BigMint

A data-first platform for commodity market intelligence

A commodity market-intelligence platform running a subscription price-reporting business and a newsroom together — hundreds of price assessments as auditable time-series, entitled correctly and delivered across web, REST API, Excel plugin and mobile.

Client: BigMint
Industry: Commodity Intelligence
Location: India
Duration: 24 weeks

Visit bigmint.co

Laravel

Platform

Price assessments published

Countries covered

Web, API, plugin, mobile

Delivery channels from one source

The challenge

BigMint is a commodity market-intelligence platform for the metals, mining and broader commodities world — price assessments, indices, futures data, news and research, statistics, tenders and events. It publishes something on the order of 800+ price assessments across 40+ countries, spanning ferrous, non-ferrous, energy, agriculture and more. A platform like this is not a website with articles; it is a data product with a website on the front.

Why this is a hard content platform

Price data is the product, and it is time-series. Spot assessments, indices and historical series are structured, frequently updated data — not prose. They have to be stored, versioned, charted and delivered, and their credibility depends on being right and traceable (the assessments are positioned as IOSCO-audited).
Access is tiered. A commodity-intelligence business runs on subscriptions and entitlements: some content is open, much of it is gated by what a given subscriber has paid for. The platform has to enforce that boundary consistently across web, API and exports.
Delivery is multi-channel. The same underlying data is consumed on the website, through an API, through an Excel plugin, and on mobile. One source of truth has to feed several very different clients.
Breadth is enormous. Dozens of commodity families, dozens of countries, and multiple content types (prices, news, reports, data, tenders, events) all coexist and cross-reference.

The core challenge is running a subscription data business and an editorial newsroom on one platform, where price series, research, news and events are all first-class and all correctly entitled.

What we built

We treat BigMint as what it is — a data platform first, a publisher second — and build the web experience on top of a properly modelled data core rather than the other way round.

Prices as structured time-series, not articles

Assessments and indices are modelled as first-class time-series: a commodity, a market/geography, a unit, a timestamp and a value, with history preserved. That is what makes the data chartable, comparable across periods, exportable, and — crucially for an audited price-reporting agency — traceable back to a specific published assessment rather than an overwritten number.

Entitlements as a cross-cutting concern

Subscription access is enforced centrally, so the same entitlement rules apply whether a user hits a web page, calls the API, or pulls data through the Excel plugin. Building this once, in the platform core, is the only way a multi-channel data business stays consistent — you cannot re-implement "who is allowed to see this price" in three places and expect them to agree.

One source of truth, many delivery surfaces

The website, a REST API, the spreadsheet plugin and the mobile experience all read from the same governed data. Heavy work — indexing, aggregation, export generation, dashboard queries — is pushed onto background queues and caching so that a spike in traffic or a large export never slows the read path that subscribers live in.

An editorial engine beside the data engine

News, research reports, statistics, tenders and events are managed as their own content types alongside the price data, cross-referenced to the commodities they concern. On a Laravel + relational + Redis foundation, the newsroom and the data product share one platform instead of fighting over two.

The outcome

The result is a single platform that runs a subscription price-reporting business and a commodities newsroom together — data, research, news, tenders and events, entitled correctly and delivered across web, API, plugin and mobile.

What the platform carries

Hundreds of price assessments and indices across many commodity families and geographies, stored as time-series that can be charted, compared and exported.
Multi-channel delivery from one source — website, REST API, Excel plugin and mobile all reading the same governed data under the same entitlement rules.
A full editorial estate — news, research reports, statistics, tenders and events — cross-referenced to the commodities they cover.

Why the data-first architecture is the whole game

For a market-intelligence provider, two properties decide whether the business works: the data must be correct and traceable, and access must exactly match what each subscriber has paid for. Modelling prices as auditable time-series and enforcing entitlements centrally is what delivers both, and it is what lets the same data safely appear in an API response, a spreadsheet cell and a web chart without three different answers.

Because the platform is built on one governed core, adding a commodity, a country, a content type or a new delivery channel is an extension of the existing system rather than a parallel build — which is exactly what a platform tracking a growing universe of 800+ assessments needs in order to keep growing.

Under the hood