Two platforms replaced by one storefront the brand controls end to end

Pashmina.com is a New Delhi–headquartered luxury house built on a craft rather than a category. The fibre comes from the underbelly of Changthangi goats grazing at high altitude in Ladakh; artisans in Kashmir hand-spin and hand-weave it into shawls, wraps and scarves using techniques — Kani, Jamawar, Sozni, Kalamkari, Zari and Tilla — where a single piece can take months. The company works with more than a hundred artisan families, reinvests a share of every sale back into them, and has each piece tested and certified by the Crafts Development Institute under the Ministry of Textiles.

In this category, the storefront is not a shopfront; it is the evidence. Eighty-five per cent of the business comes from outside India, which means the persuading happens at a distance — to a customer who cannot touch the fabric, often reading in a second language, deciding whether a piece worth hundreds is genuine. Machine-made blends are sold worldwide under the same word, so that is the one question standing between them and the purchase. And with forty-five per cent of customers returning, the experience is not a single transaction to get right; it compounds.

The systems underneath were working against that. Commerce ran on Magento, while the editorial library — the asset carrying much of the organic traffic and most of the authenticity argument — lived separately on WordPress. Two platforms, two content models, and a storefront experience assembled across a seam the customer could feel. It was also slow. The storefront scored 32 out of 100 on Google PageSpeed Insights when the engagement began — a number that costs most on exactly the traffic this business runs on: international buyers, on mobile connections, a long way from the server.