Frame
Define the user task, constraints, success criteria, and failure states. Use interviews and product evidence to sharpen the question.
Juan Hernandez · AI-native UI/UX Designer
Complex products.
Clear experiences.
I bring AI-native UI/UX, systems thinking, and a decade of growth experience to B2B SaaS. From a messy product question to an interactive experience people can use.
28 → 45%SaaS activation
57CRO & activation experiments
9 → 3.5 daysMedian idea to launch
Selected career outcomes.
Concepts below use sample data.
01 / Interactive conceptSignal. Make the next
Signal. Make the next
customer action obvious.
A customer activation workspace for SaaS teams. Inspect account health, identify friction, and try a guided onboarding flow.
Try it: filter At risk, open an account, then try its sample setup. This independent concept uses fictional data.
signalActivation workspace
Concept · Sample dataActivation overview
Find the accounts that need a next step.
Preview data state
Accounts tracked5Across 3 plans
Activated20%1 of 5 accounts
Needs attention2Below 40% completion
| Account | Status | Completion | Next action |
|---|---|---|---|
| Enterprise | At risk | Connect a data source | |
| Growth | In progress | Invite a teammate | |
| Enterprise | Activated | Activation complete | |
| Growth | In progress | Publish a dashboard | |
| Starter | At risk | Connect a data source |
Scope Independent B2B SaaS concept, 2026
Build React, Tailwind, shared tokens; Codex-assisted
Validation Browser and technical QA; user study pending
The product decision
Show the next useful action beside account health. A score tells a team where to look; an action helps move the account forward.
The design tradeoff
Focus on activation rather than an all-purpose CRM. Account details appear inline so the table stays in context.
What I would validate
Ask customer-success users to identify an at-risk account and its next action. Measure task success, time to decision, and interpretation errors. These are proposed measures, not reported user-study results.
Design decisions and validation plan
Audience: customer-success teams deciding which new accounts need help. The main task is to find an at-risk account and identify its next action.
Decision: keep the score and next action together. Use progressive disclosure for account details, then carry that account into a reversible setup preview.
Alternatives: a larger CRM would obscure the activation task; an AI chatbot would add uncertainty and data handling without helping this five-account example.
Next study: observe customer-success practitioners finding an at-risk account, explaining its next action, and completing setup. Record assistance, errors, and time. Do not claim measured business impact until real evidence exists.
Limits: sample data, in-memory choices, no integration or live AI. The showcase demonstrates shared components but is not a hosted Storybook.
02 / Experience case studyLess friction.
Less friction.
More activation.
Independent / Venture Studio Portfolio
Fractional Head of Growth · 2024–present
Connecting acquisition to first value
A product-led SaaS funnel needed a coherent path from persona landing page to trial, onboarding, and lifecycle engagement. I built that funnel and ran CRO and activation experiments across landing pages and in-product onboarding.
+17 ppActivation, 28% → 45%
+22 ppOnboarding completion
I worked with founders, designers, copywriters, and developers in weekly sprints. Across the portfolio, median idea-to-launch time fell from nine days to 3.5.
Career results supplied in my resume. Signal is a new concept, not a reconstruction of a client interface; these outcomes are not attributed to Signal.AI accelerates the build.
Judgment shapes the product.
Prototype
Sketch in Figma. Bring the brief and token context into Claude Code or Cursor to build an interactive React/Tailwind experience.
Systematize
Reuse components and semantic tokens. Check focus, disabled, empty, and error states; document reusable variants in Storybook.
Validate
Walk through the experience with users and engineering. Refine interactions, record decisions, and choose the next change from evidence.
AboutDesign with a
Design with a
business instinct.
My background spans growth and product optimization at JumpCloud, Globalization Partners, Mercer, Essity, and independent ventures. I bring that experience to AI-native UI/UX: clear journeys, reusable systems, and measurable user outcomes.
Claude Code · Cursor · Figma · React · Tailwind CSS · Storybook · User interviews · Mixpanel
Mexico City · Native Spanish · C2 English
juanhdeza93@gmail.com