Win org buy-in for design via business-case framing. Use when pitches stall, PMs push back, manager relationships need repair, or at review/promo time. Your own reports → people-management; product decision quality → product-decision-making; pitch craft → design-prototyping. Triggers: stakeholder buy-in, design pitch rejected, PM pushback, business case, managing up, promotion, performance review, wins doc, order taker, seat at the table.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
Win org buy-in for design via business-case framing. Use when pitches stall, PMs push back, manager relationships need repair, or at review/promo time. Your own reports → people-management; product decision quality → product-decision-making; pitch craft → design-prototyping. Triggers: stakeholder buy-in, design pitch rejected, PM pushback, business case, managing up, promotion, performance review, wins doc, order taker, seat at the table.
"It's not your design they don't trust, it's how you're framing the value." Rejection is rarely about craft, leadership, or politics — it's solution-first thinking: falling in love with a solution before understanding the business problem, presenting features instead of outcomes. The designers who consistently get ideas approved "think business first and design second."
Order taker → product-thinking designer. Someone suggests a feature; most designers say "Okay, I'll design it." The influential designer asks "What problem are we solving?" / "How will we know this worked?" — "That tiny shift from order taker to problem solver changes everything for your career." (Julie Zhuo framing: references/pm-partnership.md.)
Waiter → problem owner (Perri, Escaping the Build Trap, ch. 6). "It's not the customer's job to come up with their own solutions. That is your job." Served feature orders feed the product death cycle; move every request back to outcome, audience, current behavior, and success measure (references/influence-field-notes.md).
The pre-Figma gate. Before anything else, write down three things: (1) What business problem are we solving? (2) What user problem are we solving? (3) How will we measure success? "If you don't have answers to those questions, don't open Figma yet" — align with your PM first. "User problems and business problems aren't always the same thing."
Don't delegate impact (Soleio via Julie Zhuo, 2026). Designers who hand strategy and impact to PMs cannot ask to be treated as excellent; they must help shape the why, pressure-test whether the idea should exist, and use design to give definition to the what-if. The practical move is not turf war: ask the PM's questions yourself, then bring sharper evidence and options.
"Always have a pitch for your complaints" (Steven Steiner). Never bring a problem without a proposed move. Also his door-opener: "How can I help you right now?"
Reposition yourself: "not as someone who slows things down with research, but as someone who de-risks decisions." Exploratory work without metrics = frame as risk reduction ("2 weeks validating assumptions vs 3 months of wasted eng time").
"Your job isn't just to make things look good. It's to make the business better through design."
The impact framework — 6 steps: (1) identify the business problem first ("Our checkout abandonment rate is 68%, costing us $2.3 million annually" — not "this checkout flow has usability issues"); (2) connect design to metrics leadership cares about; (3) present with data — three layers (quantitative foundation → qualitative insights → solution with projected impact); (4) anticipate objections, decode stakeholder motivations; (5) clear implementation plan (phased rollout, the MVP that "delivers 80% of the value"); (6) time the pitch strategically. "Your design brief should read like a business case, not a creative brief."
The strategic narrative framework — 4 parts: problem context (what's broken, in business terms) → your insight → design solution (features connected to outcomes) → expected impact (quantify where you can, "be directional when you can't").
Four lenses for craft/design-system asks (when you hear "opportunity cost"): customer impact · company goals · team efficiency ("That's not opportunity cost, that's opportunity creation") · tech-debt timing ("the cheapest it will ever be"). Bring engineering allies so it isn't you advocating alone.
Timing: planning/budget cycles, competitor launches, bad metrics, churn spikes — and manufactured urgency. "Smart designers don't just wait for the perfect timing, they create it."
The before/after craft deck (Briggs, Dive Club): sell UI craft with side-by-side default-vs-crafted slides; the reaction to engineer: "it's so obvious now that I see the comparison." Comparison axes → references/influence-field-notes.md.
The animation-adoption case (Nabors, Animation at Work, ch. 4): skip the science debate, "talk benefits"; show, don't tell — the animated version next to the static comp. And be a gardener, not gatekeeper: the lone evangelist is a bus-factor risk — build allied experts. Full quotes: references/pitching.md.
Translate your message into each audience's motivations (Lara Hogan, Resilient Management, ch. 4): "translate your message into something that's much more digestible and easy for that audience to care about" — one elevator pitch per audience, repeated often (her web-perf case: references/pitching.md).
Before/after: "I redesigned our mobile app because the current design looks outdated" → "Mobile app engagement dropped 23% this quarter… projected to increase screen time by 40%… here's the six-week rollout plan." Same design, completely different reception.
Business translation, not design translation (Sten): if stakeholders keep choosing the "modern-looking" option over the researched one, stop explaining whitespace, systems, and flows. Translate the decision into the language they already use: revenue, recurring revenue, signups, support cost, retention, operational overhead, debt, risk, and strategic position. "Design inconsistency" may land as ; "poor content" may land as .
Sticky framing (Heath & Heath)
The last pass on a stakeholder pitch or strategy memo, never a substitute for evidence. Five checks — full quotes and examples: references/influence-field-notes.md.
Find the Commander’s Intent. One pitch gets one lead — "If you say three things, you don't say anything"; the core must be decision-useful, not merely short.
Break the right guessing machine. Name the current assumption, show why it fails, replace it with a better decision rule.
Make it concrete before asking for belief. Lead with specific people doing specific things — a customer scene, support ticket, prototype moment. "Abstraction is the luxury of the expert."
Put numbers at human scale. Give statistics a comparison frame or everyday scale ("$500/employee/year" → "one or two minutes a day").
Carry the moral in a story. A concrete incident over a slogan; stories simulate action.
The four PM phrases: "I don't like this" · "this will take too long / no resources" · "it looks great, but this isn't a priority" · "I don't think it's gonna work." One response pattern: ask why → reframe to the goal ("which of these concepts best helps us achieve that goal?") → propose a test (usability test, A/B, phased milestone).
The conflict script (framed around shared success): "I want us to succeed at [goal]. I'm concerned about X because [reason]. What if we tried Y?" "Conflict isn't personal. It's about the goal." Productive pushback builds trust.
Decode before defending: "Sometimes a push back isn't about your design. It's about a perceived gap in your process" (Florian) — fix the expectation, not the mockup. "When you understand motivations, objections become collaborative problem solving."
Cite empirical psychology when project data doesn't exist (Jon Yablonski, Laws of UX, preface): grounding decisions in named, cited research "became quite useful in convincing project stakeholders to move in the proposed design direction" — borrowed evidence: weaker than your own data, far stronger than taste-versus-taste (references/objections.md).
Negotiate roles explicitly (Miranda Slater): "These are what I feel like I should be responsible for. What do you feel like you should be responsible for?" — try lines for a quarter, run a retro, iterate.
Set the expectation early: ask to be "approached with problems and not solutions," then co-create discovery. Your PM isn't your boss; they're your partner.
Empathy-first decode (Miranda): don't conclude "the PM is just a control freak" — "It could be a past experience or lack of experience" with designers. We don't extend PMs the empathy we extend users.
"Think of your stakeholders as your users" (Miranda): of 16 things you did, pull forward the four that prove strategic value. Speak their language — with UX jargon "you're setting that barrier for yourself."
Invite yourself to the rooms you're left out of (Metts & Welfle, Writing Is Designing, ch. 8) — three scripts (listen-in, productivity enabler — "nothing opens doors like dev-team risk", helping hand), plus weekly office hours as an influence channel: references/pm-partnership.md.
"Don't confuse making a deliverable with making a difference" (WID ch. 8): meet the team's needs first; the support to meet users' needs follows.
Five seniority anti-patterns: waiting for permission · outputs over outcomes · not building relationships proactively · avoiding conflict · underselling your impact. "Outputs don't get you promoted."
Wins doc: one entry per week — problem, what you did, impact (qualitative counts). "No one's going to care about my career as much as I do."
SBI (Situation–Behavior–Impact) — the Center for Creative Leadership's technique (per Kim Scott, Radical Candor: it makes feedback "more precise and therefore less arrogant") — for every entry: situation with a number → what you specifically did → what changed. Same work, completely different picture.
Four self-reflection mistakes: responsibilities instead of contributions · outputs instead of outcomes · vague-to-seem-humble ("Vague is not humble, just unhelpful") · only writing about wins (all-sunshine "reads as unaware").
Invisible work — "There's no Figma file for prevented a bad decision": log the reversed decision, the aligning brief, the 45-minute unblock when it happens or it's gone.
The credibility ladder (Miranda): master the language → find small wins → build credibility gradually → then ask for your seat at the table ("I know planning's coming up. Can I be included…?").
Her stateful Claude-skill review logger: setup interview → artifacts converted to SBI entries tagged to competencies → attributed peer-feedback log → end-of-period synthesis. Build the review doc in the background, not at the deadline.
Career as a business case (Ryan Scott, Dive Club LOkdM6frkak)
The load-bearing gap: "There are skills that designers are getting hired for right now which are like visual design skills, high craft, but what they're getting promoted for is providing more value to the business" — treat the gap as the career strategy. The rules, one line each; full quotes and receipts: references/influence-field-notes.md.
Sell to budget-holders, not design managers — design managers rarely control budget; the buyer treats you as an investment, and that frame is fair.
Rejection is overfiltering, not competition — companies filter on one specific puzzle-piece detail.
Metric-direction language — describe work as driving a named metric in a direction, never as "a redesign."
Business value beyond metrics — competitor neutralization, velocity, quality, culture, brand, positioning; if the company launched it, it's probably producing value.
"Shock and awe" results slides — cite many kinds of value so no single number carries the pitch.
Role-convergence game theory — designers and C-suite are chess players, PMs poker players; speak chess to executives and renegotiate role lines out loud.
No-fidelity + idea sequencing — float the bare idea with allies first, widen gradually, never skip levels.
Value–Cost–Risk over RICE — cost includes organizational cost, risk includes morale/exposure/timing; mind the sub-MVP cliff.
Data practice — ask for the analytics tool yourself; chase weird outliers from the human angle.
Managing your manager
Julie Zhuo's 2024 Looking Glass note belongs here, not in people-management: the task is influencing up and repairing the manager-report contract from the report side.
Drop the executioner/white-knight model. Fear says the manager is waiting to punish mistakes; pining says the manager should defend and promote you. Both make the manager the main character in your career. Recenter on your own agency.
Define the relationship you want. Before judging the manager, write what a great manager-report relationship would look like, then describe that aspiration in a 1:1 and ask whether they agree.
Make the ask specific. "More support" is too vague. Ask for concrete help: more context on a decision, sharper feedback on a project, clearer promotion expectations, earlier visibility into priorities.
Align on both jobs. Ask: "How do you think about your role? What makes you feel you're doing a great job?" Then ask the mirror for yourself: "What would make you feel I'm doing a great job?" Trust improves when both success definitions are explicit.
Run a monthly feedback loop. Tell them what is helping, ask for more of what you need, and ask them for feedback. If strong repeated feedback is ignored, escalate to the manager's manager; if the chain loses trust, leave rather than normalizing mediocre management.
Influence modes, politics, and culture change
Zhuo's Looking Glass essays add three operating moves to the stakeholder playbook:
Know what they care about in <50 words. If you cannot answer "What does X care about?" and "Can my proposal genuinely help with that?", you are not ready to pitch. Ask first; then route your proposal through their goal.
Keep a bet ledger. Track what you pushed for, where you disagreed, what got overruled, what later happened, and what you learned. Good accounting sharpens product judgment and builds promotion evidence without relying on memory.
Advisor vs solver mode. Use advisor mode when the goal is to change someone's decision frame: words, stories, principles, alternatives. Use solver mode when the goal is to make a specific decision obvious: prototypes, experiments, narrowed pathways. As a report influencing a manager, default to solver mode; reduce decision burden.
Reframe "politics" as assumptions. In competent organizations, most politics is not villainy; it is hidden shared goals, different assumptions, and bad translation. Start conflict by asking which position is most likely to help the company win and what unique knowledge each side has.
Change culture by changing beliefs. Identify the current belief behind the behavior, why reasonable people adopted it, where it fails, and the replacement belief that serves the same goal better. Then repeat the new story until someone else says it.
Build-trap diagnosis (Perri / Bungay)
When the org keeps asking for more plans, controls, or features instead of better outcomes, name the build trap (Perri): success measured by outputs rather than outcomes — "we are rewarding shipped inventory instead of solved customer/business problems." Three gaps (Stephen Bungay, via Perri), each with a signature wrong fix: the Knowledge Gap patched with more detailed reporting, the Alignment Gap with more detailed instruction, the Effects Gap with more controls — the corrective is strategic intent: leadership sets goals and boundaries, teams decide how, evidence closes the loop. Reward learning over shipping, and remember "the unconstrained team is the most frightened and scared to act" (Jabe Bloom) — autonomy without strategy is paralysis. Full detail + Perri's six leadership-room diagnostic questions: references/influence-field-notes.md.
Org-building and the AI-era operating model
Hiring for capability gaps, the underdog binary, Core+3 flattening, five-lines-of-process, the three archetypes, designers-shipping-PRs, kill-mediocre-even-shipped, and the panel's CEO-in-the-tool diagnostic (Modisett, Worboys, Dill, Choy/Shipper/Zipper): references/team-building.md.
Ask for the ball (Soleio via Julie Zhuo, 2026) — with too many stakeholders, synthesize the input, choose a direction, explain why, and accept accountability; consensus-seeking turns design to mush.
CEO strategy diagnostic (Soleio) — before judging the design team, ask leadership "If I joined your team for a month, what would I work on?"; a vague answer means the real gap is company strategy, not design.
Strategy as sandbox (Soleio, Dive Club) — demand strategy that acts as a sandbox: competitive strategy, plan to win, resources, success measure; problem selection over problem solving.
Will this matter in a year? (Soleio, Design MBA) — triage by one-year consequence; reserve weight for the hard-to-unwind class (hires, product names).
Write before pixels (Luis Ouriach, Pn2G7JhxNKc) — writing before committing to a pixel is the best way to get buy-in.
Use writing to separate knowns from assumptions (Sten): before selling a concept, write what is known, what is assumed, and what remains unknown. Then build the story from the pattern in that list. It lowers defensiveness because the pitch is not pretending exploration is certainty.
Pick the artifact for the conversation you need (Sten): page briefs focus sections/actions/goals; wireframes invite hierarchy and structure discussion; visual designs communicate balance and feeling; concept designs share a plausible dream. A polished artifact can trap stakeholders in tiny comments; a wireframe can make them think bigger. Fidelity is an influence tool, not only a process stage.
Titles are going away (Brian Lovin, dvEwb1Ajkwo) — be the person who moves between disciplines; 2 undistracted hours/day is a competitive advantage.
Ship governance (Megan Choy, hKeDfupbA4U) — everyone can ship ≠ everything should ship; build systems that scale.
Pre-agreed regression tolerance for big bets (Katarina Batina, 0YjO7wShTkQ) — redesigns lose to years-optimized incumbents at first; negotiate the acceptable baseline reset before the work.
Data narratives are contestable (Batina) — when a metric is read against your work, contest the reading, not the number.
UI engineers report to design (Briggs, Dive Club) — staff them as translators; team design quality is capped by its UI engineers. Pairing pattern + his anti-deadline stance (his position, not house consensus) → references/influence-field-notes.md.
Team-of-one influence (Buley)
For the only UX/design person on a team, or an org too immature for formal process (method catalog routes to user-research and design-prototyping). Full quotes: references/influence-field-notes.md.
Principles over process. Pushing process on an immature org is a distraction — "invite non-UX people into your process. Listen with genuine curiosity. Make ideas tangible."
Good enough is an influence tool. "Good enough to have a conversation about it" beats polished proof; imperfect artifacts invite participation.
Run a Listening Tour before the big pitch. It lays bare others' ideas, hopes, politics; share what you heard before prescribing.
Negotiate size, not permission. Offer large study vs. small informal evaluation — the question becomes how much UX, not whether.
Build the pyramid, not the executive-only pitch. "Build an army of supporters throughout the organization"; the lunch date is the durable unit of influence.
Checklist
Pre-Figma gate answered: business problem, user problem, success measure?
Pitch leads with the business problem in one sentence, with numbers?
Pitch has one Commander’s Intent, a concrete customer/team scene, and human-scale numbers?
Three layers of evidence + implementation plan (phases, MVP, success metrics, rollback)?
Objections anticipated — and the stakeholder's goal/motivation decoded first?
Pushback handled as ask-why → reframe-to-goal → propose-a-test, using the conflict script?
Roles negotiated with the PM, with a quarterly retro — not assumed?
Wins doc current (one SBI entry/week), invisible work included?
Manager relationship aligned explicitly: what success looks like for them, for you, and for the relationship?
Bet ledger current: what you pushed for, lost, won, and later learned?
Influence mode chosen deliberately: advisor for frames, solver for proof?
Conflict reframed around shared goal + different assumptions?
Timing window chosen or urgency manufactured?
If solo/early UX: listening tour done, informal supporters named, and the ask framed as scope rather than permission?
Relationship to other skills
people-management — running your own team of direct reports (1:1s, feedback, careers, hiring/firing process, team comms); this skill is influence up/sideways and your own promotion.
personal-growth — broader reflection on the user's own life/work fit, goals, recovery, and whether to persist or change direction; this skill owns promotion, managing a manager, and stakeholder influence inside the current organization.
product-decision-making — whether the proposal is well-framed: user problem, data, success measure, optimization vs new behavior, and owner-by-outcome. This skill translates that judgment into buy-in and org movement.
user-research — runs the studies whose findings power your business case; its research-as-leadership tactics feed this skill's credibility ladder.
design-prototyping — owns session mechanics and pitch presentation craft (how to structure a review, fidelity, demos); this skill owns the business case and political context around the pitch.
building-in-public — influence with an external audience; this skill is influence inside the org.
design-craft — defines the quality bar; this skill gets the org to fund it.
design-systems — DS governance/adoption mechanics; this skill supplies the four-lens business case for DS investment.
write-clear-prose — the writing layer for briefs, self-reviews, and pitch docs.
Staleness note (Kevin's rule): auto-captioned 2022–2025 videos; AI-tooling asides and the Claude-skill packaging decay fastest — the framing frameworks (impact, strategic narrative, four lenses, SBI) and partnership scripts are the durable layer. One garbled caption (the SBI impact number in Up07knsch7I) is paraphrased, not quoted.
