D2C growth starts with a customer problem, not a channel hack

The product does not need to be revolutionary

When I first wrote about D2C growth in 2023, the central point was simple: many strong consumer businesses do not invent a new category. They remove a frustration from an existing one. A clearer assortment, easier delivery, lower perceived risk or a better buying experience can be enough to change customer behaviour.

That idea still holds. What I would add today is that the customer problem must remain the organising principle after launch. Teams often replace it with channel activity. They discuss Meta, Amazon, retail or a new country before agreeing which customer tension the expansion is meant to solve.

Separate the promise, the proposition and the operating model

A D2C model becomes easier to scale when three layers are explicit. The promise is the problem you solve. The proposition is the product, price, proof and experience that make the promise credible. The operating model is how acquisition, stock, fulfilment, service and returns deliver it profitably.

At Emma, simplifying mattress choice was not only a communication idea. A curated assortment, understandable benefits, a trial period and a convenient online journey worked together. Later, while helping scale the Beds category across more than 15 markets, I saw how important it was to protect that coherence while adapting price tiers, media, financing and operations locally.

• Promise: which customer frustration are we removing?
• Proposition: why should the customer believe and choose us?
• Operating model: can we deliver the experience at a healthy contribution?

Use data to improve the decision, not to avoid judgement

Digital commerce creates an abundance of signals: conversion, reviews, returns, search behaviour, repeat purchase and service contacts. The hard part is not collecting more data. It is deciding which signal should change the product, message or operation.

Start with a hypothesis tied to the customer problem. Then define the evidence that would support or reject it. A high conversion rate may validate the proposition, but weak contribution or high returns can still invalidate the operating model. Growth is real only when those views can be reconciled.

Expand the learning system before the footprint

A new market or channel should inherit what the business already knows: the customer tension, the minimum viable proposition, the economic threshold and the questions that still need answering. Local teams can then adapt the execution without rebuilding the logic from zero.

The useful growth hack is not a tactic. It is a disciplined loop: observe a problem, build the smallest credible answer, measure the complete economics, learn and only then widen the footprint.