A library of AI skills you can drop into your work.
# Audience AI Skill Define Area · Audience Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Audience skill helps teams decide who they are learning from, and how much to trust each source. It is the second block in the Define area of Glare's Decision Map, after User Needs and before Collecting. Teams have personas, ad segments, CRM lists, and panels, and they don't all mean the same thing. The skill asks one question first: **where did this signal come from, and what kind of decision can it support?** | Source | How it's built | Credibility for design | |---|---|---| | Helio Audience Groups | Poll questions that group people by behavior and motivation | High | | CRM / Lifecycle Segments | Purchase history, tenure, activation milestones | Medium | | Advertising Segments | Demographic and interest targeting | Low | | Internal Assumptions | Team expertise and stakeholder belief | Lowest: a starting point only | The skill also places each voice in one of four audience types, built in this order: Project Team (intent), Stakeholders (alignment), Participants (validation before launch), Customers (proof in real use). Internal voices guide; external voices validate. ### The Behavior-Over-Label Rule Define people by what they do, not by a title. "Manager" says little. "Approves budgets for a team of ten or more" says how that person works and what they need. If the source can't be named, the skill can't rate it, and it stops to ask. ## 2. Business Benefit When every source is weighted the same, work drifts toward whoever speaks loudest. Naming the source keeps a targeting segment from passing as user proof. This helps teams: - know which evidence can support a design decision - recruit the right people for Collecting and Testing - avoid averaging groups into a finding that fits none of them - carry the right level of confidence into Measure ## 3. Skill Output When used correctly, the skill produces an Audience Definition Card. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Name | Habitual mobile bankers | | Type | Participants now; Customers after launch | | Source | Helio Audience Group, checked against a CRM segment | | Defining Attributes | Log in three or more times a week; mostly check balance and recent transactions; use the app, not the website | | User Need | Findable (from User Needs) | | Connected UX Metrics | Success, Comprehension | | Credibility | High: behavioral attributes from a named source | | Gaps | No data yet on users who rely on shortcuts | | Failure Mode to Watch | Using the ad team's "young professionals" segment as the design audience. It was built for reach, not for how people use the app. | | Next Step Handoff | → glare-define-collecting to gather evidence from this audience | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Source Entry: Rate what you have *"We have a marketing persona, a CRM export, and last month's Helio panel for our mobile banking app. Using the glare-define-audience skill, name the source type of each, rate its credibility for a home screen redesign, and tell us which to use."* **Why this works:** It names each source before anyone treats it as proof. **Best for:** - teams with many overlapping segments - kicking off Define ### Prompt 2: Attribute Entry: Swap labels for behavior *"Our audience for the mobile banking dashboard is 'millennials who bank on mobile.' Using glare-define-audience, rewrite this with behavioral attributes and write the Audience Definition Card."* **Why this works:** Behavioral attributes make the audience testable and comparable. **Best for:** - personas built on demographics - recruiting for a study ### Prompt 3: Conflict Entry: Settle whose input counts *"Our stakeholders want a promotions-first home screen, but participant feedback points the other way. Using glare-define-audience, place each voice in the four audience types and tell us how much weight each should carry for this decision."* **Why this works:** It separates intent and alignment from validation, so the team knows which input settles the question. **Best for:** - stakeholder disagreements - conflicting feedback --- *Glare Framework · glare-define-audience · Define Area. Handoffs: glare-define-user-needs · glare-define-collecting · glare-define-situations · glare-measure-hunches (who to recruit)*
# Collecting AI Skill Define Area · Collecting Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Collecting skill helps teams gather evidence about their audience openly, before anyone has a hunch. It is the third block in the Define area of Glare's Decision Map, after User Needs and Audience and before Situations. Collecting reaches the defined audience through sources that can represent them, gathers more than one kind of evidence, and records where each piece came from. It works in four steps. | Step | What it does | Output | |---|---|---| | 1. Name what you want to learn | Keep the learning open, but aimed at a need or an audience | An open learning goal | | 2. Choose sources | Find sources that can reach these people, and rate each one | Named sources, with gaps | | 3. Gather | Collect more than one kind of evidence, each tied to its source | Gathered evidence | | 4. Map what you know | Record what was gathered and how far to trust it | A Source Map, handed to Situations | Techniques come in two kinds: open-ended gathering (interviews, a day in the life, diary studies, needfinding surveys, community listening, support tagging, Glare MCP evidence) and ongoing monitors (funnel review, drop-off mapping, feature adoption, retention reviews, and more). Pair at least two kinds of evidence, such as what people do and what they say. ### The No-Hunch Rule Collecting comes before any hunch. If the team already believes something about a concept, it isn't Collecting; write it as a hunch in Measure and test it. An interview with no hunch behind it is Collecting; an interview that probes a hunch is Testing. An analytics feed is Collecting; reading it against a hunch is Testing. ## 2. Business Benefit Most teams form beliefs from whatever is closest: one loud customer, last quarter's survey. Collecting widens the base on purpose, so the hunches that follow rest on more than one anecdote. This helps teams: - see what they already know before spending on new research - trace any later signal back to where it started - catch changes early through ongoing monitors - use evidence from the Glare MCP to learn more for less ## 3. Skill Output When used correctly, the skill produces a Source Map. It holds what was gathered and how far to trust it, not hunches or signals. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Learning Goal | What gets in the way when habitual users check their money on mobile? | | Audience | Habitual mobile bankers (from Audience) | | Sources | Product analytics, session recordings, support tickets, app store reviews, Glare MCP evidence on mobile banking | | Kinds of Evidence | What people do (analytics, recordings) and what they say (tickets, reviews) | | What We Know | 38% of home screen sessions end without an action; recordings show scrolling past promos; tickets ask where the balance went | | Trust | Analytics and recordings: strong. Reviews: weaker, skewed to unhappy users | | Gaps | Nothing yet from users who rely on shortcuts | | Failure Mode to Watch | Writing "the promo cards are the problem" into the Source Map. That is a hunch; it belongs in Measure. | | Next Step Handoff | → glare-define-situations to read the evidence for recurring conditions | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Open Entry: Plan open learning *"We want to understand how habitual users check their money in our mobile banking app, and we have no hunch yet. Using the glare-define-collecting skill, write the learning goal, pick sources that reach these users, and tell us which two kinds of evidence to pair."* **Why this works:** It keeps the learning open while aiming it at a real audience and need. **Best for:** - new products or audiences - teams that jump to testing too early ### Prompt 2: Inventory Entry: Map what already exists *"We have analytics, a year of support tickets, and app store reviews for our mobile banking dashboard. Using glare-define-collecting, build a Source Map: what each source tells us, how far to trust it, and where the gaps are."* **Why this works:** Most teams have more evidence than they think. Mapping it first avoids paying for research they already have. **Best for:** - teams with scattered data - before commissioning a study ### Prompt 3: Boundary Entry: Is this Collecting or Testing? *"We want to interview mobile banking users about the new home screen. Using glare-define-collecting, tell us whether this is Collecting or Testing, and how to set it up either way."* **Why this works:** The line is the hunch. The skill routes the work to the right place so the interview answers the right question. **Best for:** - interview and survey planning - mixed research plans --- *Glare Framework · glare-define-collecting · Define Area. Handoffs: glare-define-audience · glare-define-situations · glare-measure-hunches (a belief has formed) · glare-measure-questioning (Testing)*
# Situations AI Skill Define Area · Situations Block · Decision Map · Oct 23, 2026 (v1.0) --- ## 1. What the Skill Does The Situations skill helps teams name the recurring conditions behind what users do. It is the last block in the Define area of Glare's Decision Map, where User Needs, Audience, and Collecting come together. Situations judge **credibility**: is this really what happens to people like them? They own **Need** signals, which show that a real need shows up in real conditions. A situation is named from recurring Need signals; it is synthesis, not a signal on its own. Every situation has four parts. | Part | What it captures | |---|---| | What triggered it | The condition, pressure, or goal that set the situation in motion | | What they are trying to do | The progress the user wants in this context | | What stands in their way | The friction, uncertainty, or constraint that shapes behavior | | What success means here | How the user defines success, and what failure costs them | User Needs names the category, such as Findable. A situation is one level more concrete: how that need shows up for specific people, in a specific moment. ### The Three Checks and the Ladder | Check | What it asks | |---|---| | Source spread | Does the condition show up in more than one kind of source? | | Lenses that agree | Do at least two of the four lenses point to the same condition? | | Behavior behind the text | Is there something people did, not only something they said? | The checks place a situation on the confidence ladder: Weak, Possible, Emerging, Recurring, or Validated. A situation goes to Measure at Recurring. The last check is credibility itself: when you play the situation back to people in it, do they recognize it? One data point is an observation; never name a situation from a single lens or source. ## 2. Business Benefit A metric says what happened; a situation says why. The same drop in completion means one thing for trial users and another for habitual users. Without the situation, teams fix the symptom and leave the condition in place. This helps teams: - give metrics the context that makes them meaningful - align teams on what is actually happening, with less debate - spot recurring problems before they scale - hand Measure a credible problem worth writing hunches about ## 3. Skill Output When used correctly, the skill produces a named situation with its rung on the ladder. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Situation | The quick balance check | | What Triggered It | About to make a purchase or pay a bill, often between other tasks | | What They Are Trying to Do | Confirm how much money they have, in seconds | | What Stands in Their Way | Promo cards above the balance; extra scrolling on a small screen | | What Success Means Here | Seeing the balance at a glance. Failure means leaving the app, guessing, or calling support. | | Need Signals | Analytics: 38% of home screen sessions end without an action (behavior). Recordings: scrolling past promos (behavior). Tickets: "where did my balance go?" (said). All tagged Need | | Checks | Three kinds of source; two lenses agree; behavior behind the text | | Rung | Recurring: ready for Measure | | Credibility Check | Play it back to five habitual users before writing hunches | | Failure Mode to Watch | Naming the situation from support tickets alone. That is one source and only what people said. | | Next Step Handoff | → glare-measure-hunches to write hunches about concepts that address this situation | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Pattern Entry: Name the situation *"We have analytics, session recordings, and support tickets about our mobile banking home screen. Using the glare-define-situations skill, read them for recurring conditions, name the situation with its four parts, and place it on the confidence ladder."* **Why this works:** It turns scattered evidence into one named condition the team can design against. **Best for:** - teams with lots of signals and no clear story - closing out Define ### Prompt 2: Credibility Entry: Is this situation real? *"Our team believes habitual mobile banking users struggle to find their balance. Using glare-define-situations, run the three checks, tell us which rung this is on, and what evidence would move it to Recurring."* **Why this works:** It tests a belief about users against source spread, lenses, and behavior before Measure builds on it. **Best for:** - stakeholder beliefs about users - deciding whether Define is done ### Prompt 3: Split Entry: Same need, different situations *"Both habitual users and new users struggle with our mobile banking home screen. Using glare-define-situations, tell us whether this is one situation or two, and how the four parts differ for each group."* **Why this works:** The same need can produce different situations. Splitting them keeps Measure from testing one fix for two problems. **Best for:** - products with very different user groups - mixed results across segments --- *Glare Framework · glare-define-situations · Define Area. Handoffs: glare-define-collecting (more evidence) · glare-define-user-needs · glare-define-audience · glare-measure-hunches · glare-measure-concepts · glare-design-signals*
# UX Metrics AI Skill Define Area · UX Metrics Block · Decision Map --- ## 1. What the Skill Does The UX Metrics skill helps teams choose the right numbers to prove their design work is working. It sits inside the Define area of Glare's Decision Map. This is where teams decide what to measure before they start collecting data — not after. Most teams measure too late or measure the wrong thing. They wait for analytics that arrive after the sprint is over. Or they collect numbers that look good in a slide deck but never guide a real decision. The UX Metrics skill fixes that by helping teams pick metrics with purpose. Every metric belongs to one of three types. A good measurement plan includes all three. | Type | What it measures | Examples | |---|---|---| | Attitudinal | How users feel | Trust, Satisfaction, Desirability, Sentiment | | Behavioral | What users do | Completion, Comprehension, Effort, Engagement | | Performance | How well the experience works | Time on Task, Error Rate, Drop-off, Retention Rate | Use one of each. A single metric distorts the picture. High satisfaction with low completion means users like the idea but cannot finish. Strong performance with low engagement means the system works but nobody cares. Together, all three tell the real story. **The Metric Quality Rule** Not all metrics are useful. Teams often track numbers that feel important but never change what they build. The rule is simple: if a metric cannot tell you what to do differently, it is not worth tracking. Before committing to any metric, run it through four questions: - Can everyone on the team explain it in one sentence? - Can it be compared over time or across versions? - Is it a rate or ratio, not just a raw count? - Does it measure what users actually do, not just what they say? If the answer to any of these is no, replace it. --- ## 2. Business Benefit When teams choose metrics with discipline, design earns credibility. Decisions move faster because they are grounded in evidence, not debate. This helps teams: - prove that design work changed user behavior - stop tracking numbers that look good but say nothing - give product and leadership a shared language for decisions - catch problems before launch, not after - connect design outcomes to business results Metrics chosen with care become the evidence that earns the next yes. --- ## 3. Skill Output When used correctly, the skill produces a clear metric plan for a product or workflow. The plan shows: - which three metrics to track (one per type) - whether each metric is a leading or lagging indicator - which stage each metric belongs to: predictive, proxy, or analytics - any mismatches to watch for between metric types The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Attitudinal Metric | Trust — do users feel confident the balance shown is accurate? | | Behavioral Metric | Completion — can users locate transaction history within two taps? | | Performance Metric | Time on Task — how long does it take to find and act on a recent transaction? | | Leading Indicator | Comprehension score from prototype testing (collected before launch) | | Lagging Indicator | Session abandonment rate (confirmed after launch) | | Mismatch to Watch | High satisfaction + low completion = users feel good about the app but cannot finish the task. Fix the flow before assuming the experience is working. | | Next Step Handoff | → glare-define-collecting to choose the right techniques and tools for collecting each metric | The output connects directly to the other Define blocks: - User Needs tells you what each metric should prove - Audience tells you whose behavior you are measuring - Collecting tells you how to gather the data --- ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. --- ### Prompt 1 — Diagnostic Entry: Fix a broken metric plan "We're updating our mobile banking dashboard and our current metrics are monthly active users and app store rating. Using the glare-define-ux-metrics skill, tell us whether these are the right metrics to track, apply the four quality principles to each one, and recommend a replacement trio — one attitudinal, one behavioral, one performance — that would actually guide our next design decision." **Why this works:** Monthly active users and app store ratings are common vanity metrics. They count things without explaining what to do next. This prompt uses the quality filter to replace them with metrics that can change how the team builds. **Best for:** - auditing an existing metric plan - sprint kickoffs where the success criteria feel vague - any situation where teams are measuring activity instead of outcomes --- ### Prompt 2 — Timing Entry: Choose the right metric for the right stage "We are about to run usability testing on our mobile banking dashboard before launch. Using glare-define-ux-metrics, help us identify which metrics should be collected now as leading indicators, which ones we should plan to collect post-launch as lagging indicators, and how to use each to make a decision at the right time." **Why this works:** Teams that only track post-launch analytics are always learning too late. This prompt uses the leading vs. lagging framework to build a measurement plan that catches problems early and confirms results after. **Best for:** - pre-launch research planning - setting up a test with clear success criteria - building a measurement timeline across a product sprint --- ### Prompt 3 — Mismatch Entry: Diagnose a confusing result "After our last round of testing on the mobile banking dashboard, satisfaction scores were high but task completion on the transaction history flow dropped to 61%. We are not sure what to do with this. Using glare-define-ux-metrics, explain what this mismatch means, what it tells us about where the experience is breaking down, and which metric we should add to diagnose the root cause." **Why this works:** Metric mismatches are one of the most common signs that a team is measuring the wrong thing or missing part of the picture. This prompt uses the diagnostic mismatch model to turn a confusing result into a clear next step. **Best for:** - making sense of conflicting data - preparing a findings summary for a design review - deciding which metric to add before the next round of testing --- *Glare Framework · glare-define-ux-metrics · Define Area* *Handoffs: glare-define-user-needs · glare-define-audience · glare-define-collecting · glare-measure*
# User Needs AI Skill Define Area · User Needs Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The User Needs skill helps teams name what users are actually trying to accomplish, and where the experience starts to break down, before anyone proposes a solution. It is the first block in the Define area of Glare's Decision Map, followed by Audience, Collecting, and Situations. Needs are rarely stated outright. They show up in repeated behavior, hesitation, confusion, workarounds, and recurring friction. The skill sorts what the team is seeing into five layers and 20 needs. | Layer | Focus | User Needs | |---|---|---| | Basics | Foundational usability and access | Useful, Usable, Findable, Accessible | | Trust | Reliability and confidence | Credible, Secure, Reliable, Intuitive | | Personal | Relevance and context | Inclusive, Adaptable, Connected, Insightful | | Impact | Progress and meaningful outcomes | Valuable, Sustainable, Efficient, Scalable | | Feelings | Emotion and long-term connection | Desirable, Delightful, Engaging, Empowering | The layers build on each other: foundational needs usually come before emotional ones. They are not a rigid sequence, though. Most experiences have several overlapping needs at once. ### The Say-Do Rule What users say is not always what they need. If users ask for something but don't act on it, it is probably a want, not a need. Check every stated need against what people do. For example, users say they want a simpler layout, but recordings show they leave when they can't find their balance. The need is Findable, not a new look. ## 2. Business Benefit Teams often have interviews, tickets, analytics, and opinions, but no shared way to read them. Named needs give everyone the same reference point, so the team fixes the real problem instead of the loudest one. This helps teams: - turn scattered feedback into a few clear needs - separate needs from wants before building - spot foundational gaps before polishing higher layers - tie each need to the UX metrics that would show it improving - give Measure, Focus, and Lead the same starting point ## 3. Skill Output When used correctly, the skill produces a needs brief. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Layer and Need | Basics: Findable (primary); Trust: Credible (watch) | | Need Statement | Habitual users need to see their balance and recent transactions at a glance, so they can check their money without hunting for it | | Evidence | Session recordings show users scrolling past promo cards; support tickets ask where the balance went after the last update | | Say vs. Do | Users ask for a "simpler layout"; behavior shows the gap is finding the balance, not the look | | UX Metrics | Success and Comprehension (leading); sessions ending without an action (lagging) | | Failure Mode to Watch | Jumping to Feelings, such as animations or delight, while a Basics need is still failing | | Next Step Handoff | → glare-define-audience to name whose experience this need comes from and how far to trust each source | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Symptom Entry: Start from a problem in the data *"Users of our mobile banking dashboard leave within 30 seconds without doing anything. Using the glare-define-user-needs skill, look across the five layers, name the needs most likely failing, and tell us what behavior would confirm each one."* **Why this works:** Starting from behavior keeps the team from redesigning the look when the real problem is finding or trusting something. **Best for:** - post-launch problems - redesign debates with no evidence ### Prompt 2: Validation Entry: Check a stated need *"Users say they want the mobile banking app to remember their preferences. Using glare-define-user-needs, apply the say-do rule, tell us whether this is a need or a want, and which foundational needs must hold first."* **Why this works:** It checks a feature request against behavior before the team invests in it. **Best for:** - roadmap planning - feature requests without user evidence ### Prompt 3: Metric Entry: Tie needs to measures *"We've named Findable and Credible as the needs on our mobile banking dashboard. Using glare-define-user-needs, tie each one to its UX metrics, say which are leading and which are lagging, and flag the failure mode to watch."* **Why this works:** The needs are already named, so the prompt focuses on how the team will know they are improving. **Best for:** - test planning - handoffs to Measure --- *Glare Framework · glare-define-user-needs · Define Area. Handoffs: glare-define-audience · glare-define-collecting · glare-define-situations · glare-ux-metrics · glare-measure-concepts*
# Bets AI Skill Focus Area · Bets Block (formerly Decisions) · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Bets skill helps teams commit to one of several compared options. It is the last block in the Focus area of Glare's Decision Map. By now the initiative is framed, the variable is named, and Comparing has measured every option the same way. Bets judge **desirability**: of the options, which outcome is worth committing to? They own **Prefer** signals, which show which option people would choose. A bet names the option it backs, the evidence behind it, what it costs, and the outcome it expects. The skill still runs as glare-focus-decisions so existing installs keep working. This block used to be called Decisions. Bets need options. One option tested once is not a bet; it is a next move in Measure → Findings, which owns the five decision types (Implement, Refine Design, Test Iteration, Revisit Later, Do Not Pursue). ### The Prefer Rule No Prefer signal, no bet. Prefer must be measured the same way for every option, as a side-by-side choice, likely to choose (top-two box), trust, or pull-away. Desirability words or the Desirability UX metric alone don't count. Without Prefer, the next move is another comparison round with the Prefer question added. ### The Winner Rule The readout has a Use line (which option worked better) and a Prefer line (which option people would choose). The call comes from both: | Prefer | Use | Call | |---|---|---| | Wins | Wins or ties | Implement is open | | Ties | Wins | May go ahead, labeled "won on Use, preference tied"; the bet carries that risk | | Wins | Loses | A split: Refine Design or Test Iteration, never Implement | | Loses | Any | Not this option | | Doesn't separate | Doesn't separate | No winner: Test Iteration, or the business goal decides and the bet says so | A strong signal on its own is not enough. The call comes from the rule. ## 2. Business Benefit Without a bet, comparisons stay in decks and the same options come back to every meeting. A bet ends the loop: it says what the team backs, what it gives up, and what it expects to happen, with a date to check. This helps teams: - commit to a direction backed by what users chose, not only what they could do - name the cost of the choice up front - size the commitment to the evidence - give Lead a predicted outcome it can check later - rank several winning options with real lift, not guesses ## 3. Skill Output When used correctly, the skill produces a bet record. Tactical bets (a version, flow, or message) are judged on design KPIs with a product KPI to watch. Strategic bets (a direction to invest in) need a credible line to business KPIs. When being wrong is expensive, reduce risk first with a smoke test, pilot, or proof of concept. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Initiative | Optimize Navigation & Discovery on the home screen, for habitual users | | Options and Variable | Current vs. redesigned home screen (Iteration) | | Use Line | First-click success on balance: Redesign 79%, Current 61% | | Prefer Line | Likely to choose (top-two box): Redesign 68%, Current 49% | | Call | Prefer wins and Use wins: Implement is open | | Cost | Power-user shortcuts move off the home screen | | Bet Size | Tactical; low risk, reversible in one release | | Business Goal | Reduce session abandonment | | Predicted Outcome | Sessions that end without an action drop from 38% to under 30% within 60 days of launch | | Check Date | 60 days after launch, by Lead → Results | | Owner and Next Step | Design lead moves the redesign into release planning; research tracks the power-user gap | | Failure Mode to Watch | Calling Implement on first-click success alone. That is a Use signal; without Prefer it is not a bet. | When several options clear the winner rule, the skill ranks them with RICE, ICE, weighted scoring, MoSCoW, Eisenhower, or Now/Next/Later, scored with measured Use and Prefer lift. ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Readiness Entry: Is this ready for a bet? *"We tested the redesigned mobile banking home screen against the current one. The redesign won first-click success, 79% vs 61%. Using the glare-focus-decisions skill, tell us whether we can make a bet yet, and if not, what is missing."* **Why this works:** It checks for a Prefer signal before anyone commits. Here there is none, so the skill sends the team back for one more round with the Prefer question added. **Best for:** - stakeholders pushing to ship on one strong number - reviewing a readout before a decision meeting ### Prompt 2: Split Entry: Use and Prefer disagree *"Version A of our mobile banking dashboard won first-click success, 79% vs 71%. But more people would choose Version B, 66% vs 54%. Using glare-focus-decisions, apply the winner rule, name the call, and tell us what the next round should change."* **Why this works:** A split is not a win. The winner rule says Refine Design or Test Iteration, never Implement, and the skill names what a sharper round would need to show. **Best for:** - results where two metrics point different ways - teams tempted to quote the number they like ### Prompt 3: Record Entry: Write the bet *"We are backing the redesigned mobile banking home screen. Using glare-focus-decisions, write the bet record: the Use and Prefer lines, the call, the cost, the bet size, the predicted outcome with its metric and window, the check date, and the owner."* **Why this works:** A bet without a predicted outcome can't be checked. The record gives Lead → Results something to test later. **Best for:** - closing out a Focus round - handing a decision to product or leadership - building a history of bets and how they turned out --- *Glare Framework · glare-focus-decisions (Bets) · Focus Area. Handoffs: glare-focus-comparing · glare-focus-methods (Variables) · glare-focus-initiatives · glare-measure-findings (single test) · glare-lead-mapping · glare-lead-results (outcome check)*
# Comparing AI Skill Focus Area · Comparing Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Comparing skill helps teams repeat the same measurement while one thing changes, then read the difference. It is the third block in the Focus area of Glare's Decision Map, after Initiatives and Variables and before Bets. Measure designs one test. Comparing runs that test again across versions, competitors, segments, time periods, or another variable. A single score only means something next to what it is compared with. Is 72% task success good? Only a comparison can say. Every comparison measures two things for each option, the same way: - **Use:** which option works better, such as first-click success, task success, comprehension, or errors - **Prefer:** which option people would choose, asked as a side-by-side choice, likely to choose (top-two box), trust, or pull-away Without a Prefer measure, a comparison can't lead to a bet. Desirability words alone, such as "it looks nicer," don't count as Prefer. The skill helps the team pick a comparison design. Each design says what its result can prove. | Design | How it works | What the result can prove | |---|---|---| | Randomized A/B | Two versions shown to randomly split audiences | The design caused the difference, when significant | | Randomized multivariate | Several combinations of elements tested at once | Which combination works best, when significant and the sample is large | | Within subjects | The same people see every option | A real difference for the same people; vary the order | | Separate samples | Different people see each option | A real difference between options; cause not shown | | Before and after | The same measure taken again after a change | An observed change, not proof of cause | | Benchmark | Your experience measured next to others or a standard | Relative standing only | | Segment split | One study, with results split by group | Groups differ; observational | ### The Same-Measurement Rule Only one thing changes. The UX metric, the technique, the kind of audience, and the Prefer question stay the same on every side. If two things change at once, split the comparison or name the confound in the readout. A comparison that changes the audience and the design together can't say which one moved the number. ## 2. Business Benefit Decisions stall when teams look at results one at a time. A concept scores well, but is it better than what is live today? Comparing answers that, so the team has a reason to choose. This helps teams: - see what got stronger or weaker between versions - know whether a difference is real or noise - name the tradeoff instead of chasing the highest score - see where one audience responds differently from another - bring Bets a readout it can commit to ## 3. Skill Output When used correctly, the skill produces a comparison readout. The readout shows: - what was compared, and the one variable that changed - a **Use line**: which option worked better, with every option's number side by side - a **Prefer line**: which option people would choose, and how it was asked - the tradeoff that showed up - what the design can and can't prove - the next step Every signal in the readout is tagged Need, Use, Prefer, or Adopt. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Decision It Supports | Which home screen to take into the next release | | Variable | Iteration: current home screen vs. redesigned home screen | | Held Fixed | Same task (find your balance), same audience (100 habitual mobile banking users per version, Helio), same Prefer question | | Design | Separate samples: shows a real difference between versions, not what caused it | | Use Line | First-click success on balance: Redesign 79%, Current 61% | | Prefer Line | Likely to choose this home screen (top-two box): Redesign 68%, Current 49% | | Tradeoff | The redesign hides shortcuts that power users rely on | | Baseline and Cadence | Current version is the baseline; repeat after launch and again at +90 days | | Failure Mode to Watch | Changing the layout and the copy between versions in the same round. If two things change, the readout can't say which one moved the number. | | Next Step Handoff | → glare-focus-decisions (Bets) to commit to the redesign and set its predicted outcome | The output connects to the other Focus blocks: - Initiatives names the need, goal, and metric the comparison serves - Variables names the one thing that changes - Bets uses the Use line and the Prefer line to call a winner, or names a tie When Use and Prefer disagree, the skill names it and does not force a winner. Bets applies the winner rule. ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Setup Entry: Design a fair comparison *"We want to compare our current mobile banking home screen with a redesign before the next release. Using the glare-focus-comparing skill, name the variable, list what must stay fixed, pick the comparison design, and tell us how to measure both Use and Prefer the same way for each version."* **Why this works:** It sets up the repeat before any data comes in, so the result can actually be compared. It also makes sure Prefer is measured, so the comparison can lead to a bet. **Best for:** - planning a test round with more than one version - A/B test design - making sure a comparison can support a decision ### Prompt 2: Reading Entry: Make sense of a mixed result *"Our mobile banking comparison came back. The redesign won first-click success, 79% vs 61%. But likely to choose was 52% for the redesign and 55% for the current version. Using glare-focus-comparing, write the readout with a Use line and a Prefer line, name the tradeoff, say what this design can and can't prove, and tell us whether this is ready for a bet."* **Why this works:** When Use and Prefer point different ways, teams tend to quote the number they like. This prompt puts both lines side by side and names the tie instead of forcing a winner. **Best for:** - results where two metrics disagree - preparing for a review where someone will push back - deciding whether another round is needed ### Prompt 3: Repeat Entry: Track change over time *"We launched the new mobile banking home screen in March. Using glare-focus-comparing, set up a before-and-after comparison against our pre-launch baseline: what to measure, when to repeat it, what size of change counts as real, and what this design can and can't prove about the launch."* **Why this works:** Before-and-after shows an observed change, not proof of cause. This prompt sets the baseline and cadence up front and keeps the readout honest about what it can claim. **Best for:** - post-launch checks - seasonal or campaign changes - building a benchmark the team repeats every quarter --- *Glare Framework · glare-focus-comparing · Focus Area. Handoffs: glare-focus-initiatives · glare-focus-methods (Variables) · glare-focus-decisions (Bets) · glare-measure-findings (single test, no options)*
# Initiatives AI Skill Focus Area · Initiatives Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Initiatives skill helps teams name the area of work that deserves attention before anything is varied or compared. It is the first block in the Focus area of Glare's Decision Map. This is where scattered requests (redesign this page, fix this flow, test this feature) become one focused effort with a clear goal. Without a clear initiative, everything feels equally important. Tests run without a shared target, and results come back that don't connect to a decision. An initiative gives the team a container for signals, variables, comparisons, and bets. A strong initiative connects three things. | Part | What it names | |---|---| | User Need | What users are trying to do and where they get stuck | | Business Goal | What outcome the business is trying to move | | Measurable Part of the Experience | The specific area that needs to improve | All three must be present. An initiative with no user need is a roadmap item. One with no business goal is a design exercise. One with no measurable part of the experience is a direction without a target. ### The Framing Rule Don't choose a variable until the initiative answers six questions: - What part of the experience needs attention? - Who is affected by it? - What user need is involved? - What business goal does it connect to? - What signal would show progress? - What decision needs to be made? If any answer is missing, the initiative isn't ready. Frame it before moving on. The skill uses ten common initiative types as starting points: Simplify Identity & Access, Accelerate Onboarding & First-Use Success, Optimize Navigation & Discovery, Streamline Checkout & Payment, Increase Engagement & Retention, Personalize User Experiences, Strengthen Collaboration & Communication, Elevate Marketing & Growth Pages, Improve Content & Support, and Expand Device & Context Reach. Each points to a different kind of friction, business pressure, and signal pattern. ## 2. Business Benefit A clear initiative keeps design work on the problems that matter most to users and the business, and stops tests that produce results nobody can act on. This helps teams: - connect scattered requests to one shared goal - choose the right variable and metric for the work - group many concepts under one outcome without losing scope - explain findings to product and leadership - give each bet a goal it can be checked against later ## 3. Skill Output When used correctly, the skill produces an initiative brief that answers the six framing questions. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Initiative Type | Optimize Navigation & Discovery | | Area of Experience | Mobile banking dashboard home screen | | Audience | Habitual users who log in three or more times a week | | User Need | Findable: users need to see their balance and recent transactions without extra taps | | Business Goal | Reduce session abandonment | | Signal of Progress | First-click success on balance (Use), likely to choose (Prefer), and sessions that end without an action | | Concepts in Scope | Redesigned home screen, summary card, pinned transactions | | Decision Needed | Which home screen to take into the next release | | Failure Mode to Watch | Running two initiatives at once. If the team can't pick one, the work hasn't been scoped yet. | | Next Step Handoff | → glare-focus-methods (Variables) to name the one thing that changes between measurements | The output connects to the other Focus blocks: - Variables uses the decision to name what changes - Comparing uses the need and metric to measure every option the same way - Bets uses the business goal to write a predicted outcome ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Diagnostic Entry: Start from scattered requests *"We've been asked to redesign the mobile banking home screen, fix the transaction history flow, and test a new summary card, all at once. Using the glare-focus-initiatives skill, map these requests to one initiative with one user need, one business goal, and one signal of progress."* **Why this works:** Three requests with no shared frame is the exact symptom this skill fixes. It pulls them toward one area of improvement the team can test and decide on. **Best for:** - sprint planning with too many directions - connecting a list of requests to a business outcome ### Prompt 2: Typing Entry: Name the initiative type *"We're improving the mobile banking dashboard and can't tell if this is a Navigation, Engagement, or Personalization initiative. Using glare-focus-initiatives, pick the type that fits, explain what it is built to improve, and tell us which signals that type usually needs."* **Why this works:** The wrong type leads to the wrong metrics and comparisons that don't support the real decision. **Best for:** - new Focus efforts without a clear category - projects where scope keeps growing ### Prompt 3: Readiness Entry: Check before choosing a variable *"Our initiative: help habitual users find their balance faster on the mobile banking home screen, to reduce session abandonment, tracked by first-click success. Using glare-focus-initiatives, check it against the six framing questions and tell us whether it's ready for Variables."* **Why this works:** Teams often leave out the decision or the audience. Here the decision is missing, and the skill catches it before a round of testing that can't support one. **Best for:** - checking an initiative before a research sprint - preparing an initiative brief for product or leadership --- *Glare Framework · glare-focus-initiatives · Focus Area. Handoffs: glare-focus-methods (Variables) · glare-focus-comparing · glare-focus-decisions (Bets) · glare-define · glare-measure*
# Variables AI Skill Focus Area · Variables Block (formerly Methods) · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Variables skill helps teams name the one thing that changes between measurements. It is the second block in the Focus area of Glare's Decision Map, after Initiatives and before Comparing. Measure runs one test. Focus runs that test again, and the variable says what is different from one run to the next: a new version, a competitor, a different audience, a later month. Everything else stays the same, so the team can tell what moved the number. The skill still runs as glare-focus-methods so existing installs keep working. This block used to be called Methods. The skill works from 12 variables. | Variable | What changes between runs | |---|---| | Iteration | Versions of the same design | | Competitors | Your experience vs. others, on the same measure | | Timeline | The same measure across time periods: baseline, launch, +90 days | | Segments | Audiences, roles, cohorts, or need states | | Journeys | Steps or touchpoints, with the same metric at each | | Platforms / Devices | Mobile, desktop, tablet, or channel | | User Goals / Tasks | Different tasks on the same experience | | Feature Usage | Features compared on value, adoption, or use | | Geographies | Regions, markets, or languages | | User Lifecycle | New, returning, or expanding users | | Behavioral Triggers | Prompts or conditions that cause action | | Season | Timing or cycles that change needs, such as holidays or renewal windows | ### The One-Variable Rule Change one thing at a time. If the team wants to change two things, such as the layout and the audience, split it into two comparisons or name the confound up front. Start from the decision, not the test: ask "what would we need to see change to make this call?" The answer is usually the variable. ### Where the old methods went Methods used to mix four kinds of tool. The skill routes each named method to where it belongs now. | Kind | Examples | Where it lives now | |---|---|---| | Measuring: collects one data point you can repeat | Task success, first click, SUS, funnel review | Measure → Testing, or Define → Collecting for ongoing monitors | | Comparison designs: set up the repeat | A/B test, before and after, competitor benchmark, segment split | Focus → Comparing | | Mapping: shows what to vary, measures nothing | Journey mapping, JTBD, empathy mapping, service blueprint | Here, as ways to find the variable | | Prioritizing: ranks options once evidence is in | RICE, ICE, MoSCoW, Eisenhower, Now/Next/Later | Focus → Bets | ## 2. Business Benefit Most teams default to the method closest to their craft: another prototype test, another survey. Naming the variable first makes the team ask what the decision actually turns on. The right comparison might be a competitor, a segment, or a season, not another version. This helps teams: - pick comparisons that answer the real decision - avoid tests where two things changed and nobody can say which mattered - use data they already have before collecting more - stop running prioritization exercises when they need a measurement ## 3. Skill Output When used correctly, the skill produces a variable choice the team can hand to Comparing. It shows: - the decision the variable serves - the one variable, and the options along it - what stays fixed - data the team already has along this variable - any named methods, routed to where they belong The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Initiative | Optimize Navigation & Discovery on the home screen, for habitual users | | Decision It Serves | Which home screen to take into the next release | | Variable | Iteration | | Options | Current home screen and redesigned home screen | | Held Fixed | Task (find your balance), audience (habitual users), metric (first-click success), Prefer question | | Data Already in Hand | Analytics show 38% of sessions on the current home screen end without an action; use it as the baseline | | Methods Routed | Team asked for a RICE score: that ranks bets later, in Bets, once the comparison is in | | Failure Mode to Watch | Testing the redesign on new users and the current version on habitual users. That changes two things at once. | | Next Step Handoff | → glare-focus-comparing to set up the repeat and pick the comparison design | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Decision Entry: Find the variable *"Our initiative is to help habitual users find their balance faster on the mobile banking dashboard. We need to decide what to ship next quarter. Using the glare-focus-methods skill, ask what we would need to see change to make that call, name the variable, and list what must stay fixed."* **Why this works:** Starting from the decision keeps the team from defaulting to another version test when the real question is about a segment or a competitor. **Best for:** - kicking off a Focus round - teams that always run the same kind of test - initiatives with more than one possible comparison ### Prompt 2: Routing Entry: Sort a list of methods *"Our team wants to run journey mapping, an A/B test, a SUS survey, and a RICE exercise for the mobile banking dashboard. Using glare-focus-methods, tell us where each one belongs now, which one helps us find the variable, and what order to run them in."* **Why this works:** Teams often list methods that do different jobs. Sorting them shows which ones measure, which set up a comparison, which find the variable, and which rank bets afterwards. **Best for:** - research plans with a long list of methods - stakeholders who ask for a specific framework - untangling a plan that mixes measuring and ranking ### Prompt 3: Confound Entry: Check a planned comparison *"We plan to compare the redesigned mobile banking dashboard on iOS against the current version on Android. Using glare-focus-methods, check whether more than one thing is changing, and tell us how to split this into clean comparisons."* **Why this works:** It catches a comparison where two variables change together before any data is collected, when it is still cheap to fix. **Best for:** - reviewing a test plan before launch - comparisons across devices or regions - results nobody can explain --- *Glare Framework · glare-focus-methods (Variables) · Focus Area. Handoffs: glare-focus-initiatives · glare-focus-comparing · glare-focus-decisions (Bets) · glare-measure (Testing)*
# Business Goals AI Skill Lead Area · Business Goals Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Business Goals skill helps teams tie design work to the pressures leaders already track. It is the first block in the Lead area of Glare's Decision Map, followed by Workflows, Mapping, and Results. Every business runs on the same five pressures. The skill works from 20 measurable goals, four for each. | Pressure | Goals | |---|---| | Growth | Prove product fit, attract new customers, grow sales, gain market share | | Retention | Keep customers, reduce churn, earn loyalty, increase spend | | User Experience | Show value fast, drive adoption, improve customer satisfaction, prove launch success | | Efficiency | Work faster, lower support costs, scale efficiently, fund what matters | | Operations | Align strategy, lead with evidence, strengthen brand trust, prove design impact | Each goal is tracked at three KPI layers: **Design KPIs** (do users succeed in the moment?), **Product KPIs** (do they adopt and return?), and **Business KPIs** (does the company benefit?). ### The Quick Test A metric is a signal only if it passes three questions: Does it prove the design works for users? Does it show adoption or retention at the product level? Does it connect to growth, churn, or revenue at the business level? The skill also flags three pitfalls: chasing vanity metrics, mistaking correlation for causation, and reporting numbers without connecting them to an outcome. ## 2. Business Benefit Most metrics die in dashboards. Tying design work to a named pressure and goal gives every initiative and every bet a reason leaders care about, and something to check later. This helps teams: - anchor initiatives in a goal leaders already track - pick KPIs at all three layers, not just the top one - drop vanity metrics before they reach a deck - give each bet a business goal Results can check ## 3. Skill Output When used correctly, the skill produces a goal anchor. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Pressure | Retention | | Goal | Keep customers | | Design KPI | First-click success on balance | | Product KPI | Sessions ending without an action; monthly active use | | Business KPI | Account churn; support calls about balance | | Quick Test | Passes all three: users succeed, product use holds, churn and support cost are on the line | | Pitfall Avoided | App store rating, which moves without telling the team what to change | | Failure Mode to Watch | Naming revenue as the goal for a home screen change. The chain is too long to check; anchor to the nearest goal it can move. | | Next Step Handoff | → glare-lead-workflows to name the functions that own each KPI | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Anchor Entry: Pick the goal *"We're redesigning our mobile banking home screen. Using the glare-lead-business-goals skill, pick the pressure and goal this work serves, and name a Design, Product, and Business KPI for it."* **Why this works:** It ties the work to a goal leaders already track, before any test runs. **Best for:** - starting an initiative - work that came in without a stated goal ### Prompt 2: Quick Test Entry: Check a metric *"Our leadership wants to track app store rating and daily logins for the mobile banking redesign. Using glare-lead-business-goals, run the quick test on each and suggest replacements that pass."* **Why this works:** It separates metrics that guide decisions from ones that just look good. **Best for:** - dashboard reviews - OKR planning ### Prompt 3: Pitfall Entry: Check a claim *"Churn dropped the quarter we launched the new mobile banking home screen. Using glare-lead-business-goals, tell us what we can and can't claim, and what we'd need to show the redesign caused it."* **Why this works:** It stops a correlation from being reported as proof. **Best for:** - quarterly reviews - launch retrospectives --- *Glare Framework · glare-lead-business-goals · Lead Area. Handoffs: glare-lead-workflows · glare-lead-mapping · glare-focus-initiatives · glare-focus-decisions (Bets) · glare-measure-concepts*
# Mapping AI Skill Lead Area · Mapping Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Mapping skill helps teams trace how user success turns into business results. It is the third block in the Lead area of Glare's Decision Map, after Business Goals and Workflows and before Results. Mapping traces each bet's expected outcome chain from Focus. The skill builds the Chain of Proof: five rungs from what users need to what the business values. | Rung | Question it answers | Example | |---|---|---| | User Need | What are users trying to achieve? | "I want to finish onboarding quickly." | | User Outcome (Design KPI) | Did users succeed in the moment? | 85% completed setup without error | | Product Outcome (Product KPI) | Did that lead to adoption or return use? | Trial-to-paid conversion doubled | | Business Outcome (Business KPI) | Did the company benefit? | Pipeline revenue grew | | Business Goal | Which pressure did this support? | Growth | Each link between rungs gets a link type. The link type decides how strongly the team can claim the business moved. | Link type | What it means | What you can claim | |---|---|---| | Causal | A tested change moved the KPI (experiment, holdout, or staged rollout) | The change caused the move | | Correlated | The KPIs move together across users, accounts, or periods | They move together; cause not proven | | Sequential | A before-and-after read | The KPI moved after the change; other things may have too | | Linked: Defined | True by formula, such as pay rate as part of revenue | Certain, but says nothing about whether design moved it | | Linked: Stated or Planned | Someone says so, or a plan assumes it | Unproven; never report it as causal | ### The Link Rule Every link in the chain names its type, and the claim never goes past it. A Sequential link can say the KPI moved after the change; only a Causal link can say the change moved it. A Stated or Planned link is a hunch about the business, not proof. When a link is weak and the decision is big, the skill suggests a holdout or staged rollout to make it Causal. ## 2. Business Benefit Without mapping, progress stops at screens and features. With it, a single usability signal becomes part of a story leaders can trust, and the team is honest about how strong that story is. This helps teams: - show how a design change reaches a business goal - say exactly how strong each claim is - find the weak link before leadership does - give Results the links it needs to read an outcome ## 3. Skill Output When used correctly, the skill produces a Chain of Proof with a link type on every step. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | User Need | Findable: see my balance at a glance | | Design KPI | First-click success on balance (79% in test) | | Design → Product link | Sequential now; Causal if the launch is staged to half of users first | | Product KPI | Sessions ending without an action (baseline 38%) | | Product → Business link | Correlated: accounts with more abandoned sessions churn more, across past cohorts | | Business KPI | Account churn; support calls about balance | | Business → Goal link | Linked: Defined: churn is part of the Retention goal by definition | | Business Goal | Retention: keep customers | | Weakest Links | Design → Product is only Sequential until the staged launch; Product → Business is only Correlated | | Failure Mode to Watch | Reporting "the redesign cut churn." No link in this chain is Causal yet. | | Next Step Handoff | → glare-lead-results to check the bet's outcome against this chain | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Chain Entry: Build the Chain of Proof *"We're backing the redesigned mobile banking home screen to keep customers. Using the glare-lead-mapping skill, build the five-rung Chain of Proof and name the link type for each step."* **Why this works:** It shows the whole path from user need to business goal, and how strong each step is. **Best for:** - preparing a bet for leadership - quarterly impact reviews ### Prompt 2: Claim Entry: Check what we can say *"Our mobile banking team wants to report that the redesign reduced churn. Using glare-lead-mapping, check each link and tell us the strongest claim we can honestly make."* **Why this works:** It keeps the claim no stronger than the weakest link. **Best for:** - launch reports - board or exec updates ### Prompt 3: Upgrade Entry: Strengthen a weak link *"The link between abandoned sessions and churn in our mobile banking chain is only correlated. Using glare-lead-mapping, suggest how to make it causal before our next big bet."* **Why this works:** A holdout or staged rollout turns a correlation into proof when the stakes justify it. **Best for:** - high-stakes bets - teams whose impact claims get challenged --- *Glare Framework · glare-lead-mapping · Lead Area. Handoffs: glare-lead-business-goals · glare-lead-workflows · glare-lead-results · glare-focus-decisions (Bets) · glare-ux-metrics*
# Results AI Skill Lead Area · Results Block · Decision Map · Oct 23, 2026 (v1.0) --- ## 1. What the Skill Does The Results skill helps teams find out whether a bet paid off. It is the last block in the Lead area of Glare's Decision Map, and the last block of the whole map. Results judge **viability**: do people keep using it, coming back, and paying? They own **Adopt** signals: what people actually do over time, such as retention, repeat use, spend, and adoption. Test results are leading reads that belong to Findings and Bets. Results reads the lagging measures that only exist once the change is live. Results closes the loop: Initiatives and Bets come from Focus, Findings come from Measure, and Outcomes are what Results checks. The skill works in six steps. | Step | What it does | |---|---| | 1. Start from the bet's prediction | The behavior, metric, and window the bet named. No prediction, nothing to check | | 2. Set the baseline | The same metric, same audience, comparable period before the change | | 3. Read Adopt behavior over the window | Did people come back, repeat, stay, or pay? Launch day is not the outcome | | 4. Read up the chain by link type | Use Mapping's link types to decide how strongly to claim the business moved | | 5. Give a verdict | Held, Partly held, Didn't hold, or Too early | | 6. Re-rank the next round | The verdict changes what the business funds next | | Verdict | When to use it | What it does to the priority | |---|---|---| | Held | Adopt behavior and the business KPI moved as predicted, through a link strong enough to claim | Keep funding; scale the bet | | Partly held | Behavior moved, but the chain broke before the business KPI | Keep the priority; fix the broken link | | Didn't hold | The predicted behavior didn't appear | Move the priority down; send the gap back as a hunch | | Too early | The window hasn't closed, or the data is missing | No verdict yet; set the next check date | ### The Window Rule No verdict before the window closes. A lift that fades by day 60 is a different result from one that holds. Strong test results don't change that: until the window closes, the verdict is Too early. ## 2. Business Benefit Without this check, teams report launch-day numbers, call them proof, and move on. Results tells the business which bets to keep funding, which to fix, and which to drop. This helps teams: - find out whether a bet held over time, not just at launch - make claims that match the strength of the evidence - move funding toward priorities that are working - send what didn't hold back to Measure as a new hunch ## 3. Skill Output When used correctly, the skill produces an outcome check. The example below shows how this works for a mobile banking dashboard, 60 days after the bet shipped. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Bet's Prediction | Sessions ending without an action drop from 38% to under 30% within 60 days of launch | | Baseline | 38%, habitual users, the 60 days before launch | | Window | 60 days after launch (closed) | | Adopt Behavior | Sessions ending without an action: 29%. Weekly return visits held steady | | Up the Chain | Support calls about balance down 22% (Sequential). Churn flat so far (Correlated link; too soon to show) | | Verdict | Partly held: behavior moved as predicted, but the chain hasn't reached churn yet | | Re-rank | Keep the Retention priority. Fix the weak link: stage the next home screen change to half of users so the churn link can be read as Causal | | Next Check | Churn at 120 days | | Failure Mode to Watch | Calling it Held because the 79% test result looked strong. That was a leading read; this is the outcome. | | Next Step Handoff | → glare-focus-decisions (Bets) to re-rank the remaining bets | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Check Entry: Did the bet hold? *"Our mobile banking bet predicted sessions ending without an action would drop from 38% to under 30% in 60 days. It's day 60 and they're at 29%. Using the glare-lead-results skill, read up the chain, give the verdict, and tell us what to re-rank."* **Why this works:** It reads the outcome against the bet's own prediction, baseline, and window, not against how the test looked. **Best for:** - post-launch reviews - quarterly business reviews ### Prompt 2: Early Entry: Resist an early call *"Two weeks after launch, leadership wants to know if the new mobile banking home screen worked. Using glare-lead-results, tell us what we can say now, what we can't, and when the real verdict comes."* **Why this works:** It keeps launch-day excitement from being reported as the outcome. **Best for:** - pressure for early wins - launch updates ### Prompt 3: Miss Entry: When outcomes keep missing *"Our last three mobile banking bets were Partly held or Didn't hold. Using glare-lead-results, look for a pattern and tell us whether the problem is the designs or how the team works."* **Why this works:** Repeated misses usually point to how the team works, which the Design Assessment can rate. **Best for:** - teams with a string of misses - planning a process change --- *Glare Framework · glare-lead-results · Lead Area. Handoffs: glare-focus-decisions (Bets) · glare-lead-mapping · glare-lead-workflows · glare-measure-hunches (send a gap back) · glare-design-assessment*
# Workflows AI Skill Lead Area · Workflows Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Workflows skill helps teams connect design signals to the business functions that own the numbers. It is the second block in the Lead area of Glare's Decision Map, after Business Goals and before Mapping. Every company already runs on workflows: finding customers, keeping them, launching products, managing costs, reducing risk. The skill speaks each function's language across eight functions. | Function | Top metrics | Where design creates lift | |---|---|---| | Sales | Pipeline growth, conversion, quota attainment | Clearing friction from demos and trials | | Marketing | Lead quality, campaign ROI, acquisition cost | Clear messages and pages before the spend | | Product | Feature adoption, retention, product–market fit | Early proof of adoption and retention | | Engineering | Velocity, rework cost, defect rates | Catching usability failures before handoff | | Strategy | Market share, innovation rate, differentiation | Proving desirability before big bets | | Operations | Efficiency, support costs, internal adoption | Smoother flows and internal tools | | Finance | Revenue growth, margin, ROI | Making design ROI visible | | Legal | Compliance, liability cost, risk exposure | Clearer consent and accessible flows | In Focus, every bet names the workflow that owns its result, so that function knows the change is coming and which of its metrics should move. ### The Owner Rule Every KPI in the chain needs a function that owns it and acts on the verdict. A signal no function owns stays in a deck. The skill also runs the Quick Test: did users succeed without friction, did adoption or retention improve, and did it move a metric this function already tracks? ## 2. Business Benefit Design gets sidelined when it can't show its effect on the functions that run the business. Speaking their language puts design signals into decisions those teams are already making. This helps teams: - translate a finding into each function's own metrics - know who owns each KPI before the bet ships - bring functions into a test instead of reporting to them after - turn internal concerns into hunches Measure can test ## 3. Skill Output When used correctly, the skill produces a workflow map for a bet. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Bet | Back the redesigned home screen (from Focus → Bets) | | Owning Function | Product: owns sessions ending without an action and monthly active use | | Also Affected | Operations: owns support calls about balance. Marketing: loses promo space on the home screen | | In Their Language | Product: fewer abandoned sessions. Operations: fewer "where's my balance" calls. Marketing: promo clicks may drop; watch campaign ROI | | Internal Hunch to Test | Marketing believes moving promos lower will cut offer sign-ups (sent to Measure → Hunches) | | Failure Mode to Watch | Telling Marketing after launch. The function that loses something should hear about it before the bet ships. | | Next Step Handoff | → glare-lead-mapping to trace the chain from the user need to the business goal | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Owner Entry: Find who owns the result *"We're backing the redesigned mobile banking home screen. Using the glare-lead-workflows skill, name the function that owns each KPI in the chain, and which other functions the change affects."* **Why this works:** It puts an owner on every number before the bet ships. **Best for:** - preparing a bet for launch - cross-functional planning ### Prompt 2: Translation Entry: Speak each function's language *"Our finding: first-click success on balance rose from 61% to 79%. Using glare-lead-workflows, translate it for Product, Operations, and Finance using each function's top metrics."* **Why this works:** The same finding means different things to different teams. Translating it gets it into their decisions. **Best for:** - stakeholder updates - leadership reviews ### Prompt 3: Concern Entry: Turn pushback into a hunch *"Marketing says the new mobile banking home screen will hurt offer sign-ups. Using glare-lead-workflows, turn this into an internal hunch with the KPI that would prove it, and send it to Measure."* **Why this works:** Pushback becomes something the team can test instead of argue about. **Best for:** - functions that resist a change - launch risk reviews --- *Glare Framework · glare-lead-workflows · Lead Area. Handoffs: glare-lead-business-goals · glare-lead-mapping · glare-lead-results · glare-measure-hunches (internal hunches) · glare-design-assessment*
# Concepts AI Skill Measure Area · Concepts Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Concepts skill helps teams name what is being designed and why, before anyone writes a hunch. It is the first block in the Measure area of Glare's Decision Map, followed by Hunches, Testing, and Findings. A concept is a focused design effort, such as a page, flow, feature, or message, tied to a user need and a business goal. Big initiatives cross many teams. The skill breaks them into concepts small enough to test one at a time. Every concept connects two things. | Part | Where it comes from | How it is measured | |---|---|---| | User Need | Define | UX metrics, such as Comprehension or Error Rate | | Business Goal | Lead | Business metrics, such as conversion or retention | The skill also captures loose requirements: boundaries, not specs. It uses the concept catalog, grouped by Design Area (Products, Websites, Mobile Apps, E-commerce, Marketing), as a starting point. ### The Five-Minute Rule If the team can't write the user need, the business goal, and the signal it would measure in under five minutes, the concept isn't clear enough to test. Narrow it until it is, and work on one concept at a time. ## 2. Business Benefit Teams that jump straight to solutions end up with polished features that solve the wrong problem. A concept says what winning looks like before the test, so every signal that follows has a meaning. This helps teams: - break big initiatives into pieces they can test - connect design work to a user need and a business goal - agree on what is being judged before testing - avoid vanity metrics by naming the signal up front ## 3. Skill Output When used correctly, the skill produces a concept brief. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Concept | Redesign the dashboard home screen so balance and recent transactions are the first thing users see | | Design Area | Mobile Apps | | User Need (Define) | Findable: habitual users need their balance without extra taps, measured by Success and Comprehension | | Business Goal (Lead) | Reduce session abandonment, measured by sessions that end without an action | | Loose Requirements | Keep quick actions on the home screen; no new navigation patterns this release | | Signal to Measure | First-click success on balance | | Failure Mode to Watch | Writing the solution as the concept ("add a balance card") before naming the need it serves | | Next Step Handoff | → glare-measure-hunches to state the belief about this concept and the questions that would prove it | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Breakdown Entry: Split a big initiative *"Our initiative is to make the mobile banking dashboard easier to navigate. Using the glare-measure-concepts skill, break it into three or four concepts we could test one at a time, each with a user need, a business goal, and the signal we would measure."* **Why this works:** It turns a broad goal into concepts the team can test, without picking a solution too early. **Best for:** - kicking off Measure after a Focus initiative - backlogs full of large, vague efforts ### Prompt 2: Clarity Entry: Check a concept *"Our concept is 'add a balance card to the mobile banking home screen.' Using glare-measure-concepts, tell us whether this is a concept or a solution, and rewrite it with the user need, business goal, loose requirements, and signal."* **Why this works:** It catches a concept written as a fix and reframes it around the need, so the hunches that follow can test more than one answer. **Best for:** - concepts that arrive as feature requests - reviewing a concept before writing hunches ### Prompt 3: Catalog Entry: Find a starting point *"We are designing the mobile banking dashboard home screen. Using glare-measure-concepts, find the closest concepts in the Mobile Apps area of the catalog and show us how they define intent and what they usually measure."* **Why this works:** The catalog gives the team a proven starting frame instead of a blank page. **Best for:** - teams new to Glare - common design areas like dashboards, onboarding, or checkout --- *Glare Framework · glare-measure-concepts · Measure Area. Handoffs: glare-measure-hunches · glare-define-user-needs · glare-lead-business-goals · glare-focus-initiatives*
# Findings AI Skill Measure Area · Findings Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Findings skill helps teams call the hunch, then turn the results into a design signal and a next move. It is the last block in the Measure area of Glare's Decision Map. Findings judge **usefulness**: did the design help people do what they came to do? They own **Use** signals, which show whether people can understand the design, find their way, and finish the task. If a test also turns up a Need or Prefer signal, the skill records it and passes it on: Need to Define → Situations, Prefer to Focus → Bets. Every finding opens with a verdict on the hunch. | Verdict | When to use it | |---|---| | Supported | The metric moved the way the hunch predicted, on evidence strong enough for the decision | | Challenged | Results were mixed, or pointed somewhere else | | Refuted | The results showed the hunch was wrong | | No call | The test couldn't answer the hunch; go back to Testing | Then it writes the design signal from four parts: the **behavior** observed, the **metric** that captured it, the **context** that explains it, and the **direction** it points. A metric alone is not a signal. ### The Verdict-First Rule Call the hunch before writing anything else, and weigh the call by the evidence rating in the Test Plan. A Use call needs evidence of use, such as clicks, task success, or errors. A call made only on the leading test is provisional until the lagging check comes in after launch. A single test ends in one of five decision types: | Decision Type | Use it when | |---|---| | Implement | Supported, on evidence strong enough for the size of the change | | Refine Design | The direction is right, but this version isn't strong enough yet | | Test Iteration | The evidence is useful but can't settle the call | | Revisit Later | The idea has potential, but the timing or context is wrong | | Do Not Pursue | Refuted, or not worth more investment | When the team wants to compare options, versions, or segments instead, the finding goes to Focus → Variables. A single test never goes straight to Bets. ## 2. Business Benefit Data is not a signal. Without findings, teams drown in dashboards and argue about what the numbers mean. With them, every result ties back to the hunch it tested and the need and goal behind it. This helps teams: - know what each test actually settled - see who a result holds for, not just the average - connect a result to a user need and a business goal - leave every test with a clear next move ## 3. Skill Output When used correctly, the skill produces a finding. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Hunch | Moving balance above the promo cards will raise first-click success on balance for habitual users | | Verdict | Supported (provisional, leading test only): 79% first-click success on balance; Moderate evidence | | Who It Holds For | Habitual users overall; users who rely on shortcuts took longer to find transfers | | User Need | Basics: Findable | | Business Result | Fewer sessions ending without an action, toward reducing session abandonment | | Design Signal | Behavior: habitual users go straight to their balance. Metric: 79% first-click success. Context: balance now sits above promotions. Direction: keep balance first and find a home for shortcuts. Type: Use | | Next Move | The team wants to compare against the current home screen before committing → Focus → Variables | | Lagging Check | Sessions ending without an action, 60 days after launch | | Failure Mode to Watch | Presenting the 79% as proof the live product improved. It is a leading result until the lagging check comes in. | | Next Step Handoff | → glare-focus-methods (Variables) to set up the comparison, or Implement if no comparison is needed | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Verdict Entry: Call the hunch *"Our Helio first-click test on the redesigned mobile banking home screen came back at 79% success on balance, with 100 habitual users. Using the glare-measure-findings skill, call the hunch, say how sure we can be, and write the design signal."* **Why this works:** It opens with the verdict and the evidence strength, so nobody overstates what one test proved. **Best for:** - reading a new test result - writing up a round for the team ### Prompt 2: Split Entry: Find who it holds for *"The redesigned mobile banking home screen tested well overall, but some users took longer to find transfers. Using glare-measure-findings, split the result by audience, say who the design is useful for, and name the next move."* **Why this works:** Averages hide splits. Naming who a result holds for often matters more than the overall number. **Best for:** - mixed or surprising results - products with very different user groups ### Prompt 3: Next-Move Entry: Choose what happens next *"Our hunch about the mobile banking balance was supported, but the team isn't sure whether to build it, refine it, or compare it against the current screen first. Using glare-measure-findings, pick one of the five decision types or route us to Focus, and explain why."* **Why this works:** Every test ends with a named next move, so the result doesn't sit in a deck. **Best for:** - teams stuck after a good result - deciding between building and comparing --- *Glare Framework · glare-measure-findings · Measure Area. Handoffs: glare-measure-questioning (Testing) · glare-measure-hunches · glare-focus-methods (Variables) · glare-define-situations (Need signals) · glare-lead-results (lagging check) · glare-design-signals*
# Hunches AI Skill Measure Area · Hunches Block · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Hunches skill helps teams turn a belief about a concept into something they can test, and write the questions that would prove it. It is the second block in the Measure area of Glare's Decision Map, after Concepts and before Testing. A hunch is an informed guess, written as a hypothesis: **We believe that** [this change] **for** [this group] **will** [have this impact] **because** [supporting reason]. The reason after "because" should point at something the team has seen. A hunch and its questions belong together: a hunch is only testable once someone writes down what they would need to see. Every hunch attaches to a concept. Hunches about people, such as what users need or the situation they are in, belong in Define. The skill gives each hunch a type and a perspective. | Hunch Type | What it is about | UX metrics it usually tests | |---|---|---| | Business | Whether the experience serves its goal for the right audience | Intent, Desirability, Brand Score, Completion Rate | | Presentation | Whether people understand what they see | Comprehension, Expectations, Appeal, Sentiment | | Interaction | Whether people can use it, and whether it works | Usability, Success, Effort, Error Rate, Time on Task | | Perspective | Who answers | Measured with | |---|---|---| | External | Users, in a test | UX metrics | | Internal | The team and the business function that owns the number | Product KPIs and Business KPIs | ### The Falsifiable Rule A hunch must be able to be wrong. Each one cites its observations, lists what it assumes, and names the metric that would prove it. If a hunch fits two types, split it into two. Untested assumptions carry the risk, so the questions test those first. The skill writes the questions too. Each question has a type (People, Process, Product, or Problem), a mode (Exploratory, Evaluative, or Comparative), a bias check, and a UX metric. Ready-made instruments such as SUS or SEQ are not written here; Testing picks them and runs them as written. ## 2. Business Benefit Teams get stuck waiting for perfect data or leaping to solutions. Hunches give them a place to start that can be proven wrong cheaply, before anyone builds. This helps teams: - focus on the beliefs that carry the most risk - turn expert audit issues into testable hypotheses - write questions that aren't leading or loaded - test internal concerns, like launch cost or support load, with the business's own numbers - keep the reasoning next to each hunch, so nobody rebuilds it later ## 3. Skill Output When used correctly, the skill produces a hunch ready to test, with a status: Draft, Ready to test, Supported, Challenged, or Refuted. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Concept | Dashboard home screen redesign | | Observations | Analytics: 38% of home screen sessions end without an action. Session recordings: users scroll past two promo cards before reaching their balance. | | Hunch | We believe moving balance and recent transactions above the promo cards for habitual users will raise first-click success on balance because users currently scroll past promos to find it. | | Type and Perspective | Interaction · External | | Assumptions | Users leave because the balance is hard to find, not because the numbers are wrong (untested) | | Questions | Product, Evaluative: "Where would you tap to check your balance?" (Success). Problem, Exploratory: "What, if anything, slowed you down?" (Effort) | | Metric | Success (leading, in Helio); sessions ending without an action (lagging, after launch) | | Status | Ready to test | | Failure Mode to Watch | "Users like a cleaner home screen": a preference, not a falsifiable hunch tied to a metric | | Next Step Handoff | → glare-measure-questioning (Testing) to pick the technique, Stack, and tool | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Observation Entry: Form a hunch from what you've seen *"Analytics show 38% of sessions on our mobile banking home screen end without an action, and recordings show users scrolling past promo cards. Using the glare-measure-hunches skill, write one or two hunches with their type, perspective, assumptions, and metric."* **Why this works:** Starting from observations keeps the "because" grounded in evidence instead of opinion. **Best for:** - turning analytics or recordings into testable beliefs - early rounds on a new concept ### Prompt 2: Audit Entry: Turn audit findings into hunches *"Our expert audit of the mobile banking dashboard flagged that the balance is buried below promotions. Using glare-measure-hunches, rewrite this audit finding as a hypothesis, add the impact metric, and write unbiased questions to test it."* **Why this works:** Most audit findings stop at the fix. Adding the impact metric is what makes them testable. **Best for:** - expert audits and heuristic reviews - backlogs of design opinions ### Prompt 3: Internal Entry: Test a business concern *"Our support team worries the new mobile banking home screen will raise calls about missing shortcuts. Using glare-measure-hunches, write this as an internal hunch, name the function that holds it, and tie its questions to a Product or Business KPI."* **Why this works:** Internal concerns become hunches the team can check with its own numbers, instead of objections that stall the work. **Best for:** - stakeholder concerns about cost or risk - changes that touch support, marketing, or sales --- *Glare Framework · glare-measure-hunches · Measure Area. Handoffs: glare-measure-concepts · glare-measure-questioning (Testing) · glare-measure-findings · glare-define-user-needs · glare-define-situations*
# Testing AI Skill Measure Area · Testing Block (formerly Questioning) · Decision Map · Oct 23, 2026 (v1.1) --- ## 1. What the Skill Does The Testing skill helps teams put a hunch against real evidence. It is the third block in the Measure area of Glare's Decision Map, between Hunches and Findings. Hunches set it up. Findings read what it returns. The skill still runs as glare-measure-questioning so existing installs keep working. This block used to be called Questioning. Writing questions now happens in Hunches; Testing picks how to check them. A test is any deliberate check of a hunch against evidence the team didn't write itself: a study, a standard instrument, an expert review, live behavior, or reading data the team already has. Every test has three parts. | Part | What it holds | Examples | |---|---|---| | Techniques | How you test | First click, task success, prototype test, expectation probe, data analysis, Wizard of Oz | | Stacks | What you ask, ready-made | Research Stacks (SUS, SEQ, UX-Lite, NPS) and Design Stacks (Website, Mobile App, Product, E-commerce, Marketing) | | Tools | Where you run it | Helio, Maze, Optimal Workshop, UserTesting, Lookback | Before a test runs, the skill rates how strong its evidence will be. | Rating | What the test captures | Examples | |---|---|---| | Weaker | Stated opinion or intent | Preference questions, likelihood to use, NPS | | Moderate | Behavior in a simulated setting | First click, task success on a prototype, tree testing | | Stronger | Behavior or commitment in the real world | Feature stub clicks, presales, Wizard of Oz | ### The Hunch Rule No hunch, no test. Open-ended learning before anyone has a hunch is Collecting, in Define. An interview with no hunch behind it is Collecting; an interview that probes a specific hunch is Testing. Running the same test across versions or segments is Comparing, in Focus. The skill also names a leading and a lagging metric. Behavioral and Attitudinal metrics lead: they can be read in a test before launch. Performance and Intelligence metrics lag: they are read in the live product after launch. Naming both up front makes sure the lagging check still happens once the leading test looks good. ## 2. Business Benefit Every hunch gets a test that fits it, and every result carries its source and strength. That keeps the team from treating a survey answer like proof of behavior. This helps teams: - pick the fastest test that can settle the hunch - say up front how much the result can be trusted - use standard instruments without breaking their scores - plan the after-launch check before the first test runs - set up a test that Focus can repeat later ## 3. Skill Output When used correctly, the skill produces a Test Plan. The example below shows how this works for a mobile banking dashboard. | Field | Example Output (Mobile Banking Dashboard) | |---|---| | Hunch | Moving balance above the promo cards will raise first-click success on balance for habitual users | | Technique | First-click test on the redesigned home screen | | Stack | SEQ (Single Ease Question), asked as written after the task | | Tool | Helio, 100 habitual mobile banking users | | Evidence Rating | Moderate: behavior in a simulated setting | | Leading Metric | Success (first click on balance), read in Helio before launch | | Lagging Metric | Sessions ending without an action, read in analytics 60 days after launch | | Failure Mode to Watch | Rewording SEQ to fit the brand voice. Changing a standard instrument breaks its score. | | Next Step Handoff | → glare-measure-findings to call the hunch when results come back | ## 4. Prompt Strategies The prompts below show different ways to use this skill. Each example uses a mobile banking dashboard update. ### Prompt 1: Match Entry: Pick the right test *"Our hunch: moving the balance above promo cards will raise first-click success for habitual mobile banking users. Using the glare-measure-questioning skill, pick a technique, a Stack if one fits, and a tool, rate the evidence, and name the leading and lagging metrics."* **Why this works:** It matches the test to the hunch and its metric, and says how strong the answer will be before any money is spent. **Best for:** - turning a ready hunch into a Test Plan - teams that default to the same test every time ### Prompt 2: Strength Entry: Upgrade weak evidence *"We were going to ask users which mobile banking home screen they prefer. Using glare-measure-questioning, tell us how strong that evidence is for our hunch about finding the balance, and suggest a test that captures behavior instead."* **Why this works:** Stated preference is weak evidence for a Use call. The skill moves the team toward a test of what people do. **Best for:** - plans built on surveys alone - large decisions resting on stated intent ### Prompt 3: Data Entry: Test with what you already have *"We have six months of mobile banking analytics and support tickets. Using glare-measure-questioning, show us how to read them against our hunch about the buried balance, what that test can and can't prove, and when we'd still need a study."* **Why this works:** Reading existing data against a specific hunch is a real test, and often the fastest one. **Best for:** - tight timelines - hunches about live products --- *Glare Framework · glare-measure-questioning (Testing) · Measure Area. Handoffs: glare-measure-hunches · glare-measure-findings · glare-define-collecting (no hunch yet) · glare-focus-comparing (repeat the test) · glare-ux-metrics*