Case · Mass Customisation and Print on Demand

Print on Demand: When Design Became Free, the Model Lost Its Moat

A category whose core input dropped to zero cost overnight — and the uncomfortable lesson about what was actually scarce all along.

At a glance

Customised Products · Global · Category reading

01

The Category and Its Promise

Chitrangana has worked inside this category. What follows is not a reading of any one business — it is a reading of the category itself.

Print on demand removed inventory risk from product commerce entirely. A creator uploads a design; the platform prints it onto a mug, poster, or shirt only when someone orders; the creator earns a royalty and never holds stock. For the platform it was better still — a marketplace of infinite catalogue, funded by other people’s creativity, with production triggered by demand.

The scarce input holding the whole model together was design: original work that people wanted enough to buy.

Why it broke. Generative AI made the scarce input free, and every downstream part of the model — discovery, creator economics, production — was built assuming it never would be.

  • The scarce input stopped being scarce. Generative AI made competent design instant and effectively free. Catalogues that took a decade to build can now be replicated in a weekend. Every marketplace in the category faced the same arithmetic: unlimited supply against fixed demand.
  • Discovery collapsed under volume. Marketplace search cannot rank millions of near-identical designs meaningfully, so visibility became a lottery. Platforms have responded with rules rather than economics: enforcement against what one marketplace calls “AI spam farm” accounts, mandatory AI disclosure on another, tagging requirements for generative content, and curated submission processes filtering low-effort work.
  • The creator’s economics were always thin, and thinned further. Royalties on a commodity product, on a platform the creator does not control, in a catalogue with unlimited competitors. Even established platforms are now characterised as requiring long-term SEO and keyword strategies for patient niche sellers rather than offering fast or large sales — an accurate description of a category where the median participant earns very little.
  • Production stopped being a differentiator. Distributed local production is no longer a novelty; cross-border sellers now reasonably expect orders to print in the buyer’s country. When fulfilment quality and speed equalise across providers, a platform whose advantage was its print network has no advantage left.

02

What Changed

The value in this category has moved from making the design to having a reason for it to exist. Three things are now scarce where design is not: rights, relationship, and occasion. Licensed intellectual property cannot be generated. An audience that trusts a specific creator cannot be prompted into being. And a product tied to a real moment — a wedding, a team, a company milestone, a personal photograph — has a buyer who wants that exact thing and no substitute.

Meanwhile the marketplaces’ own filtering has created an opening: platforms actively separating human, curated, or verified work from generated volume are effectively creating a premium tier, and someone will own it.

The renewed opportunity. The strongest positions are now: licensed and IP-backed merchandise, where rights are the moat; creator-owned storefronts where the audience relationship — not marketplace search — drives sales; occasion and personalisation commerce, where the design is about the buyer rather than about taste; and corporate and B2B merchandise, an unglamorous, high-repeat, relationship-driven segment that generative flooding barely touches because the buyer is a procurement decision, not a browsing consumer.

India’s opening sits squarely in the last two: corporate gifting at scale, and regional occasion commerce — weddings, festivals, institutional merchandise — where local design sensibility and local production together beat a global generated catalogue.

03

Chitrangana’s Transformation Advisory

For customisation platforms and product businesses, in order:

  1. Move from selling designs to selling rights, relationships, or occasions. If a competitor can generate your catalogue tonight, your catalogue is not the business. Identify which of the three scarce assets you can actually own and build there.
  2. Escape marketplace discovery. Owned storefronts, creator audiences, B2B relationships — any channel where a buyer arrives asking for you rather than searching a category. Marketplace ranking in a flooded catalogue is not a strategy; it is a queue.
  3. Build for the personalisation that AI cannot fake. The customer’s own photograph, their team’s name, their company’s milestone, their wedding date. Generated design competes on aesthetics; personal meaning has no competitor.

Repositioning a customisation business away from commodity design is eCommerce Consulting work; understanding how generative systems change what is scarce in a category — and acting before rivals do — is AI Consulting.

When the design costs nothing, the design is not the product. The reason someone wants that exact one is.

Chitrangana

Building in this category?

Every engagement begins with the Business Architect.

A working session on your model, not a pitch. We map where the money is actually made, then agree what to build first.

01ThinkWhere the model earns, and where it quietly leaks.
02ValidateTest the thesis against your numbers before anyone builds.
03ExecuteDeploy it, then hand you the operating system for it.

Each case describes the business and its model as they stood during the period the case draws on; a company’s subsequent history is its own. The cases on this page draw on Chitrangana’s professional work and study, carried out directly or with partner firms, with the working record held in confidence. Brand names and trademarks belong to their respective owners; their appearance does not imply endorsement, affiliation, or partnership.