๐Ÿคฏ Snaphunt: 8M Users Couldn't Save a $2.8M Funding Gap

Singapore AI recruiter Snaphunt signed up 8M job seekers, then shut when a promised $2.8M came too late.

New here? Join us to become the top 1% founder by learning from startup failures:

P.S. Syfe wanted a room full of its exact customers, so we built one. 177 founders applied, we turned 70 away, and 53 were in the room.

๐Ÿ‘‰๐Ÿป See the full Syfe case study

Hey Founders,

Welcome to The Runway Ventures โ€” a weekly newsletter where I deep dive into failed startup stories to help you become the top 1% founder by learning from their mistakes with actionable insights.

Today's story is about a Singapore AI recruiter that signed up 8 million job seekers but earned just US$109K in a quarter, and shut down when a promised $2.8M came too late. Let's get to it! ๐Ÿš€

Today at a Glance:

  • โ˜ ๏ธ 1 Failed Startup โ†’ Snaphunt

  • โš ๏ธ 2 Mistakes โ†’ Confused distribution with a business

  • ๐Ÿง  3 Lessons Learned โ†’ Committed money isn't cash until it clears

  • ๐Ÿ”— The Runway Insights โ†’ The guerrilla hiring playbook

  • ๐Ÿ’ฐ Southeast Asia Funding Radar โ†’ Nexstrom secures $12M (Seed) to develop equipment for manufacturing 2D semiconductor materials at scale for next-gen AI chips

โ˜ ๏ธ 1 Failed Startup: Snaphunt

The Snaphunt team

๐Ÿš€ The Rise of Snaphunt

๐Ÿ‡ธ๐Ÿ‡ฌ Founded by Tulika Tripathi in 2017, with technology partner Prakash Sanker helping bootstrap the early build, Snaphunt was a Singapore-based AI recruitment platform that wanted to replace expensive hiring agencies with an algorithm โ€” matching candidates to jobs without the bias, at a fraction of the agency fee, and without the human recruiter in the middle.

๐Ÿ•บ๐Ÿป Foundersโ€™ Story

Tulika had spent nearly 15 years in traditional recruitment โ€” Michael Page across Switzerland, India, and Singapore, then Managing Director for Asia at Hudson Global.

She knew the industry inside out. And she saw the same problem everywhere: companies spending too much time and money to find talent, and still often hiring the wrong fit.

๐Ÿ’ผ So in 2017, she walked away from the corporate job, pooled her personal savings with technology partner Prakash Sanker, and started building what she thought recruitment should look like โ†’ fast, data-driven, borderless, and without the recruiter in the middle.

  • The Problem โ€” ๐Ÿ˜ฉ Traditional recruitment was expensive, manual, and brutal for everyone involved.

    • Agency fees put a good recruiter out of reach for most startups and SMEs.

    • Human bias crept into sourcing and screening.

    • Candidates sent applications into a black hole and heard nothing back.

  • The Solution โ€” ๐Ÿค– Snaphunt built an AI-powered Applicant Tracking System and talent-matching engine that automated the whole hiring funnel.

    • Algorithmic matching on skills, experience, notice period, and salary expectations.

    • Snapsych psychometric scoring, plus one-way video interviews.

    • Martha โ€” an AI voice agent that handled reference calls and spat out a structured report.

๐Ÿ‡ธ๐Ÿ‡ฌ๐Ÿค– In short, Snaphunt was a recruitment agency without the recruiter โ€” algorithmic, borderless, and pitched as 90% cheaper than the human kind.

Tulika Tripathi (Founder & CEO)

๐Ÿ’ฐ In February 2019, Snaphunt went live in Singapore and raised a US$1 million seed led by BEENEXT. More than 500 companies had already signed up, including Microsoft, Antler, and Safilo.

By early 2020, Snaphunt was operating in 10 Asian markets. Then COVID hit, remote hiring exploded, and the company rode the wave.

In January 2021, Snaphunt announced another US$1 million from BEENEXT.

In September 2021, with 3,500+ employers on the platform, Tulika opened a Series A fundraise.

๐Ÿ”๏ธ At its peak, Snaphunt:

  • had served 15,000+ employers

  • had 8 million+ registered professionals on the platform

  • connected talent to jobs in 140+ countries

Its website also claimed 50,000+ successful hires.

On paper, it looked unstoppable. 8 million users. 140 countries. A clean AI story in a post-ChatGPT world.

But the revenue tells a different storyโ€ฆ

๐Ÿ“‰ The Fall of Snaphunt

In Julyโ€“September 2025, Snaphunt had 8.5 million job seekers and 10,000 employers on the platform. It earned US$109,130 that quarter.

Xelpmoc, an Indian tech firm that owned about 8% of Snaphunt, reported those numbers to its own investors.

๐Ÿ“Œ Hereโ€™s what happened to Snaphunt:

โ

We had secured investor commitments, but the funding was delayed beyond the point where the business could be sustained.

โ€” shared by Tulika Tripathi (Founder & CEO)

๐Ÿš€ From Hudson Global to a Singapore MVP

  • 27 Feb 2017 โ€” ๐Ÿข Snaphunt was incorporated in Singapore.

  • 2017 โ€” Tulika resigned as MD Asia at Hudson Global and bootstrapped the MVP with Prakash using personal savings.

  • Feb 2019 โ€” ๐Ÿ’ฐ Raised US$1M (Seed) led by BEENEXT as the platform went live in Singapore. 500+ companies had signed up, including Microsoft, Antler, and Safilo.

  • Jan 2020 โ€” ๐ŸŒ Operating in 10 Asian markets โ€” Singapore, Malaysia, Indonesia, India, Philippines, Bangladesh, Sri Lanka, Vietnam, Thailand, and Hong Kong โ€” with a team of 15, all hired through Snaphunt itself.

  • Jan 2021 โ€” ๐Ÿ’ฐ Announced another US$1M from BEENEXT.

  • 15 Sep 2021 โ€” ๐ŸŽฏ Tulika opened a Series A fundraise.

    • At that point: 3,500+ employers, ~1M candidates, jobs in 57 countries, ~30 employees.

โณ More products, thin revenue

  • 2022 โ€” ๐Ÿ› ๏ธ Launched Snaphyre, a fully managed hiring service. Snaphunt's pricing deck now offered 4 ways to pay: free credits, subscriptions from US$6,000/year, Flexpert at US$2,000/job, and Snaphyre at 20% of a hire's annual pay.

  • 2024 โ€” Removed its freemium plan. Paid customers grew 1.5x.

  • Julโ€“Sep 2025 โ€” ๐Ÿ“Š Revenue for the quarter: US$109,130, against 8.5 million job seekers on the platform (based on Xelpmoc's investor presentation).

  • 6 Jul 2026 โ€” โš ๏ธ Snaphunt's directors declared that the company "cannot by reason of its liabilities continue its business."

  • 7 Jul 2026 โ€” Tan Wei Cheong and Lim Loo Khoon were appointed provisional liquidators.

  • 30 Jul 2026 โ€” ๐Ÿšซ Members passed the resolution placing Snaphunt into Creditors' Voluntary Winding Up, with the same 2 liquidators from Deloitte.

  • 17 Aug 2026 โ€” ๐Ÿ’” The Straits Times reported that a planned US$2.8M (S$3.6M) round had been delayed. When operations stopped in early July, only 3 employees in Singapore remained, including Tulika herself.

A committed round isn't cash. It's a promise with a settlement date, and if that date slips past your runway, there's no second chance.

๐Ÿ™๐Ÿป Snaphunt had 8 million users in 140 countries, and none of it mattered when the money didn't land on time.

The delay was the trigger. But a company earning US$109K a quarter had little room for a delay in the first place.

Want to learn more about Snaphuntโ€™s downfall?

โš ๏ธ 2 Mistakes

Mistake 1: Confused distribution with a business

Snaphunt built an audience at scale. 8.5 million job seekers. 10,000 employers. The numbers looked incredible.

But in Julyโ€“September 2025, Snaphunt earned US$109,130. That's about US$11 per employer for the whole quarter.

๐Ÿค” Build a big enough funnel and monetisation will take care of itself. It usually doesn't.

Snaphunt only removed its freemium plan in 2024, 7 years after it was incorporated.

๐Ÿ’ฐ Scale without pricing power is just a mailing list.

Mistake 2: Ran out of runway before the money landed

By Tulika's account, Snaphunt had secured investor commitments for a new round. Then the investors' money was delayed, and Snaphunt couldn't wait.

๐Ÿคฆ๐Ÿปโ€โ™‚๏ธ The round was committed. The cash wasn't in the bank yet.

Snaphunt had no buffer to absorb the slip. On 6 July 2026, the directors declared the company could not continue because of its liabilities.

๐Ÿ˜ญ A commitment is not cash.

๐Ÿง  3 Lessons Learned

Lesson 1: Don't scale users you can't monetise

Snaphunt went live in Singapore in 2019, giving it years to pressure-test a paid model. 6 years later, with 8.5 million job seekers on the platform, it earned US$109,130 in a quarter.

๐ŸŒฎ Key Takeaways:
  • Pressure-test willingness to pay in year 1. The longer you delay, the harder it is to charge users you trained to expect free.

  • If quarterly revenue is a rounding error against your user count, you have a mailing list on your hands.

๐Ÿ› ๏ธ Operator Playbook:
  • ๐Ÿงช Run a willingness-to-pay sprint

    • Before shipping new features or scaling marketing, interview 20-30 target buyers and ask them to commit to a specific price for a specific outcome.

    • Use the Van Westendorp Price Sensitivity Meter to find your acceptable price range.

    • If fewer than 30% commit at your target price, your value prop or your pricing is broken.

  • ๐Ÿ“Š Benchmark your ARPU against focused competitors

    • For HR tech / ATS SaaS, compare your pricing against focused players like Workable and Greenhouse.

    • If your ARPU is less than 10% of the industry's leading paid tier after 24 months, your free users are subsidising a broken model.

Lesson 2: Committed money isn't cash until it clears

Snaphunt had investor commitments for a US$2.8M round. The funding was delayed, Snaphunt ran out of cash first, and the company went into liquidation.

๐ŸŒฎ Key Takeaways:
  • A term sheet or a verbal yes is a promise with a settlement date. Count runway from cash in the bank, not from commitments.

  • Rounds slip for reasons you don't control: an investor's own fundraising, legal reviews, market shocks. Plan for the delay, not the closing date.

  • If one late wire can end the company, you need a plan B before you need it.

๐Ÿ› ๏ธ Operator Playbook:
  • ๐Ÿงพ Run a cleared-cash runway

    • Calculate runway using only money in the bank. Treat committed-but-unpaid money as a forecast.

    • Start your next raise with at least 12 months of runway, so a 3-6 month slip can't kill you.

    • Ask for a staged close: part of the round wired at signing, the rest on fixed dates.

  • ๐Ÿ›Ÿ Set your fallback before the round is at risk

    • Line up a bridge option early: a convertible note from existing investors, insider pro-rata, or venture debt.

    • Agree a trigger with your board in advance: "If the round hasn't closed by [date], we cut burn by [X]%."

    • Run the "Default Alive" test from Paul Graham's essay. If you're default dead, plan as if the round won't close on time.

Lesson 3: Depth in one market beats breadth across 10

By early 2020, Snaphunt was operating in 10 Asian markets with a team of 15. By late 2021, slightly less than half of its employers were Singapore companies. The home market was one of 10 long before it was ever a stronghold.

๐ŸŒฎ Key Takeaways:
  • 10 markets means 10 different go-to-market motions a small team cannot execute.

  • Dominate your home market until you're the default choice before expanding abroad. If your home base is less than 70% of customers, you don't have the dominance to export.

๐Ÿ› ๏ธ Operator Playbook:
  • ๐ŸŒ Treat each new country as a new company with its own P&L

    • Each market gets its own gross margin target, CAC payback target, and 12-month proof window.

    • Benchmark against regional SEA peers like Glints and JobStreet.

    • If a new market can't hit break-even contribution margin within 18 months, close it before it drains the core.

๐Ÿ”— The Runway Insights

  • The guerrilla hiring playbook (Read)

  • What TypeSafe got right with the Jev launch (Read)

  • What it's like to work at an AI-native company (Read)

  • There are no more moats. You never had one (Read)

  • The SaaStr AI guide to building a top-tier inbound AI agent: 17,000 conversations, ~600 meetings booked, and 60% more new business (Read)

๐Ÿ’ฐ Southeast Asia Funding Radar

  • Nexstrom secures $12M (Seed) to develop equipment for manufacturing 2D semiconductor materials at scale for next-gen AI chips (More)

  • TacnIQ.ai bags $1.5M (Pre-Seed) to build tactile AI foundation models, grow its engineering team, and scale commercial deployments (More)

  • WILDER raises $500K (Pre-Seed) to scale itโ€™s performance hydration across Southeast Asia (More)

  • iPiD raises $16M (Series A) to verify your payments before sending them (More)

๐Ÿ’ƒ Rate Todayโ€™s Edition

What'd you think of today's edition?

Your feedback helps me create better content for you!

Login or Subscribe to participate in polls.

Thatโ€™s all for today

Thanks for reading. I hope you enjoyed today's issue. More than that, I hope youโ€™ve learned some actionable tips to build and grow your business.

You can always write to me by simply replying to this newsletter and we can chat.

See you again next week.

- Admond

Disclaimer: The Runway Ventures content is for informational purposes only. Unless otherwise stated, any opinions expressed above belong solely to the author.

Reply

or to participate.