. Synthesis treats it as a hypothesis worth investigating—not dismissed because n=1, but not acted on blindly either.

That said, guard against the loudest voice dominating. If one particularly articulate customer describes a problem vividly, it can overshadow feedback from others. Journey mapping helps—you see all the pain points for a step, not just the most memorable quote.

Build Beyond AI Groupings

AI-assisted synthesis tools are genuinely useful. They surface patterns you might miss and process volume you couldn’t review manually.

But they have limits:

They group by surface similarity, not underlying cause
They don’t know your strategy or business context
They can’t judge which patterns matter most

Use AI groupings as input to your synthesis, not as the synthesis itself. Review what the tool suggests. Ask: “Does this grouping reflect a real customer need, or just similar language?”

Connect Individual Feedback to Macro Data

Good event tracking in your application helps contextualize individual feedback.

When a customer reports a problem:

Can you see in your analytics how many users encounter that step?
Can you measure success/failure rates for that flow?
Can you identify which segments are most affected?

The combination of qualitative depth (why it matters) and quantitative breadth (how many it affects) is more powerful than either alone.

Make Judgment Calls

Synthesis requires decisions. Where does this feedback fit in the customer’s story? Which journey does it belong to? What is the underlying need behind the stated request?

These are judgment calls. They can be wrong. But making them is essential—synthesis that defers all interpretation isn’t synthesis.

Document your reasoning. When you decide that a piece of feedback reflects a specific pain point in a specific journey, note why. Future you (or future teammates) can revisit the judgment if new information emerges.

Shareout Best Practices

Synthesis generates insights. Shareouts turn those insights into shared understanding and aligned action.

Format: Canvas-Based Workshops

For team synthesis shareouts, I have found that interactive canvas tools (Miro, FigJam, or similar) work better than slide decks.

Why canvas-based workshops:

Everyone can see the full picture simultaneously
Team members can add their own observations
Voting and prioritization happen in real-time
The artifact remains useful after the session

Workshop structure:

Present the synthesis: Walk through journey maps with mapped pain points
Discuss and refine: Team adds perspectives, challenges interpretations
Vote on priorities: Which pain points are most important to address?
Define next steps: What enters the current work? What becomes a roadmap item?

These workshops help the team align on focus and objectives. They also create artifacts to refer back to—when someone asks “why are we working on this?”, you can point to the synthesis session.

Audience: Right Information to Right People

Different audiences need different formats:

Audience	Format	Frequency	Content
Trio + immediate team	Canvas workshop	Per research phase / project	Full synthesis, interactive discussion
Stakeholders	Slide deck	Periodic (monthly/quarterly)	Key findings, implications, recommendations
Leadership	Executive summary	As needed	Strategic insights, resource implications

Don’t over-share raw synthesis with stakeholders who need conclusions, or under-share with teammates who need context. Match the format to the audience’s needs.

Letting Users Speak for Themselves

The most powerful moments in shareouts are when customers speak in their own words.

Include:

Direct quotes (or close paraphrases)
The name of the person (with appropriate privacy considerations)
Links to original recordings or transcripts

The point isn’t to prove your synthesis is correct—it’s to keep the customer present in the room. When the team hears a customer’s actual words, they connect emotionally in ways that summarized insights can’t achieve.

Balance raw voice with synthesis:

Element	Purpose
Journey map with pain points	Shows the synthesized picture
Direct quotes at each pain point	Grounds synthesis in real voices
Links to full interviews	Enables deeper exploration

After the Shareout

Workshops generate alignment, but alignment fades. Capture the outcomes:

What priorities did the team agree on?
What items move into current work vs. roadmap?
What questions remain open for further research?

Document these decisions in a place the team can reference. The workshop artifact (the Miro board, the journey map) should remain accessible—not buried in someone’s personal folder.

Summary: Research & Synthesis Principles

Area	Core Principle	Practical Application
Timing	Synthesize immediately, not weeks later	Two streams: immediate tickets + broader insights to system
Ownership	PM or UX researcher owns; team reviews	Synthesis is full-time work, not side task
Method	Journey mapping is the workhorse	Filter all feedback through customer stories
Multiple journeys	Customers have many stories	Map distinct journeys for setup, daily use, troubleshooting, etc.
Context preservation	Link back to original sources	Every insight should be traceable to its origin
Pattern threshold	One customer is signal, not noise	Use first report as hypothesis; validate further
AI assistance	Use as starting point, not endpoint	Review and build beyond AI groupings
Quantitative + qualitative	Combine depth and breadth	Individual feedback + event tracking data
Shareouts	Canvas workshops for teams	Vote, align, document next steps
User voice	Include direct quotes with attribution	Keep customers present in the room

Synthesis is where customer understanding becomes actionable. Without it, research is just conversation. With it, research becomes the foundation for every product decision.