₹0 to ₹25L: The Founder Finance Mistake That's Cheapest to Fix Now

I have sat with enough early founders now to recognise the pattern before they finish describing it. Someone has built a Shopify store, listed on Amazon or Flipkart, and revenue is trickling in, ₹2 lakh one month, ₹4 lakh the next. They are answering customer DMs, packing orders themselves some evenings, and paying vendors from whichever account has money in it that week. When I ask a simple question, how much did you actually sell last month versus what hit your bank, the answer is almost always a pause followed by "roughly the same, I think." That gap between "roughly" and "exactly" is where the whole finance problem for a D2C brand begins, and it begins far earlier than most founders expect.

The uncomfortable truth is that this works, for a while. At low order volumes a founder can genuinely remember why ₹40,000 left the account, a vendor advance, a Meta ad top-up, a personal Swiggy order paid from the business card by mistake. The number of channels is small, the number of SKUs is small, and errors stay invisible because nobody is looking closely enough to find them. What breaks first is not a system, because there usually is no system yet. What breaks first is memory-based finance itself, the founder's ability to reconstruct the story behind every transaction after the fact.

I have reviewed enough founder-run books to say this plainly: the businesses that struggle most at ₹1 crore are almost always the ones that never drew a hard line between the founder's wallet and the company's bank account at ₹5 lakh. It starts innocently. The business card gets used for a personal Uber ride because it is the card in hand. A friend invests ₹50,000 informally and it lands in the founder's personal account, then gets moved into the business a week later as an unlabelled transfer. None of this feels dangerous in the moment. It becomes dangerous the day someone, a CA doing your annual filing, a bank underwriting a loan, an investor doing diligence, has to answer the question "what did this business actually earn," and the honest answer requires untangling a year of mixed transactions from memory.

The fix at this stage is not a finance system, it is a single decision: open a separate current account for the business on day one, route every rupee of sales and every rupee of expense through it, and treat any personal draw as a formal, labelled transaction rather than an invisible one. It costs nothing and it is the single highest-leverage habit available to a founder below ₹25 lakh.

This is the mistake I would flag first to any founder reading this before they list their first product on Amazon or Flipkart. Most people assume, reasonably, that the familiar ₹20 lakh services threshold or ₹40 lakh goods threshold buys them time before GST registration becomes mandatory. That assumption is correct if you are selling only through your own website with no e-commerce operator involved. It stops being correct the moment you sell through a marketplace that is required to collect Tax Collected at Source under Section 52 of the CGST Act. The Act's registration provisions expressly extend to persons making supplies through such an operator, threshold or no threshold.

So a founder who lists a ₹500 product on Flipkart in month one, having sold nothing before that, is technically required to be GST registered from that first sale, not from the day cumulative turnover crosses ₹20 or ₹40 lakh. I have seen this catch out founders who registered "when the business felt real enough," usually around ₹15-20 lakh, and then had to go back and sort out invoicing and TCS reconciliation for months of sales that were never GST-compliant in the first place. The safer default, if a marketplace listing is anywhere in your plan for the year, is to register for GST before your first sale on that channel, not after.

Here is the scenario I have walked through with more than one early founder, and it always lands the same way. A brand sells ₹1 lakh worth of product on a marketplace in a week. The marketplace deducts its commission, a logistics fee, a small ad spend recovery, 0.1% income tax TDS under what is now Section 393 of the Income-tax Act, 2025, and 0.5-1% GST TCS under Section 52, and settles roughly ₹78,000 into the bank a week later. If the bookkeeper, or the founder acting as bookkeeper, records ₹78,000 as "sales" for that period, three things go wrong at once. The P&L understates revenue by ₹22,000. The GST return, if it is even being reconciled against the bank rather than the actual invoices raised, risks understating taxable turnover. And the TDS and TCS that were deducted at source, real money that belongs to the business as a tax credit, simply vanish from view because nobody recorded that they were ever taken out.

I would treat this as the single most expensive habit a founder can carry forward from ₹5 lakh in revenue to ₹50 lakh, because by the time volumes are high enough to notice the gap, the discipline of recording gross sales separately from net settlement has to be rebuilt from scratch, across months of transaction history. It is far cheaper to build the habit now: download the marketplace's order-level and settlement report every week, however tedious it feels at twenty orders a week, and record gross sales, fees and tax deductions as separate line items from day one.

I am wary of telling a founder at ₹8 lakh in revenue to go buy accounting software or hire a finance manager, because at this stage that is usually the wrong kind of spend. What I would recommend instead is closer to a discipline than a system. Separate the bank accounts. Register for GST correctly, accounting for the e-commerce operator rule above rather than the general threshold. Every time a marketplace or payment gateway settles money, save the underlying order and fee report before it ages out of the seller portal, because most platforms do not keep this data available indefinitely and reconstructing it later from memory is close to impossible.

And then, honestly, the highest-value habit of all: sit with your CA or bookkeeper once a month, even for thirty minutes, and ask them to walk you through what actually happened to the money, not just whether the GST return got filed. Compliance-only conversations are useful for staying out of trouble. They rarely tell you whether the business is actually making money, and at this stage, when everything still fits in one founder's head, that thirty-minute habit is what keeps it fitting for a little longer, and what makes the next stage, when the marketplace settlements start arriving in bulk and impossible to eyeball, less of a shock.