How to Actually Shut Down a Startup in India

I've written before about the emotional side of shutting a company down, the part where you stop pretending an ending is a personal verdict on your character. This is the other half. The tactical half. The part nobody writes about because it isn't dramatic, it's just a sequence of decisions and filings that have to happen in something close to the right order, or you spend the next two years cleaning up a mess that a slightly different sequence would have avoided.

I wound down Perfume Lounge after growing it to nearly ₹55 lakh in monthly revenue. Here is the actual order of operations, not the sanitised version.

Step one: decide, then don't announce yet

The decision to stop is made in a single afternoon, once the numbers are honest and the alternative is clearly worse than stopping. But the day you decide is not the day you announce, and conflating the two is the first mistake founders make.

Between deciding and announcing, you need a plan for the sequence below. Announcing before you have that plan means every subsequent conversation happens under pressure, with people who now know the company is ending and are making their own decisions faster than you're ready for.

Step two: people, before anyone else finds out

Your team should hear it from you, individually, before they hear it from anyone else, including each other in a group setting. This is the least efficient way to deliver the news and the only acceptable one.

Have clarity, before these conversations, on what you can actually pay: final salaries, notice period pay if applicable, any accrued leave encashment required under the relevant state's Shops and Establishments Act, and provident fund settlement if you're registered under EPFO. Indian labour law doesn't have a single unified "startup shutdown" statute, and the specific obligations depend on your entity type, employee count, and state, which means this is the one section where a labour lawyer's two hours of time is worth more than any generic article, including this one.

Be honest about timeline. If you can pay everyone immediately, say so. If it will take sixty days because you're recovering receivables, say that too, and then keep the specific promise you made, because a founder's word during a shutdown is the only currency that survives the company.

Step three: vendors, in the order they extended you trust

Vendors who gave you credit terms did so on a bet about your company's future. That future changed. They deserve to hear it directly, not discover it when a payment doesn't arrive on schedule.

Prioritise by who extended the most trust relative to their own size, not by who's shouting loudest. A small vendor who gave you sixty-day terms when they didn't have to took a real risk on you, and settling with them in good faith, even partially, matters more than placating a large vendor who can absorb the loss without noticing.

Where you genuinely cannot pay in full, negotiate directly and honestly rather than going silent. Going silent is the single most common founder mistake at this stage, and it converts a solvable relationship problem into an adversarial legal one.

Step four: customers with open orders or warranties

Any customer with an order in flight, a subscription that's been paid for, or a warranty claim outstanding needs a clear resolution path before you start the compliance shutdown below. This is partly ethical and partly practical: unresolved customer obligations are exactly the kind of thing that surfaces as a consumer complaint months later, at the Consumer Disputes Redressal Commission, well after you assumed the matter was closed.

Step five: the compliance sequence

This is the part that has an actual, filing-by-filing order, and getting the sequence wrong costs months.

GST cancellation. File for cancellation of your GST registration once you've stopped making taxable supplies, and file your final GSTR-10 return within the prescribed window after cancellation. Missing this deadline attracts a penalty even though the business has already stopped operating, which is a genuinely common and avoidable mistake.

TDS and income tax closure. Ensure all TDS returns are filed and TDS certificates issued for the final period. Your company's income tax return for the year of closure still needs to be filed even though the company no longer operates, covering the period up to cessation.

ROC filing, if you're a private limited company. This is where most shutdowns actually stall. You have two realistic paths: Fast Track Exit under the Companies Act, formally called striking off the company's name via Form STK-2, which works if the company has no assets, no liabilities, and hasn't been operational for a defined period. Or formal winding up, which is required if there are outstanding liabilities, disputes, or assets that need to be liquidated and distributed. Most founders assume they qualify for the faster route and discover during the process that an unresolved vendor dispute or an unsettled loan disqualifies them, which is exactly why steps two and three above need to happen before this step, not after.

Bank account closure. Close accounts only after every payment above has cleared and reconciled. Closing early because it feels like the final symbolic act, and then discovering a pending settlement needs that account, is a specific and common unforced error.

Statutory registrations beyond GST. Professional tax, ESIC if applicable, Shops and Establishment registration, and any state-specific licenses relevant to your category all need their own closure filings. Each one is small. Missing several of them compounds into a genuinely large cleanup later, usually discovered when you try to start your next company and an old registration is still technically active and accruing penalties.

Step six: the founder's own paperwork

If you were a personal guarantor on any loan or lease, confirm in writing that the guarantee is released once the underlying obligation is settled. If you held a company credit card, cancel it and confirm final settlement in writing. If you have data, customer or financial, that carries retention obligations under applicable law, plan for its secure storage or destruction on the correct timeline rather than deleting it the day the company closes or keeping it indefinitely out of inertia.

The timeline you should actually expect

The decision takes an afternoon. Steps two through four, if handled directly and honestly, take four to eight weeks. Step five, the compliance sequence, realistically takes four to twelve months depending on which exit route you qualify for and how clean your filings have been historically. Founders consistently underestimate this timeline, expecting a shutdown to be finished within the quarter they decided to close, and the mismatch between that expectation and reality is where most of the frustration in this process actually lives.

The decision to stop is made in a day. The stopping itself is a project, with a plan, a sequence, and a timeline, and treating it as anything less than that is what turns a difficult but manageable ending into a genuinely painful one.

What this sequence actually protects

Getting this order right doesn't make the shutdown less real or less final. What it protects is your ability to start again. Founders who handle a shutdown badly, who go silent on vendors, who leave compliance filings incomplete, who let employee settlements drag past their promises, carry that reputation into their next venture in a market as relationship-driven as India's. Founders who handle it in the sequence above, however painful the underlying decision was, are the ones other founders, investors, and vendors are willing to work with again.

The ending is not optional once you've made the decision. How you sequence it entirely is.