Consulting engagement · ~20 months, within 2018 to 2024

fromKhet

Product Management Consultant · Agri-tech: local farm-to-home delivery

fromKhet delivered fresh, organic, hygienically handled vegetables and fruit from farms to households, a local farm-to-home delivery service (khet is Hindi for farm). I worked with its distributed team of ~7 for ~20 months as a freelance, remote product management consultant.

In brief

fromKhet’s founders had sourcing, warehousing and farm partnerships in place during the first COVID-19 lockdown and wanted to build an app straight away.

I recommended validating demand on WhatsApp first, which ran for ~3 months before app development began, and designed the operations, including the logic of a delivery route algorithm, before the consumer app.

Retention reached ~60% at D14 and ~50% at D30, with the third order as the activation threshold, and spoilage and discard stayed below 5% throughout.

The problem

The co-founders started fromKhet during the first COVID-19 lockdown. They had already worked out sourcing, warehousing and farm partnerships, and wanted to build an app straight away. I recommended validating demand on WhatsApp first.

I defined the target user as tech-savvy, health-conscious millennials working remotely and living with their families. Service was limited to fewer than seven adjacent residential areas, because only three team members were in the city and the founder delivered in his own car.

What I did, phase by phase

Phase 1July to September 2020, ~3 months

WhatsApp validation

Pamphlets designed around user-behaviour principles were posted across the target areas, with the founder’s number as the contact. The first week brought 10 to 15 orders, after which orders settled at 5, then 7, then 8 a week, growing ~10% week on week. Orders went to a central WhatsApp group and an Excel sheet, and farmers received next-day requirements by text.

Orders placed by 10 PM were delivered between 7 and 10 AM the next morning. At 4 to 5 AM the founder went by car to a cluster of fewer than 50 partner farmers on a highway outside the city, brought the produce to a makeshift warehouse on his rooftop, cleaned, sorted and packed it with one helper, planned the route by pinning addresses on Google Maps, and delivered personally, finishing around 11 AM, seven days a week.

Local farming was controlled by unofficial community leaders who expected a cut, and small farmers needed their permission to work with fromKhet. The team resolved this well enough to operate, but it remained a source of friction and later contributed to sourcing problems.

Phase 2September to December 2020

Operations first, then the app

The first technical build addressed the founder’s workload, and the consumer app came after it.

  • A local transport vendor took over farm pickups, and a second vendor took over last-mile delivery.
  • A route planning algorithm processed each day’s delivery addresses and produced a route optimised for distance and time. Addresses changed daily, so manual planning did not scale. I designed the algorithm’s logic; I did not write its code.
  • After delivery, the vendor marked the order done and the customer received an automated WhatsApp confirmation.

App and web app planning began around September 2020. The central design problem was inventory uncertainty at the farm: an item shown as available in the evening might be unavailable at 4 AM, and a refund would make for a poor experience. The system was designed to handle that uncertainty.

Phase 3December 2020 to March 2021

App launch and growth

The app and web app launched around December 2020, after ~5 months of WhatsApp-only operation (July to December 2020).

  • Referral programme: the referred friend and the referrer each received INR 100 off an order of at least INR 500. From January 2021, ~15% of new users came through referral codes.
  • Funnel and UI/UX fixes brought a ~20% higher signup rate and a ~20% lower bounce rate.
  • User acquisition grew ~10% a week, reaching ~15% in weeks with on-ground marketing.
  • With no intermediaries in the supply chain, produce sold at market prices with a slim margin after all operating costs.
Phase 4April to June 2021

Second COVID wave and an evening slot

Demand rose as concern about hygiene grew. Order growth reached ~25% week on week for a short burst; working back from ~85 to 90 orders a week in March 2021 and a peak of ~175 in May to June, the rate lasted about two to three weeks. Referrals rose to ~20% of new users. COVID restrictions, administrative problems and licensing requirements limited fulfilment to ~50% of orders.

In June 2021 fromKhet added same-day evening delivery, a decision informed by customer conversations on WhatsApp: order by 10 AM for delivery between 7 and 10 PM, alongside the next-morning slot. Order volume rose ~20%. Peak demand reached ~25 orders a day (~175 a week) in May to June 2021, with ~12 to 13 a day fulfilled, and the peak was not sustained.

Phase 5

Retention and the third order

I tracked time to first, second and third order and repeat behaviour. A user who placed a third order was retained, which made the third order the activation threshold. D14 retention was ~60% and D30 ~50%. Retained users ordered every 3 to 4 days, about 2 orders a week, and spoilage and discard stayed below 5% throughout.

Phase 6From July 2021

Decline and shutdown

  • The farm laws passed in September 2020 led to protests through 2021 until their repeal in November 2021. Together with local farming politics, this made sourcing unreliable from July 2021.
  • Quick-commerce services expanded into the city around August to September 2021 and offered delivery in minutes, which fromKhet’s delivery windows could not match.
  • Reach stayed limited to fewer than seven localities, a single delivery vendor, and no instant delivery.
  • Margins stayed slim, leaving nothing to fund expansion.

The founders shut fromKhet down. It broke even over its lifetime, and selling direct to consumers added ~15% to 20% to small farmers’ income.

Results

  • ~3 months

    WhatsApp validation before app development began, July to September 2020.

  • ~5 months

    WhatsApp-only operation before the app launched, July to December 2020.

  • about 60% and about 50%

    D14 and D30 retention. A user who placed a third order was retained.

  • ~15% to 20%

    Share of new users from referral codes: ~15% from January 2021, rising to ~20% in the second COVID wave.

  • ~20%

    Rise in signup rate after funnel and UI/UX fixes. Bounce rate fell ~20%.

  • ~20%

    Rise in order volume after the same-day evening delivery slot launched.

  • ~25 a day

    Peak demand in May to June 2021 (~175 orders a week), of which ~50% could be fulfilled.

  • <5%

    Spoilage and discard, throughout.

  • ~15% to 20%

    Added to small partner farmers’ income by selling direct to consumers.

My role and the team

Product Management Consultant, freelance and remote. The team of ~7 was distributed, with 3 based in the city of operation. The founders made the final decisions; I presented options with my reasoning and evidence.

My work covered product management, the target user definition, the WhatsApp-first recommendation, operations design, the logic of the delivery route algorithm, app and web app design, funnel optimisation, the referral programme, and marketing and growth strategy. fromKhet overlapped with my FindX engagement from July to December 2020.