Case Study: Packaging for a Madras Filter Coffee Brand | Kavin Arulraj
Kavin Arulraj monogram Kavin Arulraj Hire me →

Packaging · Retail · 2026

A coffee brand that had to work on a 60g pouch

A small-batch filter coffee roaster wanted to move from plain sticker-on-foil packs into proper retail packaging, without losing the regulars who recognised the old pack across the counter.

Client

Filter coffee roaster

Scope

4 SKUs + label system

Timeline

3 weeks

Result

Picked up by 3 stores

The brief

Four blends, sold at the roastery counter and increasingly through two neighbourhood provision stores. The existing pack was a printed sticker on a foil pouch — fine at the counter where someone explains the difference between the blends, useless on a shelf where nobody does.

The ask was straightforward: packaging that a stranger could understand in three seconds, and that a regular could still find without reading.

The constraint

The smallest SKU is a 60g trial pouch. Its printable front face is roughly the size of a playing card, and legally it has to carry the FSSAI mark, the net weight, the ingredients, the roast date, the address and a barcode. Whatever design idea survives that is the design.

Design rule that came out of it

Design the smallest SKU first. Scaling a working small pack up to a 500g bag is easy; shrinking a 500g design down to 60g never is.

What I did

  1. 01 Kept the existing colour. Regulars recognised the deep red before they read anything, so it stayed and became the brand's fixed element across all four blends.
  2. 02 Coded the blends by a single band of colour rather than four different pack designs, so one artwork template covers the range and new blends cost nothing to add.
  3. 03 Set Tamil and English at matched optical weight — the blend name reads equally in both, instead of English leading with a Tamil translation underneath.
  4. 04 Moved every mandatory declaration to a single back panel block, drawn to the printer's dieline so nothing falls into a seal or a gusset fold.

The shelf test

Before approval I printed the front panels at actual size, taped them onto dummy pouches and stood them in a row with six competitor packs bought from a shop nearby. Then I photographed the row from two metres and looked at it as a thumbnail on a phone.

Two of the three concepts vanished into the row. That is a cheaper way to find out than a print run.

The surviving concept was the one with the largest area of flat colour and the fewest elements — which is almost always the answer, and almost never the one that looks most impressive in a presentation.

Outcome

The range went into production three weeks after the first call. Two more provision stores picked it up in the following month, and the template has since absorbed a fifth blend without any new design work — which was the actual point of building it as a system rather than four separate packs.

Gallery

More case studies

All work →

WEB · HEALTHCARE

Booking-first website for an OMR clinic

Cutting a seven-page site down to one decision.

VIDEO · WEDDING

14 hours of footage into a 4-minute film

Selecting for feeling instead of chronology.

Got a product going on a shelf?

Send the dieline, or just the pouch size and the blends. That's enough to quote from.

Start a packaging project See rates