Six SKUs, one funding round, one question: does the startup file seven trademarks or one? Get it wrong one way and you overspend ₹50,000 before revenue. Get it wrong the other way and one bad product review bleeds into everything you sell.
This is brand architecture — the choice between a house mark that covers everything and product marks that stand alone. Tata and Amul built empires on the first model. The world's biggest FMCG companies built theirs on the second, with soap, shampoo and detergent brands that never mention the parent on the front of the pack.
The right answer for an Indian startup is usually neither extreme. Here is the framework, with the filing economics in rupees.
House mark vs product mark: the two architectures
A house mark strategy — the branded house — puts one master brand on everything. Tata Salt, Tata Motors, Tata Consultancy Services: one name, one reputation engine. Amul runs the same play across butter, milk, cheese, ice cream and chocolate. Every product deposits trust into the same account, and every launch withdraws from it.
A product mark strategy — the house of brands — gives each product its own name and keeps the corporate parent invisible. The company behind several of the shampoos, soaps and detergents in your bathroom is often the same one; the brands never say so, by design.
Legally, the architecture decision is a filing decision. It sets how many wordmarks you must register, renew every 10 years, watch in the journal, and defend in court. Choose the architecture first; the filing list falls out of it.
A house mark spends once and compounds. A product mark spends per launch.
The filing economics, in rupees
Government fees are per application, per class: ₹4,500 for e-filing by individuals, DPIIT-recognised startups and MSMEs, ₹9,000 for other companies.
Run the math on our six-SKU founder. Branded house: one mark across four classes as a startup — ₹18,000 in government fees. House of brands: six marks averaging two classes each — ₹54,000. Same company, triple the government cost, before a rupee of professional fees.
The recurring bill multiplies the same way. Every separate mark carries its own renewal on Form TM-R every 10 years, its own journal watch, its own objection replies and opposition fights. Estimate the full stack with the trademark cost calculator, and see the complete fee breakdown in our guide to trademark costs in India.
Every extra mark is a subscription, not a purchase.
When the branded house wins
- Service businesses. Trust attaches to the firm's name in consulting, education, healthcare and software. Clients hire the reputation, so splitting it across product names dilutes the only asset that closes deals.
- Tight budgets. One registration covering your core classes protects everything that carries the name. For most pre-revenue startups this is the entire correct strategy.
- Trust transfer. Launch number seven borrows the credibility of launches one through six. Amul can enter a new category and skip years of introduction because the name arrives pre-trusted.
- One watch, one fight. Policing one name in the journal and the market is dramatically cheaper than policing six.
The cost of this concentration is fragility. One scandal, one recall, one viral complaint touches every product on the shelf. And a name famous for one thing stretches only so far — the market resists a snack brand selling skincare.
When the house of brands wins
- Risk isolation. A recall or controversy on one brand stays on that brand. The rest of the portfolio trades untouched the next morning.
- Segment freedom. The same maker can sell a ₹99 value line and a ₹999 premium line without either buyer feeling cheated — because the brands never meet.
- Clean exits. A standalone brand is a sellable asset. A trademark assignment transfers the mark and its goodwill to an acquirer without touching the rest of the company. Acquirers pay for brands, not SKU lists.
- Category conflicts. If your house name is blocked or weak in a class you need, a standalone product name solves in one filing what an objection fight might not.
The price is gravity. Every new brand starts at zero awareness, the marketing budget divides by the number of names, and unloved marks rot on the register until renewal lapses quietly kill them.
Investors buy companies. Consumers buy brands. Architecture decides whether those are the same asset.
The hybrid most Indian startups actually need
The endorsed-brand model splits the difference. The house mark does the legal and reputational heavy lifting; products launch as descriptive names under it — "Glow serum by Meadow" — and only graduate to their own registrations when revenue proves they deserve one.
The umbrella filing move: register the house mark broadly, covering your current classes plus the classes on the 18-month roadmap. At ₹4,500 a class for a DPIIT startup, buying the runway now is cheaper than fighting a squatter later — India is first-to-file, and roadmaps leak.
Hero SKUs earn their own filings on evidence, not affection: repeat purchase rates, search volume for the sub-brand name alone, distributor pull. When customers start asking for the product by its own name, file it — that is the market voting for a second trademark registration.
Six products and one filing budget? We will map which marks to file now and which can safely wait.
Plan my filing strategy →Licensing and exit change the answer
If the plan includes licensing a product line — co-manufacturing, regional partners, brand extensions — standalone marks make the deal clean. A trademark licence with proper quality-control clauses, recorded through the registered-user route on Form TM-U, is far simpler when the licensed mark is not also the company's own name.
The rule of thumb: franchises run on house marks, sell-offs run on product marks. If a strategic acquirer might one day want your hero product but not your company, that product needs its own registration, held cleanly.
Held cleanly means held by the entity. Investor diligence checks that marks are registered to the company, not to a founder's personal name — and that the architecture on the pitch deck matches the architecture on the register. Fix mismatches before the term sheet, not during it.
The five-question checklist
- Will trust flow between products? If a customer who loves product A becomes likelier to buy product B, a house mark captures that flow. File the house name broadly.
- Could one product embarrass the others? Regulated categories, experimental lines and acquisition targets deserve their own names and their own risk.
- Will you sell or licence any line separately within five years? If yes, that line needs a standalone mark registered to the entity now.
- Do your segments conflict? Premium and value, kids and adults, veg and non-veg — where audiences clash, separate the brands.
- Can the budget carry every mark's full life? Count renewals, watches and enforcement per mark. If the honest answer is no, run the house mark plus graduated heroes.
File the name you cannot afford to lose. Graduate the rest when revenue votes.
Brand architecture sounds like a marketing seminar until the first copycat, the first recall or the first acquisition offer — then it is suddenly a filing history. Write yours deliberately, starting with the one name that appears on every invoice.
Your brand is only yours when you file it.
10,000+ Indian brands filed with IPForte. 48-hour turnaround. 130+ countries via Madrid Protocol. First call is free, no commitment.