The Inbox15 May 2026

— Asked of the desk

The questions buyers
ask first.

Honest answers about what we do, where we win, and where we don't. Each one signed by who answered it — Anaita for the buying-floor questions, Yash for the business and data ones.

A designer's pinboard crowded with folded handwritten question slips in marigold, sage, maroon, and ivory paper

Questions, as they arrive.

— Inbox to date

47

Buyer questions answered since launch. A small number, growing honestly. Counted as of May 2026.

— Filter by topic

Showing 5 questions on forecasting.

01

— Chapter 01

First, what this is.

Two answers that set the frame — what we are, and the voice the engine speaks in.

  • What is TrendSense, exactly?

    Asked of Anaita Verma · senior-buyer desk

    A buying-decision platform for Indian D2C womenswear brands. It pairs your own sales history with the Indian festive calendar, current runway, fashion media, and competitor cataloguesto tell you what to buy in the next 30 / 60 / 90 days, how deep to go on each SKU, and where the season's energy is moving. The buyer briefs are written in the voice of Anaita Verma — the senior-buyer character TrendSense was built around.

  • Who writes the buyer briefs?

    Asked of Anaita Verma · senior-buyer desk

    Anaita Verma — the senior buyer voice TrendSense was built around. She's a composite character drawn from twelve years of Indian D2C womenswear buying decisions: the ones that worked, the ones that left ₹80L of bandhani in the warehouse. Every brief she writes cites the signals it leaned on, names the price band, says how many SKUs to load, and explicitly flags what could go wrong. When the data is thin she'll say “early signal — watch, don't load” rather than make something up.

02

— Chapter 02

Under the hood.

How the forecast is made, what feeds it, and how the Indian festive calendar lifts the maths.

  • How does the festive calendar lift the forecast?

    Asked of Anaita Verma · senior-buyer desk

    Forty-two windows in the Indian commercial calendar (Diwali, Karva Chauth, Eid, Onam, wedding season, EOSS, Republic Day sale, Independence Day sale, etc.) each have category multipliers — “festive 1.6×, party 1.3×, wedding-guest 1.2×” — baked into the forecast at the affected horizon. The lift is broken out per product on the Forecasts dashboard and per category on the Assortment Plan view.

  • How accurate are the forecasts?

    Asked of Anaita Verma · senior-buyer desk

    Every account's accuracy is visible on the Forecasts dashboard: average error (Mean Absolute Percentage Error), revenue-weighted error (Weighted Absolute Percentage Error), and what percent of actuals landed inside the 80% prediction band. Below 6 weeks of history the engine falls back to analog-based forecasts — other products in your catalogue that look most like the new SKU. Numbers are directional until 12+ months of operational data accumulate.

04

— Chapter 04

Edge cases & honesty.

Sustainability scoring, what happens to your data if you leave, and where we differ from the global incumbents.

  • How is this different from a global trend platform?

    Asked of Anaita Verma · senior-buyer desk

    The global incumbents (built for European retail) sell a worldview at a per-seat price: their calendars centre Christmas, their fabric coverage is light on silk-blend banarasi and pure georgette, and their runway tagging stops at the four big international weeks. TrendSense is built to sit on the Indian buying floor. The festive calendar drives the forecast horizon. The fabric library was indexed against what Indian D2C brands actually load on. The runway feed watches the four current Indian designer seasons, not last September in Milan. We are not a cheaper version of the global tools — we are a different tool, made for a different desk.

    OnTrendSenseGeneralist platform
    Calendar42 Indian commercial windows, weighted per-categoryChristmas, Black Friday, Easter, Mother’s Day
    Fabric coverageBanarasi, chanderi, kanjivaram silk, georgette, kota, linen-blends — indexed against Indian D2C loadsCotton, polyester, denim, wool — Indian fabrics rolled up as "ethnic"
    Runway taggingFour current Indian designer seasons, 30 designers, per-look attribute taggingThe four international weeks; Indian shows tagged at couture-week level only
    Where the desks differ.
The questions we cannot yet answer are more interesting than the ones we can. They are the next issue.
— From the editor's desk

— Asked, not yet answered

The open queue.

Questions in the inbox we haven't yet committed to in print. Listing them is the editorial move — it admits where we are still working. Push the one that matters to you and we will pull it forward.

  1. 01

    How do you handle saree blouse SKU permutations — colour × sleeve × neckline?

    In the queue · taxonomy team is sketching the variant model.

  2. 02

    Can the engine see Instagram Reels signal yet?

    Not yet · prototype shelved until the audio side is solved.

  3. 03

    What about Tier-2 and Tier-3 city buyers — does the signal hold outside the metros?

    In the queue · waiting on three pilot brands selling outside the top eight cities.

  4. 04

    When will there be a Hindi or Tamil version of the brief?

    On the roadmap · the brief generator is being audited for translation first.

  5. 05

    Can you forecast on a category I make up — say "festive co-ord set under ₹2,500"?

    Partially · custom price-band slicing works; custom category creation does not.

Have one of your own?

Submit your question

— Still curious

Want a question answered live?

The next issue of the inbox is, in part, made of the replies we get. Write to us — or read this week's brief while you wait.