Diamond 1 — Choosing Where to Look

How the Halo Alert team picked a community to search, and proved they could reach it

Five days, three stages, one decision. The team widened to fourteen candidate communities, narrowed to one, and then spent forty-eight hours finding out whether they could actually reach it before committing a month to the attempt.

See Halo Alert at a Glance for where this sits in the whole expedition, and the demonstration overview for how much of this record is reconstructed.

Diverge — Casting a Wide Net

19 February. Guide: Cast a Wide Net.

Anchor

We’re exploring friction around personal safety and vigilance during the transitional segments of daily travel for women moving between transit and destination in urban and campus contexts.

Shortlist Table (Top 14)

#	Group label	One-line profile	Orbit roles	Likely access channels	Access	Why this group might hurt
1	Professional women walking part of commute	Subway→office walkers avoid side streets after dark	Roommates, property manager, transit security	Station exits (AM/PM), coworking email list, LinkedIn women-in-tech	MED	Lighting gaps, sparse foot traffic, exit anxiety
2	Female grad students (off-campus)	Late campus→housing walks with variable schedules	Lab mates, dept admins, campus police	Grad Slack, dept newsletters, library exits	HIGH	Staggered hours, budget routes, thin lighting
3	Retail closers (evening shift)	Close store, cash-out, walk to bus/subway after 9–11pm	Shift leads, mall security	Employer break rooms, union/worker groups, bus stops	MED	Fixed late hours, predictable routes, fatigue
4	Night-shift nurses	3am lot crossing when staffing is sparse	Nursing supervisors, hospital security	Staff Slack, union board, hospital lots	LOW	Distance + fatigue + predictable patterns
5	Campus staff (custodial/night ops)	Solo building exits to parking or transit ~10pm–2am	Facilities mgr, campus police	Facilities briefings, timeclock areas	MED	Dim pathways, tool bags, routine routes
6	Rideshare drop-off walkers	Dropped at arterial, walk last 2–6 blocks home	Drivers, building doormen	Rideshare driver forums, building lobbies	LOW	Micro-routes, inconsistent lighting
7	International students new to city	Unfamiliar routes; language/apps not fully internalized	Intl office, peers, RA	Intl student office, WhatsApp groups	HIGH	Map anxiety, time-of-day mismatch, signals
8	Suburban park-and-ride walkers	Walk from remote lot/bus to workplace pre-dawn	Lot attendants, employer security	Park-and-ride kiosks, employer listserv	MED	Low density, long uncovered stretches
9	Hotel housekeepers (AM start)	Transit + early arrival before dawn	Housekeeping leads, HR	Staff entrances, HR bulletin	MED	Early hours, uniforms, predictable cadence
10	Restaurant workers (late)	Post-midnight cleanup → transit	Shift leads, door staff	Back-of-house boards, industry FB groups	MED	Cash tips, low foot traffic windows
11	Early-morning joggers (women)	Pre-sunrise routes; phone not always in hand	Run clubs, park stewards	Strava/Runclub posts, trailheads	LOW	No pockets, routine paths, sparse help
12	Hospital residents	Rotating late calls; garage/lot to housing	Chiefs, parking office	Residency emails, parking kiosks	MED	Irregular hours, exhaustion, badge access
13	Tourists near transit	Unfamiliar neighborhoods at dusk	Hotel concierge, tour ops	Hotel lobbies, concierge boards	LOW	Map reliance, visible cues, valuables
14	Childcare workers with split shifts	Transit + midday/night returns	Center directors, parents	Center break rooms/newsletters	MED	Unpredictable returns, carry supplies

Notes: Prioritized groups with (a) recurring late/eary segments, (b) observable routes, (c) plausible access channels, and (d) signals of vigilance load (cognitive tax from staying alert).

Orbit Snapshots (3 of 14)

A) Female Grad Students (off-campus)

• Orbit roles: Lab mates (walk buddies), department admins (listserv reach), campus police (patrol patterns)
• Influences/constraints: Late lab hours; grant deadlines; campus-to-city lighting gaps
• What this unlocks: Group safety norms (check-ins), preferred paths, “avoid zones,” willingness to share ETAs

B) Retail Closers (evening shift)

• Orbit roles: Shift leads (schedule owners), mall/strip security (escorts), transit operators (last bus timing)
• Influences/constraints: Fixed closing times; cash handling; uniform visibility; last run bus
• What this unlocks: Predictable windows for observation/intercept; back-of-house access via managers

C) Professional Women Walking Part of Commute

• Orbit roles: Roommates/partners (check-in routines), building management (lighting), transit security (presence cues)
• Influences/constraints: Seasonal darkness; micro-detours; elevator/door entry lag
• What this unlocks: Rich narratives about exit anxiety, lighting choices, text-to-friend routines

Access Channels to Try (Fastest → Slowest)

1. Grad Slack / Dept newsletters (HIGH) — lightweight message with study invite + 10-min intercept at library exits
   
2. Campus facilities briefings (MED) — ask to attend 5-min pre-shift, recruit 1-on-1s; post QR flyers at timeclocks
   
3. Retail break rooms (MED) — manager permission + short sign-up; pair with safe-walk offer or coffee cards
   
4. Employer listservs / HR bulletins (MED) — custodial, hotel housekeeping, hospital parking office
   
5. Station exits intercepts (MED) — clipboard + quick screener + opt-in for 20-min interview later
   
6. International office WhatsApp groups (HIGH) — multilingual invite + map anxiety prompt

Fallback (LOW): Rideshare driver forums; tourist hotel lobbies (permission friction)

Micro “Access Smoke Test” Scripts

• Digital (Grad Slack):
  “Hi all — quick study on feeling safe during the short walks between transit and home or lab at night. 10-minute chat at [library exit, 8–10pm], $10 coffee card. Not selling anything; anonymized notes only. DM ‘yes’ for details.”

• Physical (Retail back room):
  “We’re studying late-night walks from store to transit or parking. 3 questions, 5 minutes, off-the-clock, $10 coffee card. OK to say no. Who has closed this week and taken transit?”

• Intercept (Station exit):
  “Two-question study: When you exit here after dark, what do you change? (route/lighting/check-ins) Would you be open to a 20-minute follow-up later this week?”

Bias Check (Applied)

• Not just “people like us”: included night staff, caregivers, service workers, internationals, commuters outside downtown cores
  
• Mixed digital/physical channels to avoid only-convenience sampling
  
• Retained LOW access groups (joggers, tourists) as stretch targets for later

Next Moves into Converge

• Email/DM two HIGH channels this week (Grad Slack, Intl Office)
  
• Run two MED physical intercepts (station PM exit, retail back room with manager OK)
  
• Bring counts: invites sent, replies, yes to 20-min, actuals scheduled → decide focus group in Converge

Attachments (Data Room pointers)

• Shortlist CSV (20+ rows): Candidate communities, quick profiles, orbit roles, access notes.
• Bias Check Additions (3 non-obvious groups): Documented rationale for why they were added.
• Draft Access Notes (txt): Where/how the team might reach each group (at least one viable channel per group).
• Brainstorm Capture (optional): Screenshot/photo of Miro board, sticky-note clustering, or whiteboard output used to generate candidate groups.

Bias Check

To counteract convenience bias (only focusing on obvious commuting women in NYC), we deliberately added three non-obvious groups:

1. International students on urban campuses
   • Quick Profile: Often commute on foot late at night after study groups; culturally less likely to report safety fears.
     
   • Why Added: Visibility is lower; needs may differ due to cultural norms.
     
   • Access: Campus cultural clubs, international student office.
2. Elderly women commuting to part-time jobs
   • Quick Profile: Commute early/late on limited incomes; may use bus or subway.
     
   • Why Added: Overlooked due to age; different safety perceptions and physical vulnerabilities.
     
   • Access: Senior centers, church groups, job placement programs.
3. Immigrant women working late service shifts
   • Quick Profile: Women employed in restaurants, cleaning, or retail who often walk or use transit home after midnight. Language barriers and documentation status can compound vulnerability.
   • Why Added: Frequently invisible to mainstream “professional women” framing; experience late-night exposure and systemic neglect.
   • Access: Worker advocacy orgs, ESL classes, churches, and immigrant community associations.

Reflection:
These adds diversify our exploration, forcing us to see beyond the default category of “professional young women.” They highlight cultural, generational, and identity-driven differences that mainstream safety products rarely account for.

Draft Access Notes

• Female college students (night classes)
  Access via: campus org newsletters, student unions, safety escort programs.

• Professional women (late subway commute)
  Access via: LinkedIn groups, coworking spaces, women-in-tech Slack groups.

• Hospital shift workers (nurses/doctors)
  Access via: nurse associations, hospital HR postings, professional listservs.

• Gig economy workers (delivery, rideshare)
  Access via: app-based worker subreddits, Discord groups, drivers’ advocacy collectives.

• Elderly women (part-time jobs)
  Access via: senior centers, AARP groups, community bulletin boards.

• International students
  Access via: international student offices, ESL classes, WeChat groups.

Notes:
- These are draft channels, meant to test reachability in the Access Test.
- Actual reliability will need validating (response rates, willingness to engage).

Converge — Choosing a Community

24 February. Guide: Commit to a Community.

Context

This document demonstrates how the Halo Alert team moved from divergent exploration of possible communities to a converged choice of focus. It follows the Convergence Guide and illustrates each step with real decisions, reflections, and evidence from the team’s early work.

1. Compare: Spot Patterns and Weigh Possibilities

Candidate groups under review (from Diverge stage):

• College women walking home at night
  
• Professional women commuting on foot in mid-sized cities
  
• Female university students crossing campus after dark
  
• Shift workers returning home before sunrise
  
• Immigrant women using late-night buses
  
• Elderly women commuting by public transit
  
• Women in rural areas walking to jobs without cars

Patterns noticed:

• Recurring theme of vigilance: across nearly all groups, women adapted routines for safety (texting friends, carrying keys, rerouting).
  
• Time-of-day effect: concerns intensified at night or in dark/isolated areas.
  
• Transit + walking overlap: many groups blended public transit with walking — friction arose especially in “last mile” segments.
  
• Uneven visibility in mainstream solutions: tech apps and products largely targeted “generic commuters” or “college safety,” not nuanced subgroups.

Team reflection Professional women commuters and female students showed layered patterns: safety behaviors woven into daily life, often invisible to outsiders. These weren’t dramatic events, but steady frictions.

2. Empathize: Notice the Emotional Pull

Groups that lingered after work sessions:

• Professional women commuters — stories of texting roommates, rerouting streets, balancing career pressure with constant background vigilance.
  
• College students — raw vulnerability late at night, but often constrained by campus access rules (harder to study without IRB).

What stuck emotionally:
The team couldn’t shake the quiet burden of professional women who “looked fine” yet carried routines of self-protection every day. Unlike campus students, these women rarely voiced complaints — they had normalized fear.

Signal of empathy Team members reported “feeling protective” and a sense of injustice — why should these capable professionals still have to carry keys between their fingers?

3. Assess Access: Can You Actually Reach Them?

Quick tests conducted:

• Draft LinkedIn messages to small networks of women in New York and Boston.
  
• Pilot outreach in coworking Slack group: short invite for 15–20 min conversations.
  
• Intercept attempts outside subway exits near campus.

Results:

• LinkedIn: 3 out of 5 messages received thoughtful replies within 48 hours.
  
• Slack group: 8 women volunteered stories or referrals.
  
• Subway intercept: 2 agreed to short chats, though one declined recording.

Assessment:

• Observable in public settings.
  
• Reachable through professional and digital networks.
  
• Willing to engage, offering thoughtful accounts.

4. Select Your People

Decision:
Halo Alert chose to focus on professional women commuting on foot in mid-sized cities.

Rationale:

• Clear patterns of persistent, subtle safety burdens.
  
• Emotional resonance: the team felt drawn to their lived experience.
  
• Proven access: multiple channels worked quickly and respectfully.

Note: This was treated as a working commitment, not a final lock-in. The team acknowledged definitions would evolve as pain points clarified.

Attachments (Data Room Pointers)

• Comparison board (Miro capture) — clusters of candidate groups & overlap patterns.
  
• Empathy notes — debrief docs where team recorded “lingering voices.”
  
• Access tracker sheet — outreach attempts, response rates, notes on tone of replies.
  
• Draft outreach messages — short LinkedIn / Slack scripts.

Researcher Reflection

• Surprises: Volume and thoughtfulness of replies from professional women via Slack — faster and deeper than expected.
  
• Updated assumptions: Safety concerns aren’t always voiced; they’re embedded in daily rituals.
  
• Next step: Launch ethnographic exploration (conversations, observations) with this group to surface unmet needs.

Attribution: Convergence documented by the Halo Alert team, 2025-02-20.

Test — Confirming Access

24 February. Guide: Test Your Access.

1. Define Your Group

• Chosen people/community: Professional women commuting on foot as part of their daily routine in mid-sized U.S. cities
  
• Why this group matters: Their commute includes segments where safety feels uncertain — walking from transit stops, through crowded streets, or after dark. These women use quiet but persistent workarounds (text check-ins, detours, vigilance) that suggest unmet needs.
  
• Typical settings or gathering places: LinkedIn professional groups, coworking spaces, local women-in-tech Slack channels, evening transit stops, women’s safety forums.

2. Minimum Access Conditions

• We can find them (professional networks, transit hubs, coworking lists).
  
• We can engage them (via LinkedIn messages, coworking newsletters, Slack forums).
  
• They respond and open up (first tests yielded thoughtful, detailed replies).

Result: Passed.

3. Test Design

• Channels tested:
  • LinkedIn: short DMs to 10 women in relevant commuting geographies
    
  • Coworking space email list: 1 announcement inviting brief chats
    
  • Women-in-tech Slack group: posted an open-ended “commuting challenges” question
    
• Sample size goal: 10–15 responses across 3 channels
  
• Method: Direct outreach (DM/email) + forum post
  
• Timeline: 48 hours

4. Execution Notes

• LinkedIn: 10 messages sent → 5 replies; 3 agreed to interviews within a week.
  
• Coworking list: 1 group email → 6 replies; 4 substantive, 2 brief.
  
• Slack forum: 1 post → ~15 views, 4 replies, 2 sharing detailed personal stories.
  
• Total outreach: ~21 people touched; 15 replies; 9 strong leads for conversations.

Surprises:

• Faster, warmer response from coworking list than expected.
  
• Slack replies included safety experiences outside commuting (night walks, travel), broadening context.

5. Results & Signal Strength

• Strong access (green light): Multiple channels yielded replies within 24–48 hours; women were responsive and detailed in sharing.
  
• Weak/blocked access indicators: None at this stage; only challenge was scheduling.

Your assessment: Green light.

6. Knowledge Update

• Validated: Professional women commuters are reachable through both digital networks and physical coworking channels.
  
• Overturned: Expected LinkedIn to be slow/cold; instead, replies were immediate and engaged.
  
• New insight: Slack posts revealed adjacent contexts (late-night walking, solo travel) worth exploring later.

7. Next Steps

• Schedule first round of exploratory conversations with respondents.
  
• Prepare observation plan at evening transit stops to triangulate.
  
• Consider expanding to student commuters as a comparison group.

Attachments (Data Room pointers):

• Screenshot of LinkedIn DM template
  
• Copy of coworking email invite
  
• Slack forum post + anonymized replies
  
• Quick tally of response counts (CSV)

What Diamond 1 Produced

A community the team could name and justify, and evidence they could reach it: twenty-one people approached across three channels, fifteen replies, nine willing to talk at length.

What it did not produce is a customer. The choice was recorded as a working commitment rather than a final one, and the people the team ends up serving are settled in Diamond 2, by who turns out to carry the pain.