DirectShelf

DirectShelf is an early-stage Indian platform selling warehousing, fulfilment and delivery to direct to consumer brands, pitched on quick-commerce speed: "2-4 hour deliveries. Zero setup cost, go live in 72 hours." It is a fulfilment provider using its own multi-city warehouse network rather than a carrier in the parcel sense. There is no API, no developer documentation and no public technical surface of any kind. The marketing site is a static single-page React application with a Google Form as its only backend, and the seller application host app.directshelf.in resolves but returned HTTP 504 Gateway Time-out on every request made for this page. Treat this as a very early or currently non-operating platform.

At a glance

What it is

The positioning, taken from the site's own title and metadata: "One-click platform for Warehousing, Full-fillment and Delivery solution for D2C Brands", and "Transform your D2C brand with 2-4 hour deliveries. Zero setup cost, go live in 72 hours."

The model is shared infrastructure rather than owned infrastructure. The pitch is that a brand uses DirectShelf's existing warehouse network in several Indian cities instead of building its own, and can therefore "Launch in new cities within 72 hours using our existing infrastructure" and "Scale to new cities without infrastructure investment". Named benefits include "Strategic placement reduces delivery time by 40%", "Smart routing ensures 99.5% on-time delivery rate", "Smart demand forecasting" and "Multi-City Warehousing".

The service description is "End-to-end order processing with intelligent assignment for guaranteed 24 hour delivery", alongside a separate "DirectShelf Quick Commerce Service" whose process is given as "On-board, Connect, Store, and Pick & Pack". So there are two speed tiers: quick commerce at 2 to 4 hours, and standard fulfilment at 24 hours.

One of the stated pain points is revealing about the intended positioning: "Relying on third-party platforms means losing control over customer experience and data", with the counter-benefit "Complete control over customer experience and data". That is the dark-store alternative to selling through Blinkit, Zepto or Instamart, aimed at brands that want quick-commerce speed on their own storefront.

The business is not only logistics. The same site sells "Enterprise Office Setup" and "Workspace Infrastructure": managed offices and coworking, office sourcing, lease structuring and renewals, employee relocation support, and "Large-format office setup and multi-city expansion support for enterprises and GCC operations". A fulfilment platform and a commercial real estate services business under one brand is an unusual combination, and it suggests a services company with a technology front end rather than a technology company.

Related brand. The marketing bundle references zippstor.com, which presents itself as "ZippStore, India's First Retail-as-a-Service Company". The relationship between DirectShelf and ZippStore is not stated anywhere, but the cross-reference is in DirectShelf's own code.

No company registration, legal entity name, founding date, city list, warehouse count or client name could be found on the site. The testimonials name roles ("Chief Operating Officer", "Head of Supply Chain", "Supply Chain Operations Manager") without naming companies.

API access

None. The specific checks:

• api.directshelf.in, docs.directshelf.in, dashboard.directshelf.in and directshelf.com do not resolve in DNS.
• app.directshelf.in does resolve, to k8s-default-appingre-...ap-south-1.elb.amazonaws.com, which is an AWS load balancer fronting a Kubernetes ingress. Every request to it, over both HTTP and HTTPS, on / and on /track, returned 504 Gateway Time-out from nginx. The infrastructure exists; nothing is answering behind it.
• directshelf.in/sitemap.xml returns the single-page application's index.html rather than a sitemap, which is the usual signature of a static host with a catch-all rewrite. There is no page index to enumerate.
• The site is a Vite-built React single-page application deployed from a Netlify site (direct-shelf.netlify.app) with assets on an S3 bucket named websiteassets-wisetutor. Its contact form submits to a Google Form.
• Searching the compiled JavaScript bundle for HTTP endpoints returns only app.directshelf.in/track, zippstor.com, a Google Forms URL, and standard React and SVG namespace URLs. There is no API client in the front end at all.

There is no published sales contact, phone number or email either, only the embedded form. Onboarding is described as a "Simple onboarding process to scale your D2C brand across cities" with no detail.

Authentication

Not published. Nothing exists to authenticate against.

Objects we can read

Nothing is readable. The only referenced surface is a tracking page at app.directshelf.in/track, and that host is not responding.

Orders, order items, products and listings

None exposed. A fulfilment platform of this shape would necessarily hold orders and line items internally, since it picks and packs them, but nothing is published.

Inventory

This is the object that would matter most. A dark-store or multi-city warehouse model means DirectShelf holds the brand's stock at its own locations, so stock-on-hand per city is the thing a unified commerce database would most want from it. Nothing is published, and there is no evidence of an inventory interface.

Shipments and tracking

A customer-facing tracking page is referenced. No machine-readable tracking, no scan events, no status vocabulary, no AWB model. Delivery is described as "intelligent assignment" and "smart routing", so a carrier allocation layer exists internally, but which carriers or own riders are used is not stated.

Returns and cancellations

Not mentioned anywhere on the site.

Payments and settlements

Not mentioned. Pricing is "Zero setup cost" with no published rate card.

Customers

Not exposed.

Locations

"Ready-to-use infrastructure across major Indian cities" and "Strategic warehouse network across major cities with zero upfront investment", with no city list, no address list and no count.

Writing back: listings, price and stock

Not applicable, this is a fulfilment and delivery provider, not a sales channel. DirectShelf owns no buyer-facing listing and no price. Catalogue and price writes belong to the brand's own storefront connector.

The write path on a platform of this kind would be a fulfilment order create (send an order to be picked, packed and delivered), a cancel, and probably an inbound stock advice so DirectShelf knows what is arriving at its warehouses. None of those is published, and with the application host down there is nothing to test against.

Webhooks and notifications

Not published. No event catalogue, no subscription mechanism, no signature scheme, nothing to poll.

Rate limits and pagination

Not applicable. No API exists to limit or paginate.

Mapping to the unified model

Nothing can be mapped today. The shape to request, if the platform turns out to be trading and willing to integrate, in priority order for a commerce database:

listings and products are out of scope for this platform.

Gaps and open questions

• Is the company trading? The seller application returns 504 on every path tried. That could be a transient outage, a service scaled to zero, or a wound-down product. This question comes before every other one.
• No legal entity, no city list, no client names. Nothing on the site identifies the operating company, the warehouse locations or a single named customer. For a platform that would be holding a brand's physical inventory, that is a due-diligence problem well before it is an integration problem.
• The relationship with ZippStore. DirectShelf's own bundle links to zippstor.com, "India's First Retail-as-a-Service Company". Same team, parent, or spin-out is unknown.
• Logistics and office real estate in one brand. The same site sells warehousing and fulfilment alongside managed offices, lease negotiation and employee relocation. Which is the primary business is unclear, and it affects how much engineering investment the logistics side is likely to receive.
• Inventory visibility is the whole value. If DirectShelf cannot expose stock per location by API, a brand using it loses visibility of its own inventory, which is worse than the third-party-platform problem the site says it solves.
• Carrier or own fleet. "Intelligent assignment" and "smart routing" for 2 to 4 hour delivery implies either an own rider fleet or hyperlocal carrier partners. Neither is named, and it determines what tracking data could ever exist.
• Web search was unavailable for this research pass. The checks were direct: DNS, HTTP, the compiled front-end bundle, sitemaps and aggregator directories. A search pass might surface funding news or a registration record that settles the trading question.

Sources

• DirectShelf home page, positioning, service descriptions and claims
• https://directshelf.in/assets/index-B9Cghivz.js, the compiled front-end bundle, which is the only source of the site's full copy and confirms there is no API client and no backend beyond a Google Form
• https://app.directshelf.in/ and https://app.directshelf.in/track, both returning HTTP 504 Gateway Time-out from nginx behind an AWS ap-south-1 Kubernetes ingress
• ZippStore, the related brand referenced from DirectShelf's own bundle
• Negative results, verified directly: api.directshelf.in, docs.directshelf.in, dashboard.directshelf.in and directshelf.com do not resolve; directshelf.in/sitemap.xml returns the application shell rather than a sitemap; the ClickPost carrier index, enumerated from its sitemap, contains no DirectShelf page; Unicommerce's public integration list does not name it