Off-peak metro relay that moves many carriers' parcels into the city core
TL;DR Metrorelay runs a shared parcel relay on metro lines. Cages of sorted parcels from several carriers ride off-peak trains from a depot-side hub to small station counters, and e-bike riders finish the last few kilometres. Carriers pay per parcel and the metro earns from capacity it already has.
The priciest stretch of a city delivery is the van crawl from an edge-of-town hub into dense inner neighbourhoods. Meanwhile metro trains run nearly empty in late mornings and at night, on lines that go straight into those areas. Delhi Metro has noticed. It signed an MoU with a major express courier in March 2025 after short trials on the Blue Line, with plans to widen to other lines[1], and the deal includes micro parcel hubs at stations in off-peak hours[2].
But that's one carrier on one network. Smaller carriers, regional 3PLs and D2C brands with their own riders can't each negotiate with a metro corporation.
And no Indian metro has published a freight tariff, volumes or results.
Metrorelay is the neutral middle layer. It books off-peak freight windows from a metro corporation and then sells space on them to any carrier by the parcel.
Carriers drop pre-sorted parcels at a Metrorelay hub next to a depot or terminal station. Parcels go into lockable cages sized for the train doors and ride in a set coach during the agreed window. A small counter at each destination station receives the cages, and electric cargo bikes deliver within about 2–3 km, the radius planners use for micro-depots[3].
The software does two jobs - it fits each day's parcels into the booked train windows and it plans rider routes from each station. That's where the seed's AI-powered routing belongs.
I've reframed the seed, which pitched transit hubs as micro-fulfilment centres holding stock. Station space is small, rented from the metro and shared with passengers, so storing inventory there is wishful thinking. Metrorelay holds nothing overnight at stations and only passes parcels through.
Carriers pay per parcel moved, and the price has to beat what a van run into the core costs them today. There's no public price for that comparison, so you'll have to get it from carriers directly. The metro corporation is a partner and not a customer, earning a share or a fixed fee for each window. Brands with their own same-day riders are a second buyer for station-to-door delivery.
The March 2025 MoU between DMRC and that courier is the first time an Indian metro has formally opened its trains to freight[1]. DMRC called it the first such service in South Asia Pacific and said it was learning from Madrid Metro's parcel pilot[2]. Demand keeps climbing too, with online retail at about US$80B in FY26, up 21% on the year[4].
Nobody publishes a figure for parcels moved by metro in India, or for what a freight window costs. The narrowest evidence is qualitative - one national carrier has signed with DMRC and is trialling[1].
Online retail of about US$80B (roughly ₹7.7 lakh Cr) in FY26[4], most of it ending in city deliveries, is a neighbour for context only.
The courier's DMRC arrangement is the closest, but it's a single-carrier deal[2]. DMRC tried a different idea in 2016, with automated parcel terminals at 10 stations for e-commerce click-and-collect, and published no results[5]. A locker maker offers local delivery and an edge-warehousing beta in nine cities[6], but it doesn't use trains.
I found no operator in India that pools parcels from many carriers onto metro trains.
| Difficulty | Hard (56/100) |
| Starting capital | ₹3–6 Cr |
| Time to launch | 12–18 months |
The software is the easy bit. The hard part is a contract with a public metro corporation that will want security screening, fixed windows and cleanliness standards, and that will move at its own pace. You also need enough carriers on day one to fill each window, and transfers at both ends that save more time than they cost.
There's no courier licence, but you can't run a single cage without the metro corporation's agreement.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
AI shouldn't clear parcels for security on its own. The metro's screening rules need a trained human to sign off every exception.
Want to work out the AI layer for Metrorelay? Talk to Yash →
Here's what can go south -
This is for a founder with logistics operations experience and real patience for public-sector contracts. If you want a quick software sale then I'd look elsewhere.
Before raising, prove one thing on one line - that a pooled cage reaches the destination station faster than the carrier's own van, for at least three carriers.
Plan for roughly ₹3–6 Cr, covering hubs, cages, station counters, cargo bikes and the software, before the first metro contract pays back.
No courier licence, but you need a signed agreement with the metro corporation. Add company registration, GST and DPDP compliance for recipient data.
Yes, if the city's transit operator will sell off-peak capacity. The contract terms, screening rules and data protection law will be local, so start there.
The free Metrorelay validation kit has the line-by-line capital breakdown, the effort scoring, the metro-contract and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for carriers and metro officials, and 5 assumptions to test. Download the kit →
Thinking about building Metrorelay or something close to it? The make-or-break work is winning freight windows from a metro corporation and filling them with several carriers at once, and I can help you plan both sides of that. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.