Our First Four Months Building Underdog.io

Joshua Goldstein
6 min readAug 27, 2014

Four months ago, we released underdog.io, a two-way curated marketplace for talent. Here’s how we got started and where we are today.

In the beginning

Underdog.io was never supposed to be our main gig. We were too embarrassed to bring it up when we met with a well-known NYC angel investor the morning we went live.

Instead, we were meeting to discuss Porter, the software tool that our team had been building for residential property managers.

Unfortunately, I use the word “building” lightly. Very little had been coded.

None of us had spent a day in real estate. So while we had a network that touched some of the biggest real estate players, we were spending much of our time trying to understand how things worked.

Every time we had to make a product decision, we’d have to ask for feedback. We had no instincts in the space. As if that wasn’t enough, the space was notoriously slow to sell to.

There’s no need to fear, underdog.io is here

Initially, we thought that underdog.io was a product that could provide relatively passive income while we battled the long real estate sales cycle. It grew from our first-hand experience interacting with external technology recruiters.

We had all been involved in hiring at our previous startups. We understood how difficult it was to find awesome, talented folk from the minute you turned on the “I’m hiring” sign. Traditional recruiting firms, while perfect for some companies, always felt too expensive at our early-stage companies.

The recruiter model also didn’t feel right on the candidate side of the marketplace. High-quality people tend to get inundated with unsolicited LinkedIn requests, emails, and sometimes even phone calls from recruiters and others. When it came time to actively look for a job, there was so much noise in the marketplace that it was difficult to see past a few trees in the forest.

Launch day

Looking back, what we did may seem disingenuous. We built and launched underdog.io in less than a week. We had 0 startups lined up. We had 0 candidates on board. What we had were a few hypotheses:

(1) If we offer a frictionless job application process, people will sign up.

(2) If we did a good job curating candidates, companies would pay us.

(3) If we charged a monthly subscription fee instead of a traditional recruiting fee, companies would appreciate that.

We went live with underdog.io on April 15, 2014 on Hacker News and Reddit.

Of the 2,288 unique visitors that day, 86 applied to be candidates.

Within the first seven days, we had 6,746 uniques. 495 people applied to be candidates.

Anecdotally, our first candidate was a highly qualified developer from the Bay Area. Out of our first five startup applications, three had raised over $5 million.

Better to be lucky than good

We got lucky at first. Two days after launch, Jordan Crook caught wind of underdog.io and wrote about us on TechCrunch without warning. It was awesome. And terrifying.

We didn’t have a CrunchBase profile, and the post was linking to the wrong company in some places. We had a funny exchange with Jordan when we realized what was going on.

https://twitter.com/jordanrcrook/status/456904020880154624

The next day, I posted underdog.io to Product Hunt. To give you an idea of how fast that community has grown, we had fewer than 50 votes and we were the top product of the day.

We had name-brand startups attempting to pay, but we hadn’t built payments. We quickly integrated with Stripe (which we’ve been using since). Within the first week, payments were coming in.

It was still early, but we knew underdog.io was something that required more attention. The initial validation was nice. The fact that we had experienced the pain of the customer was more important. We finally had instincts.

Time to build some product

Underdog.io is still very much an MVP. We built an internal tool that helps us tag, rank, and analyze candidates. It’s buggy as hell, but it does the job.

We’re working on candidate profiles. If you apply to underdog.io, you will soon have the option to update your work status and add more information. The initial form will still take 60 seconds to fill out.

Candidates are finding us from all over. We continue to spend money testing different job boards (or at least the ones that will have us) and ad campaigns, but many of the quality candidates are finding us through word of mouth.

Whenever a candidate applies and we’re all together, I yell, “Candidate, coming in!” It’s super annoying, but it’s good to remember that candidates are the lifeblood of this whole thing.

Hypothesis testing

Checking back on those hypotheses:

(1) If we offer a frictionless job form, people will sign up. SO FAR, YES

(2) If we did a good job curating candidates, companies would pay us. SO FAR, YES.

(3) If we charged a monthly subscription fee instead of a traditional recruiting fee, companies would appreciate that. NOT ALWAYS

We’ve had a number of companies ask to pay a placement fee. We added it as a second option, but we’re thinking about removing it. We still believe that typical placement fees are too expensive, so even though we’re only charging 10%, it still feels wrong. We’re in a spot where some customers are telling us one thing and we want to do another. Once we figure out how to reconcile the two, we’ll share.

We should mention that our pricing model will change. Most of our monthly customers have told us we’re too cheap. Until we make significant modifications, there’s an awesome opportunity for startups to apply.

Four months down, many more to go

Underdog.io has been our full-time focus for almost 4 months now. In that time, we’ve had thousands of candidates apply. Some are so high quality, it surprises even us.

We’re working with more than 50 of the top startups in NYC. We’ve turned away just as many companies that aren’t a fit for our candidates. We’re focused on curating both sides of the market. It’s not about volume right now.

And best of all, a bunch of our candidates have received offers.

It feels good to help candidates find jobs. It feels equally good to know that we’re saving startups on average $25,000–$30,000 per hire had they used a traditional recruiting agency.

We’re looking forward to continuing our underdog story.

If you enjoyed this article, we’d love for you to share it on Twitter.

If you’re interested in applying to underdog.io as a candidate or a startup, there’s that too!

Thanks to @marckohlbrugge, @ajwaxman, @GrantCov, @CyrusWired, and @GoldsteinZach for providing feedback on this post.

--

--

Joshua Goldstein

Co-founder of @underdogjobs. Past: @Bondsy. #ProPoker. @UNC. @EmoryUniversity.