discovery-research-synthesis
Turning research artifacts into actionable PM insight. Customer interviews, user research notes, support ticket reviews, sales call transcripts, survey data, in-app feedback, all synthesized into the decisions they are meant to inform.
Install / Use
npx skills add rampstackco/claude-skills --skill discovery-research-synthesisInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Customer SupportSupported Platforms
Our assessment of discovery-research-synthesis
discovery-research-synthesis scores 86/100 on our quality scale, 219th of 319 Customer Support skills we index.
Its SKILL.md is 25 KB long, well organised into 15 sections and no code examples: a thorough specification that gives an agent plenty to work with.
It has 902 GitHub stars, a meaningful sign that others use it.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so discovery-research-synthesis is actively maintained.
- Our last check on 2026-11-03 found the source still online.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-10-04. Automated pattern scan on 2026-10-04. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
discovery-research-synthesis compared with similar skills
All 4 of these similar skills score higher than discovery-research-synthesis; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| discovery-research-synthesis (this skill)by rampstackco | 86 | 902 | 2mo ago | SKILL.md |
| x402by browser-use | 100 | 121.0k | 3d ago | SKILL.md |
| algorithmic-artby anthropics | 100 | 177.9k | 55d ago | SKILL.md |
| pptxby anthropics | 100 | 177.9k | 55d ago | SKILL.md |
| designby nextlevelbuilder | 100 | 149.0k | 6d ago | SKILL.md |
Frequently asked questions
- How do I install discovery-research-synthesis?
- Run
npx skills add rampstackco/claude-skills --skill discovery-research-synthesis. The install tabs above show the steps for each supported agent. - Which AI agents does discovery-research-synthesis work with?
- It is written for Zed, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is discovery-research-synthesis safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is discovery-research-synthesis still maintained?
- The repository was last updated about 2 months ago, so discovery-research-synthesis is actively maintained.
Skill content
View source on GitHubname: discovery-research-synthesis description: "Turning research artifacts into actionable PM insight. Customer interviews, user research notes, support ticket reviews, sales call transcripts, survey data, in-app feedback, all synthesized into the decisions they are meant to inform. The discipline of moving from raw discovery data to clear product direction without losing signal in the synthesis or fabricating insight that was not actually there. Triggers on research synthesis, customer interview synthesis, user research analysis, discovery readout, research insights, sales call analysis, support ticket analysis, qualitative data analysis. Also triggers when a team has done research but cannot turn it into decisions, when synthesis is producing pretty decks but no roadmap movement, or when an upcoming PM decision needs to be grounded in research already conducted." category: research catalog_summary: "Synthesizing customer interviews, research notes, and support tickets into actionable PM decisions. Distinguishes data-dump (no synthesis) from insight-theater (overpolished narrative) from actionable synthesis (decision-grade clarity)" display_order: 4
Discovery Research Synthesis
A senior PM's playbook for turning research artifacts into decisions. Customer interviews, user research notes, support ticket reviews, sales call transcripts, survey data, in-app feedback, all synthesized into the product direction they are meant to inform.
Most discovery research never produces decisions. The team conducts interviews; the transcripts pile up; a researcher hands product the raw artifacts (data-dump) or builds a polished readout deck (insight-theater); the deck gets a 30-minute review meeting and is never referenced again. Two months later the team is making the same product decisions that the research was supposed to inform, with the same gut-feel inputs the research was supposed to displace.
The discipline is in the synthesis. Synthesis is where research earns its keep: where transcripts become tagged observations, observations cluster into patterns, patterns get named, named patterns surface product implications, and implications drive specific decisions. Without that sequence, research is performance art.
This skill covers one-off discovery research projects: a 12-week customer development sprint, a sales-call review for an onboarding redesign, a support-ticket audit informing a roadmap quarter. Different from user-feedback-aggregation, which covers ongoing feedback streams; different from jtbd-framing, which is a specific framing technique often applied within synthesis but narrower in scope.
The voice is the senior PM or staff product researcher who has run synthesis well and seen plenty of teams fail at it. Honest about where polish becomes performance and where pattern-naming slides into pattern-fabrication.
When to use this skill: synthesizing a recent batch of customer interviews, auditing why prior research has not produced product decisions, designing the synthesis output for a multi-week discovery sprint, or establishing the synthesis discipline a team currently lacks.
What this skill is for
This skill spans research-to-decision synthesis. The PM-skill distinction:
discovery-research-synthesis(this skill) is one-off research synthesis: turning a discrete batch of artifacts into a discrete set of decisions.user-feedback-aggregationis ongoing feedback streams: continuous synthesis across always-on channels (support, NPS, in-app, sales).jtbd-framingis a framing technique often applied during synthesis; this skill is broader.pm-spec-writingis downstream: specs use synthesized insights as input.roadmap-planningis downstream: roadmap uses synthesized priorities as input.experiment-designis the quantitative-validation discipline that often follows qualitative synthesis.
The audience: senior PMs, product directors, in-house product teams, agencies running discovery work for clients, researchers handing off to product teams.
What is not in scope: conducting the research itself (interview design, recruiting, moderation), the broader discovery strategy that decides which research to commission, or the long-running feedback channels that user-feedback-aggregation covers.
Data-dump vs insight-theater vs actionable-synthesis
The keystone framing. Two failure modes plus the discipline.
Data-dump. Research artifacts handed to the product team raw. "Here are 47 transcripts and a spreadsheet of tags; figure it out." No synthesis, no signal-finding, no implications drawn. The product team either does the synthesis itself (often badly, often months later) or skips it (the most common outcome). Output: a research investment that produced artifacts but not insight.
Insight-theater. Overpolished synthesis dressed as insight. Pretty quote walls, designed personas, branded readout decks, no decision-driving conclusions. The research deck that gets a 30-minute review and never gets referenced. Synthesis that performed for the executive audience but did not inform the product audience. Cost: real research budget converted into a slide deck; the organization "did the research" without doing the research.
Actionable-synthesis. Synthesis that drives decisions. Each pattern has a "so what" attached; each finding leads to a product implication; the document gets referenced in roadmap discussions, spec drafts, prioritization debates. The synthesis is short, opinionated, and load-bearing. Six months later, team members can quote the patterns that shaped what shipped and what got cut.
The litmus test. Three months after a synthesis ships, can team members name the decisions the synthesis informed? If yes, the synthesis was actionable. If team members remember the deck but cannot name the decisions, it was insight-theater. If nobody remembers anything because the synthesis never happened, it was data-dump.
Discovery research types
Five common research types, each with different artifacts and synthesis needs.
Customer interviews. 30-90 minute structured or semi-structured conversations. Transcripts plus moderator notes plus recordings. Highest-quality qualitative data; most expensive to collect; most rewarding to synthesize when done well. Typical batch: 8-15 interviews per discovery cycle.
Support ticket reviews. Auditing recent support tickets for patterns. Lower per-artifact richness than interviews; higher volume; surfaces the friction users actually hit (not just what they say in interviews). Typical batch: 100-500 tickets per audit.
Sales call analysis. Recording or transcript review of recent sales calls. Surfaces objections, pricing-conversation patterns, competitive mentions, and the gap between marketing positioning and what sales actually leads with. Typical batch: 15-30 calls per audit.
Survey data. Quantitative responses with optional qualitative comments. Best for validating qualitative patterns at scale; weak as primary discovery (surveys reveal what users say, not what they do). Typical batch: 100-2,000 responses.
In-app feedback. Users submitting feedback through in-product channels. Volume is high; quality varies; signal lives in patterns across many low-effort submissions. Typical batch: ongoing; synthesized in periodic audits.
Different research types compose. A discovery cycle often pairs interviews (depth) with support ticket review (volume) with survey data (validation at scale). The synthesis combines them.
Detail in references/research-types-and-when-each-fits.md.
The synthesis sequence
Six stages from raw artifacts to decisions. Skipping stages is where synthesis fails.
1. Transcribe and prepare. Convert artifacts into a synthesizable format. Interview recordings to transcripts; sales calls to summaries; support tickets to a tagged dataset. The prep work is unglamorous; teams that skip it never reach the later stages.
2. Tag at the artifact level. Read each transcript or ticket and tag it with the topics, problems, and quotes that surface. First-pass tagging is descriptive: what did this person say, what problem did they hit, what were they trying to do. Tags are messy; that is fine.
3. Cluster across artifacts. Group tags into themes. Multiple users mentioned the same friction; multiple tickets reference the same workflow; multiple interviews returned to the same struggling moment. The clusters surface the patterns.
4. Name patterns. Each cluster becomes a named pattern. The naming is the work: a pattern named badly disappears into the synthesis; a pattern named well becomes the framing the team references. "Onboarding configuration friction" is a pattern name; "users had trouble" is not.
5. Infer product implications. For each pattern, what does this imply for the product? Implications are the bridge from observation to decision. A pattern without an implication is a finding; a pattern with an implication is actionable input.
6. Name the so-what. For each implication, the specific decision it informs. "Should we prioritize the onboarding redesign in Q2?" The so-what is what makes the synthesis decision-grade.
The sequence is non-negotiable. Teams that skip tagging and jump to pattern naming hallucinate patterns; teams that skip implications produce findings without product direction; teams that skip so-what produce documentation, not synthesis.
Detail in references/synthesis-sequence-walkthrough.md.
Tagging and clustering discipline
The middle stages of the synthesis sequence have specific failure modes.
Tag at the artifact level first; do not pre-design the taxonomy. Tags emerge from the artifacts. Teams that arrive with a pre-built tag taxonomy ("we will tag for usability, performance, pricing, support") force artifacts into categories that may not fit the data, and miss patterns the taxonomy did not anticipate.
Allow tag proliferation; collapse later. Early tagging may produce 80-150 tags across a batch. That is correct. The clustering stage collapses them. Premature consolidation ("we had 4 onboarding tags; let me merge them") loses distinctions that turn out to matter.
Cluster on patterns, not on tag overlap. Two tags that co-occur in many artifacts may or may not be the same pattern. Cluster on the underlying observation: are these the same struggling moment, or do they share a tag because the tag was loose?
Name the cluster after clustering, not before. The cluster name comes from the data, not from a hypothesis. Teams that arrive with named clusters and slot data into them produce confirmation; teams that name clusters from the data produce discovery.
Tag dissent and contradictions explicitly. Some users will contradict the dominant pattern. Those contradictions are signal: either there is a sub-segment with different needs, or the pattern is weaker than the volume suggests. Tag the dissent so the synthesis can address it.
Detail in references/tagging-and-clustering-discipline.md.
Pattern naming
The work that distinguishes synthesis from documentation.
A pattern name is not a category label. "Onboarding" is a category. "Users abandon configuration before reaching the success state because the third configuration step requires data they do not have on hand" is a pattern. The pattern names the specific behavior the data reveals.
The name should make the implication legible. A reader who sees only the pattern name should be able to guess the product implication. If the name is too abstract, the reader cannot reach the implication; if the name is too concrete, the synthesis becomes a list of micro-observations.
Avoid category bloat. Synthesis with 30 named patterns is usually under-
Truncated for display — read the full file on GitHub.
Related Skills
x402
121.0kSet up Browser Use Cloud payments with x402 — pay per request from a crypto wallet (USDC on Base mainnet), no signup or API key.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
pptx
177.9kUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an em…
design
149.0kComprehensive design skill: brand identity, design tokens, UI styling, logo generation (55 styles, Gemini, Atlas Cloud, or MuAPI AI), corporate identity program (50 deliverables, CIP mockups), HTML presentations (Chart.js), banner design (22 styles, social/ads/web/print), icon design (15 styles, SVG…
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
