When we opened the beta of our local professional networking platform, Treatings, in New York City last summer, we didn’t have a specific timeline for when to open it more broadly. We wanted to run experiments to determine the best framework for facilitating one-on-one meetups over coffee.

We figured NYC would make for a nice laboratory, although there wasn’t much premeditation to starting the company in New York. My co-founder Paul and I had quit our corporate jobs and built bunk beds in the room we shared, so we had established roots in NYC.

One problem with keeping our beta restricted until finding the best framework is that “best” is nebulous. I’m not sure we could ever say we found the best way to connect people, just better or worse than previous iterations. In truth, we’ve found comfort operating beneath the “NYC beta” umbrella. We’ve deflected questions about our user growth by pointing out that “Well, it’s a closed beta. We’ve wanted to iron out the product before shifting focus to user growth.”

Constraining our beta to New York has been like slapping a governor on a car. We’ve implemented this speed limiter to create the perfect environment for controlled testing and tweaking of our platform. We actually haven’t wanted runaway growth!

Well, maybe we wouldn’t mind runaway growth.

The reality is that many of the experiments we’ve been running require a certain amount of liquidity to prove meaningful results. We’ve facilitated hundreds of coffee meetings in New York, but that needs to be thousands. We’ve embraced Paul Graham’s notion of Do Things that Don’t Scale, signing strangers up in coffee shops, manually facilitating connections and taking our users out to coffee in order to get feedback. We’ve had engagement, but need to prove that it’s sustainable without our heavy hand.

When I’ve asked entrepreneurs who have experience building and scaling marketplaces when to expand to other markets, I’ve heard conflicting views. In one conversation, I was nodding my head as the entrepreneur explained that “You should never grow to another market until you’ve built one self-sustaining community.” No sooner am I embracing the focus-on-one-market strategy that I hear “What’s the barrier to you rolling out in SF? Are you actually partnering with coffee shops to set up the meetings? Doesn’t feel like you need to do a local market rollout go-to-market like an operationally-intensive business like TaskRabbit has to do.”

Since we’ve accumulated signups outside of New York City, I’ve also heard contradictory views from users on whether restricting our beta for seven months was a good or bad idea.

You can guess what camp this person was in, emailing me that “To be told I cannot use your service because you’re currently in NY only is a lost opportunity when I could still be communicating with people of interest… That’s just limiting your business.”

I appreciated that feedback, but had to reconcile that with notes like this one: “Just go slow and nail it, then get the cash and expand. NYC is a great place to start.”

New York has been a good environment to experiment with Treatings because of the breadth of industries represented here. Before we started coding, we spent time approaching strangers in our target market, asking how they meet professionals outside their network. The unsurprising answer was that people didn’t have a problem leveraging their network to meet people in their industry. But, if looking to learn about new skills, find collaborators or explore career opportunities outside their industry, their preexisting network was less helpful. For that reason, we decided that Treatings would be industry agnostic from day one.

Where better to build a platform that requires representation of many industries than NYC, a city that has an investment bank, two fashion brands, three startups and four frozen yogurt shops on the average block. If we had started Treatings in San Francisco, where people are predisposed to being early adopters of technologies, we would have run the risk of building a framework non-transferrable to other cities. We’ve found our largest barriers to be social-based, as opposed to technological, as we seek to create an environment where people feel comfortable proposing meetings with strangers. While I don’t have empirical data to support this, I think we would have found the social barriers to be lower, and therefore not representative of the average city, in San Francisco.

Don’t get me wrong, New York has its own nuances. There’s a crazy person peeing on my leg as I type.

What began as seeming like a decrease in work, limiting ourselves to one city, turned into a headache. It’s taken time to maintain the geo-fences that keep everyone but residents of NYC out of Treatings. Our focus on improving our NYC beta has been to the detriment of the user experience for users outside NYC. They have been unceremoniously signed out after creating a user profile.

So, with all this in mind, we’ve decided to open Treatings to everyone. I’m hoping that users who previously signed up from outside New York, for whom we didn’t design a proper experience the first go-around are willing to give Treatings a second try as we continue to mold our product. Our product is far from complete. We haven’t found that elusive “best” framework to connect people, but we have learned a lot.

We’ve learned that people need to see activity on the site to feel comfortable engaging. We’ve learned that we can’t make users do work up front before providing any value (This is one of many things we were warned about but had to learn first-hand.) We’ve learned that underlying our innumerable design flaws, people still appreciate a good conversation with a side of caffeine.

With the benefit of hindsight, we’ll be able to look back on our NYC beta and judge whether we kept it restricted for too short or too long. For the record, Paul has been in favor of opening everywhere sooner rather than later, while I’ve been more reserved. I’m just looking forward to the healthy process of one of us hand-delivering “I told you so’s” to the other.

[Image via Wikipedia.]