12 — Roadmap

What gets built when, in what order, and what each phase is allowed to delay. Phases are not sprints; they are gates. Each phase must ship working, on the ground, before the next opens.

Phase 0 — Brand refresh (≈ 1–2 weeks)

Goal. Reposition the existing scroll-narrative as the front door of the platform, without touching the React code beyond what is needed.

In scope

• Hero (components/sections/Hero.tsx) gains a secondary CTA:

Subscribe to a plot (→ /subscribe) + Come for a weekend (→ /stay). Existing Reserve a seat preserved.

• New Three pillars interlude section, placed between Founder and

Seasons (see FR-5.1).

• Navigation (components/Navigation.tsx) adds Subscribe, Stay

items.

• The Invitation section becomes a dual CTA; behaviour unchanged.
• Light copy edit on the footer to acknowledge the three pillars.

Out of scope

• Any database or backend work.
• Any change to the seven chapters or any illustrations.
• Removing any existing section.

Done when

• A first-time visitor can reach /subscribe and /stay from the

homepage in two clicks.

• The seven-chapter narrative reads untouched to anyone who knew

the old site.

Phase 1 — Subscription MVP (≈ 2–3 months)

Goal. Run a working weekly-subscription product on a single 1-acre at Lokkanahalli, with manual delivery and human-driven photo updates.

In scope (FR-1.0 … FR-1.6 + FR-4.0 … FR-4.4)

• /subscribe marketing + /zones map + zone deep pages.
• /subscribe/checkout (auth + recurring payment, see NFR-5).
• Subscriber authed area: /my-plot, /my-plot/delivery,

/my-plot/manage.

• Photo updates: admin uploads → email digest (web push optional).
• Admin: zones, deliveries, subscribers, photo updates.
• A single payment provider (Razorpay recommended, but Q-8).
• A basic delivery routing spreadsheet (manual, weekly ops ritual).

Out of scope

• Stay booking, events, shop — pushed to Phases 2 / 3.
• Multi-tenant / multi-site.
• iOS / Android apps.
• Subscriber chat.

Done when

• 30 active weekly Plot-tier subscribers receive a basket every

Saturday.

• Each subscriber has received at least 4 photo digests from their

zone.

• Pause / change / cancel all work end-to-end.
• One full kharif-to-rabi season has been completed.

Phase 2 — Experience (≈ 3–4 months, after Phase 1's first full season)

Goal. Bring the workshop's two-day ritual into a permanent, bookable surface, and add excursions.

In scope (FR-2.0 … FR-2.4)

• /stay marketing + /stay/book booking flow.
• Two cottage rooms retrofitted (Z7), kitchen upgrade.
• /stay/excursions: BRT safari, Jenumutti temple, Tibetan settlement.
• /journal for events + workshops.
• Subscriber stay credits (FR-2.4).
• /about extracted from the current Founder section.

Out of scope

• The shop (Phase 3).
• The Homa / cattle surfaces on the website (Phase 3).
• International tourists / visa-letter flows.
• A mobile app.

Done when

• A subscriber can book a weekender with a BRT safari excursion in

one checkout.

• A non-subscriber can book, period — without needing to subscribe

first.

• Quarterly stay credits are auto-issued to active subscribers and

redeemable.

Phase 3 — Sustenance (≈ 4–6 months, overlapping with Phase 2)

Goal. Complete the model. The farm begins to pay for its own inputs. The first value-added SKUs ship. The first Homa records land in the system.

In scope (FR-3.0 … FR-3.3 + Gau Shala surfaces)

• Shop (/shop, PDPs, cart, checkout) with a narrow catalogue

(ghee, honey, 3 millets, 2 pulses, 2 pickles, 1 soap).

• D2C fulfilment from Bamboo Cafe → curated Bengaluru stockists in

Phase 3 mid.

• GauShalaCattle + HomaRecord admins (low-fi).
• A yearly soil report per zone, written, on the website

(decision: Q-16).

• Cattle on site; first milk→ghee→soap loop closed.
• Biogas plant installation (decision: Q-15).

Out of scope

• A wholesale portal.
• A multi-language site.
• A livestream of the Homa corner (decision: Q-17).

Done when

• The first jar of ghee is sold through /shop.
• The first Homa year has zero undocumented gaps.
• The cattle and compost loops are visible on the platform (without

romanticising them).

Phase 4 — Replication (≈ open-ended, after Phase 3 is steady)

Goal. Make the seed farm exportable. The first sister site is onboarded.

In scope

• Publish the playbook (11_FARM_ZONING.md + 10_FARM_OPERATIONS.md)

as a public document with redactions for trade secrets.

• Onboarding flow for a sister site: zone selection, partner

agreement template, admin training.

• Multi-tenant admin (the seed farm + the first sister site at the

same time).

• Optional: extract the admin into its own codebase; the marketing

surface stays on Next.js.

Out of scope

• A franchise system. Each sister site is sovereign.
• A SaaS pivot. We don't sell the platform as a product.

Done when

• The first sister site has been seeded without anyone from the seed

farm being on-site for the first month.

Things we deliberately never ship

So future-us doesn't argue against them later:

• No pesticides advertised as a feature.
• No imported "wellness" SKUs.
• No D2C meat / non-veg SKUs.
• No international shipping.
• No auto-captions on photo updates.
• No notifications that don't earn their place. If a subscriber

has notifications turned off at the OS level by week 3, our product isn't doing its job.

• No chat with the farmer from outside the platform. Email + the

contact form on /my-plot is enough; chat would tax a small team.

• No livestreams of the farm. It is too easy to romanticise the

life; the platform stays concrete.