Finding Your First 10 Customers, with Pulkit Jain of Vedantu

Overview

Two parts: a short deck on the current GTM and distribution landscape (how AI search and platforms like Reddit and YouTube are reshaping acquisition), followed by a fireside chat with Pulkit Jain, Vedantu co-founder, on how he found Vedantu's first customers, and what still applies to founders starting today.

Speaker: Pulkit Jain, Co-Founder, Vedantu
Event: The GTM School, a joint initiative by AI Learn Circle and Women in Product Marketing, running August to October 2026
Venue: Vedantu HQ, Bengaluru

---

Part 1: the current distribution landscape

Three things replaced old-style SEO

• About 16% versus about 1.8%, fewer visitors, far higher intent. Traffic arriving from ChatGPT converts roughly 9x better than organic search. Someone who already read the answer and still clicks is a fundamentally different, higher-intent visitor. You need far less traffic than before.
• 3x: get mentioned, not linked. Brand mentions now predict AI-answer visibility roughly 3x more strongly than backlinks do. Being talked about beats being linked to. This is now a PR and community problem, not just an SEO one.
• Reddit and YouTube: be where the models actually read. Reddit accounts for over half of all social citations in AI answers; YouTube is about a fifth and growing fastest. For Vedantu specifically, YouTube was already their biggest owned asset, so it is now doing double duty.
• The underlying warning: every channel between you and your customer can change its mind, the way Google did. The fix is to own the relationship outright (email list, WhatsApp group, an app on their phone). Nobody can switch that off.

Where to actually go

• The channel is chosen by your audience, not by you. Devs and B2B go to X, GitHub, Discord, Reddit. Indian consumers go to Instagram and YouTube. Parents go to WhatsApp groups and schools. Enterprise goes to the one conference they all attend.
• Organic wins, and it's slow. Vedantu's own SEO took years; Zerodha's education engine, Varsity, took years too. Start it in week one anyway. Compounding only begins once you begin.
• Founder brand helps. It isn't a prerequisite. Nithin Kamath built a public founder brand; Zoho scaled for years without Sridhar Vembu being a public figure. You don't need 50,000 followers before you can sell anything.
• Master paid, or it will kill you. Done badly, CAC goes astronomical and quietly eats the business. Done well, it's a controllable lever. Know your CAC and payback period before you scale spend.

Rented versus owned distribution

Rented distribution gets you your first customers. Owned distribution gets you a company.

---

Part 2: finding your first 10 customers

Sell before you build

An early example he walked through: he didn't build the product first. For a few months, the only "product" was videos, slides, and pitching directly to people. He pre-sold 5 units on that basis, then built. Sold 20 more, watched engagement, and because engagement kept increasing, ordered more stock. The sequence was sell, observe, build, repeat, not build then sell.

The hard part isn't building

Talking to people, making the cold pitch, that's the ugly, exhausting part, not the easy part. Building is comparatively the easy half.

A crowded market isn't necessarily a saturated one

A product already existing in the market doesn't mean the market is fully penetrated. It may be badly under-served, in which case your product will still work even with direct competition already present.

Brand is trust, and credentials substitute for it early on

Brand, at its core, is about building trust. Before you have a brand, your credentials do that job instead:

• Domain experience
• A track record of execution
• Passion and mission
• Proximity to the problem, using whatever access is genuinely available (your own network, existing relationships). "Good enough" credentials beat waiting for perfect ones.

Founders become good storytellers eventually. This can be learned, not just a starting trait (Nithin Kamath's podcast was cited as an example of the craft).

Belief matters more than the idea

An entrepreneur is someone who solves a problem. To keep going, you need a real, non-fickle belief system, not FOMO-driven idea-hopping, that the specific problem you've picked can actually be solved by what you're building.

Ask the right question about who to sell to

Don't ask "who needs this most." Ask "who can say yes without breaking something?" Chase the buyer with the least friction to a yes, not the buyer with the greatest theoretical need.

Finding the first 10 is a proximity problem

• At the first-10-customers stage, it's fundamentally a proximity problem. You don't need a marketing or sales function yet, and it doesn't need to scale.
• Trust your own gut over other people's opinions.
• Use whatever credential is available to you: proximity, your existing network, anything that gets you in the door.
• Unscalable is correct at 10 customers, fatal at 1,000. What legitimately works to land your first 10 is supposed to be unscalable. That's fine, and expected. It only becomes a problem once you're trying to repeat it a thousand times over.

Own your distribution, or CAC will crush you

• SEO is declining fast because search itself is shifting from SEO to AEO and GEO (answer-engine and generative-engine optimisation): being cited inside an AI-generated answer rather than ranked in a link list.
• You have to own the distribution channel (SEO, YouTube, community, a free product, email, WhatsApp, Reddit), otherwise customer acquisition cost will eventually crush the business.
• Reddit specifically is the fast-growing, newly "impactful" channel right now, because AI systems cite it heavily; Twitter/X and Discord matter too.
• Content volume into Meta platforms has exploded over the last year and is still increasing.
• Vedantu's own channel arc over time: radio, then SEO, then free live YouTube classes, then paid digital, used for long-term customer value rather than one-off acquisition.
• Products come and go. Vedantu has killed many along the way. The funnel and the owned distribution are the real asset, not any single product built on top of them.

Building philosophy

• Whatever you're building today, the first idea will definitely change, but you can afford to iterate fast if you already have distribution.
• Distribution matters more than everything else.
• Simple things work, and execution is usually fine, but "simple" has to be simple in the customer's head, not just in yours.
• Close the promise-to-delivery gap to zero.
• In consumer businesses, people have simple needs. Solve those needs, don't over-engineer.
• You're really competing on one question: "will this replace what they already do?" It's difficult to displace anyone from a workflow they're already set on. Your product is competing against their current habit, and you have to make their life measurably easier to win that switch.
• Own the relationship outright: an email list, a WhatsApp group, an app on their phone. Nobody can switch that off (this echoes the owned-distribution framing from Part 1).
• For consumer products: understand the customer, do right by them, while also actively catching attention and building engagement. Both matter, not just one.

Closing advice: trust your own gut

Ultimately, trust your gut. A lot of mental and emotional calculation happens in the background and simply surfaces as a gut feeling. Listen to it and act on it. Don't get swayed by the knowledge of other people's success stories; it's your own gut, your own life, your own thinking, so follow it. You can't know what will happen in the future, lessons only arrive in hindsight, so act on your own judgment now rather than waiting for certainty.

---

Key takeaways

• Sell before you build. Pre-sold 5 units on nothing but pitching and slides, then built; sold 20 more once engagement validated demand.
• Distribution matters more than the product idea. Owned channels (SEO, YouTube, community, email, WhatsApp, Reddit) compound; rented channels (ads) only buy a start.
• What works to get 10 customers is meant to be unscalable. It only turns fatal once you try to force it to 1,000.
• Search is shifting from SEO to AEO and GEO. Brand mentions now predict AI-answer visibility better than backlinks, and Reddit plus YouTube are the two channels AI systems cite most.
• Credibility before brand exists comes from domain experience, execution history, passion and mission, and raw proximity, not polish.
• Chase "who can say yes without friction," not "who needs this most." The real competition is against the customer's current habit or workflow, not other vendors.
• Trust your own gut over other founders' stories. Hindsight is the only place lessons come from; you have to act before you have certainty.