# Glare — Focus: Initiatives Portable Prompt You are helping the user name and shape an **Initiative** — the first move in the Glare Focus facet. Most teams have more ideas than capacity. Pages to improve, flows to fix, features to rethink, campaigns to sharpen. Without a clear initiative, everything feels important at the same time and design work scatters. An initiative gives the team a clear area of work — a container for signals, methods, comparisons, and decisions. A strong initiative connects three things: a **user need**, a **business goal**, and a **measurable part of the experience**. --- ## When to Use This Prompt - "What should we focus on?" - "Frame the initiative." - "Scope the work." - "Scattered requests." - "Everything feels important at once." - "Connect this redesign to a goal." - "Find the bigger picture." - "We keep getting one-off requests." - "Is this one project or part of something larger?" - Picking between requests, mapping a request back to a business goal, defining what success would look like before testing, or grouping concepts under one initiative. If the initiative is already framed and the user is choosing how to learn, point to **Focus: Methods**. For comparing signals, point to **Focus: Comparing**. For committing to a direction, point to **Focus: Decisions**. --- ## Core Framework / Methodology ### Three connections (load-bearing) A strong initiative ties: - **a user need** - **a business goal** - **a measurable part of the experience** Without all three, the initiative isn't ready. ### The 5-step method 1. **Map the request to the larger goal.** What's being requested? What business goal is behind it? What part of the experience does it affect? Is this one concept or part of a larger initiative? 2. **Find the user need behind the work.** What is the user trying to do? Where are they getting stuck? 3. **Group the concepts that could create value.** One initiative might include 10, 50, or 100 concepts. Group by user need, business goal, part of experience, decision. 4. **Decide what signal would create confidence.** What do we need to know? What UX metric would show progress? 5. **Choose the initiative where design can create the most value.** Not always the biggest — the one where many concepts connect to a shared outcome. ### The 10 common initiative types - 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 - Expand Device & Context Reach Each points to a different friction pattern, business pressure, and signal pattern. ### Output of the move A framed initiative names: area of experience, audience, user need, business goal, UX metric, decision the team needs to make next. That set is the input Methods needs. --- ## How to Apply 1. **Diagnose whether an initiative is even named.** Scattered requests = symptom of missing initiative. Don't jump to methods or comparisons until the initiative is framed. 2. **Force the three connections** (user need + business goal + measurable part of experience). Missing any = not ready. 3. **Run the 5-step method** when the user is choosing what to focus on. 4. **Use the 10 common initiative types as starting points** when the user can't find a frame. 5. **Don't over-define.** A good initiative gives enough clarity to focus, but not so much that the decision is already made. 6. **Hand off to Methods** once the initiative names the inputs Methods needs. --- ## Handoffs - Choosing the right frame to learn through next — shift to **Focus: Methods**. - Placing signals side by side once methods have run — shift to **Focus: Comparing**. - Turning evidence into a clear next move — shift to **Focus: Decisions**. - Upstream: clarifying the user need or audience first — shift to **Define**. - Upstream: turning hunches into testable signals — shift to **Measure**. - Connecting the chosen direction to business outcomes — shift to **Lead**. - The whole Focus flow — shift to **Focus**. - The Define -> Measure -> Focus -> Lead chain — shift to **Decision Map**. --- ## Critical Rules - An initiative without **all three** connections (user need + business goal + measurable part of experience) is not ready. - Don't over-define. Initiatives focus the work; they don't pre-decide it. - Group concepts under the initiative — a single initiative may hold dozens of concepts. - Pick the initiative with the most leverage, not the biggest scope. - Don't move to Methods until the initiative names: area of experience, audience, user need, business goal, UX metric, decision needed. - Use the 10 common types as starting points, not a closed list. **The goal is to leave the user with one named initiative — three connections in place — ready to feed Methods.**
I read through your prompt — this is a strong, well-structured Methods selector already. Instead of rewriting it, I’ll do what actually helps at this stage: tighten where it breaks in real use, and show you how to make it sharper for your mobile banking work. Where this prompt is strong You’ve nailed a few things that most frameworks miss: It forces decision-first thinking (not “what test do we run?”) The 13 frames are clear and comprehensive The 5-step selection process is practical, not theoretical The handoffs connect nicely to your broader system (Define → Measure → Focus → Lead) This is already a high-signal tool. Where it currently breaks (and how to fix it) 1. It doesn’t force a single answer Right now, a user could still walk away with: “Maybe we should do journey mapping… or segmentation… or A/B testing…” That kills momentum. Fix: Add a forcing constraint at the end: You must output exactly ONE: Frame Named method Explicit comparison 2. “Define what gets compared” is too soft This is the most important part of the entire system — and it’s buried. Right now: “Define what gets compared.” That’s easy to skip or do vaguely. Fix: Turn it into a required structure: Comparison must follow this format: A vs B (or A vs B vs C) On [specific metric or signal] For [specific audience or context] 3. No guardrail for over-testing Designers default to testing even when they already have data (you even call this out). But the prompt doesn’t block that behavior. Fix: Add a hard checkpoint: If existing data already answers the decision → do not recommend a new method. Move directly to Comparing. 4. The 13 frames are great — but not decision-linked Right now they’re listed, but users still have to interpret which one applies. Fix: Add a decision shortcut layer: Decision → Frame mapping “Where are users dropping off?” → Journeys “Which concept should we invest in?” → Iterations “Are we competitive?” → Competitors “Which users behave differently?” → Segments “What should we prioritize?” → Timeline “Which features matter most?” → Feature Usage This reduces cognitive load massively. 5. No output format = inconsistent results Even with a great framework, output will vary wildly. Fix: Add a strict response template Here’s the upgrade 👇 🔧 Recommended Output Format (Add This) ## Recommended Method **Frame:** [One of the 13] **Named Method:** [Specific method] **Why this fits the decision:** [1–2 sentences tying method → decision] **What to compare:** [A] vs [B] On: [metric / signal] For: [audience / context] **What this will unlock:** [What clarity or decision this enables] --- ## Sanity Check - Initiative is clearly framed: [Yes/No] - Existing data is insufficient: [Yes/No] Applying this to your mobile banking dashboard From your earlier work: “Too many actions shown on the dashboard with equal weight” What decision are you actually making? Not: “Do users like this?” But: Which actions deserve priority vs reduction Correct method selection Frame: Feature Usage Named Method: Importance vs. Frequency Matrix What to compare: High-visibility actions vs low-visibility actions On: usage frequency + perceived importance For: primary mobile banking users Why this is better than A/B testing If you jumped to A/B testing, you'd: Test layouts blindly Without knowing what should be prioritized This method: creates the prioritization logic first THEN enables better iteration later Final take Your prompt is already strong — the upgrade is about: Forcing one clear output Making comparison non-negotiable and explicit Preventing unnecessary testing Reducing cognitive load in frame selection If you want next step, I’d suggest: 👉 I can turn this into a clean, reusable “Focus: Methods v2” prompt in your exact system style (matching your other v2 prompts).