Business model — who would pay for Wearly¶

Why this document exists. SEIS 666 best-practice #6: "if your project solved a real problem for a real organization, who's the buyer? What's the ROI?" This page answers that without overclaiming.

The problem Wearly addresses¶

People spend a measurable amount of time and mental energy on outfit decisions every day. Existing tools fragment the problem:

Weather apps tell you the forecast — not what to wear in it.
Calendar apps tell you what you're doing — not what's appropriate.
Closet-organizer apps catalog items — they don't reason about them.
Generic styling AI tools suggest outfits — but with no knowledge of your closet, your fit preferences, or your schedule, and with no audit trail.

Wearly composes those four data sources into one rule-driven recommendation with a full reasoning trail.

Three potential buyers¶

1. Consumer subscription — primary path¶

Aspect	Value
Who	Time-pressed professionals who own a meaningful wardrobe and want it to feel intentional
Price hypothesis	$4–$8 / month (validate via interviews — not claimed)
What they get	The Streamlit experience with cloud-persistent profile, calendar/weather integrations, wear history, and reject-and-regenerate
Comparable products	Cladwell ($5/mo), Acloset ($3–6/mo) — neither offers Wearly's reasoning-trail/audit-trail differentiator

2. Personal-stylist tool — B2B SaaS¶

Aspect	Value
Who	Independent stylists, capsule-wardrobe consultants, "uniform" coaches
Price hypothesis	$20–40 / month per stylist
What they get	A multi-client dashboard, the compact-KG export per client, the Skill package they could load into their own LLM agent
Differentiator	Rule citations + verified evidence categories give a stylist defensible reasoning to share with a client

3. Corporate wellness / DEI tooling — niche¶

Aspect	Value
Who	HR programs, return-to-office initiatives, employee-resource groups
Price hypothesis	License-based — not directly inferred
What they get	The body-positive language contract becomes a feature, not a constraint — Wearly is the rare styling tool that, by design, never frames a body as a problem
Differentiator	Documented contract, runtime enforcement, citation-traced inclusive-design references (see Hokka 2024 in evidence-and-references §3.4)

Why the audit trail is the moat¶

Every other styling app gives you a recommendation. Wearly gives you a recommendation plus:

A numbered reasoning line per decision.
Each line citing a Skill rule (e.g. [occasion-rules#R3]).
Each rule traceable to an evidence category.
Each evidence category carrying verified primary citations (Springer 2022, ACM 2017/2018, ECCV, ISO 11664-4, EUR-Lex GDPR, Cavoukian 2009).

For a paying customer this matters because:

Consumers can disagree with a recommendation and have a coherent conversation back with the system (reject-and-regenerate surfaces what changed and why).
Stylists can defend their advice in front of a client.
Corporate buyers can audit the tool against inclusive-design and privacy standards.

The course principle "structure beats prompts" cashes out here as a real product moat: anyone can wrap GPT in a prompt; almost nobody has a rule-cited reasoning trail with verified evidence.

ROI framing (illustrative, not claimed)¶

Buyer	Their cost without Wearly	Their cost with Wearly
Consumer	~15 min/day deciding × 5 days = 1.25 hr/week	<1 min × 5 = 5 min/week
Stylist	30 min per client outfit recommendation × 20 clients = 10 hr/week	5 min review per recommendation × 20 = 1.5 hr/week
Corporate program	Manual "what to wear to return-to-office" guidance	Self-service tool with body-positive contract built in

These numbers are illustrative and would need user-research validation before being claimed in marketing. Wearly's documentation discipline (no invented citations, "tends to work well" not "is correct") applies here too.

What this prototype is not — yet¶

Not a transaction platform (no real retailer catalogs, no checkout).
Not a multi-user product (single-user local-only data).
Not a mobile-native app (Streamlit is mobile-responsive, but the production version would be a true iOS/Android app).
Not a personalization engine that learns implicitly (Wearly is deliberately explicit-only; see Hu, Koren & Volinsky 2008 cited in evidence-and-references §3.8).

All three of these are paths the roadmap (book/11-product-roadmap.md) names as future production work, not promises this prototype delivers.