From Tech to Design Leadership: Decisions, AI & Building at Headout

Core theme: Design decisions as differentiation vs. AI

• Rama (Senior Design Director at Headout) frames the key “human” advantage as design choices shaped by experience, empathy, and intuition—areas AI can’t match “at par.”
• Headout’s design ethos includes “craft in all we do” and a belief that small, high-delight details can matter beyond pure conversion.

Design & product craft choices: “Ship / Fix / Kill” mindset

This decision framework guides what the team does with ideas and features:

• Ship it: If the change creates real delight/trust impact and the effort supports user experience quality—not just aesthetics.

• Fix it: If the idea is good but the UX is mechanically “bad” (e.g., overwhelms users or relies on deceptive/low-trust patterns).

• Kill it: Not explicitly chosen in the excerpts, but implied as removing features that neither users benefit from nor align with the product’s long-term trust/culture.

Example: Onboarding UX with many steps

• Scenario: “Beautiful onboarding experience with seven, eight steps.”
• Decision: Fix, unless those steps are genuinely necessary and users don’t feel burden/pressure.
• Heuristic mentioned: Progressive disclosure / avoid showing too many options at once.
• Actionable example: A Typeform-like approach—showing only one question at a time, even when the overall questionnaire is long.

Example: “Dead space” features executives love

• Scenario: “Feature nobody uses, executives absolutely love.”
• Decision: Ship it if it creates delight, uniqueness, and trust signals, even if it doesn’t directly drive CTR/CVR.

Concrete examples

• Foursquare badges: Used more for “fun” in some markets (e.g., India), but built as a craft-driven delight system (e.g., badges tied to museum/multiplier concepts).
• Hidden door / cobweb interaction: A playful interaction that reveals old badges via a “wipe to clean” gesture—turning legacy content into something engaging.

Internal principle

• Headout’s “craft in all we do” applies even to documentation, experiments, and throwaway prototypes.

Example: Dark pattern that increases conversion

• Scenario: “Dark pattern that increases conversion significantly.”
• Decision: Fix, not kill—conversion gains may be real, but the team avoids shifting into unethical UX.

Headout context: “tickets and tours”

• Tickets: entry tickets; generally less educational.
• Tours: guided experiences with stories/immersion (example referenced: Hagia Sophia cues becoming learnable only through a guided tour).

Rejected plan (quality/values mismatch)

• AI-generated guide names
• AI-like review snippets
• AI-subtext claiming “years of experience”

These didn’t “stick well” with the team’s intent/standards.

Ethical fix approach (using real user data)

• Build a small pipeline:

  1. Pull real reviews for a guided tour.
  2. Run them through language models to detect actual guide name patterns (prefix/suffix around the word “guide”).
  3. Filter assumptions:
     • Names likely appear near the word “guide”
     • Names should be a noun (not a verb)
  4. Prioritize authenticity:
     • Use real review quotes/reviews rather than generating AI-written reviews.

• Validate intent/outcome:

  • Determine whether conversion is driven by the dark pattern alone or by a combination of smaller effects.
  • Fix and still ship, while removing the unethical behavior.

Operations / team process

“Full pulse” daily ritual/process

• Each day, the team takes new data from Headout systems and showcases it to the entire organization to maintain alignment and continuous learning.

Leadership & career-driven leadership practices (indirect but business-relevant)

Rama’s leadership at Zeta (prior role) emphasized repeatable design methods:

• Activity-oriented design methods (AODM):

  • Teach one design method per month to individuals/teams.
  • Learners complete it end-to-end, then run a session for the full team.
  • Run an activity to test applicability.

• Publish learnings on a blog to reinforce and spread “design culture.”

Rationale

• Create repeatable processes so design methods scale beyond a subset of people with a “design background.”

Company growth / scale metrics mentioned

• Headout international presence: approximately 41 countries
  • A goal referenced as 50 was later reduced to 41 due to execution realities.
• OKRs concept: everyone uses OKRs with a “North Star” goal, but targets can adjust as execution realities emerge
  • No specific numeric business KPIs (e.g., revenue/CAC/LTV/churn) were provided in the excerpt.

Presenters / sources

• Rama — Senior Design Director at Headout (primary speaker)