Walking Away

When I mentioned that Bonafide was shutting down, some people were surprised.  They were surprised that we’d disband a strong and cohesive five-person team, that we’d shut down a startup with enough cash to survive another year, that we’d throw away 2.5 years of work from great engineers, that we’d throw in the towel just a few months after getting started on a new and promising product.

Yet on a warm Monday morning, my co-founder Brian and I sat under the bright sun outside Philz in downtown Palo Alto and we agreed to shut down Bonafide, the company we’d been running together since late 2013.  We had a lot going for us — capital from great investors, a talented and devoted team, a helpful set of early adopters — but we failed to build a business and we knew that closing Bonafide was our least bad option.

Everyone asks what I learned from that journey. I’m still waiting for divine inspiration to strike me with a bulleted list of 14 amazing insights (#7 being especially shocking). Instead, I’ve tried to answer three questions:

Question 1: We had a lot of things going for us at Bonafide. Why shut down and not change paths or try to sell the company?

In December 2015, two years into our startup, we started building a shared address book for corporations. This new product, we hoped, would be useful for sales people eager to find the right contact or introduction through a colleague. We knew which features we needed to build, meaning our product strategy was focused for the first time in many months. I framed our situation to our team and our investors as Season 3 of Bonafide’s journey: a new year and a new set of opportunities.

After several months of work on the new product, we demoed it at a board meeting.  Our lead investor loved it and excitedly congratulated us for our great product design and impressive technology. I should have been thrilled: we’d gotten over a rough patch and seemed to be on the path to something great.

Instead, I felt a surprising, hollow indifference. This felt like the beginning of a long hike towards a destination I didn’t want to reach. Even with four terrific colleagues by my side, I couldn’t get myself psyched about building an enterprise product for sales teams.  I wasn’t able to step back and say “this product really needs to exist, and I really need to be the person building it.” Brian’s feelings were similar, and that meant neither of us was doing top quality work (though Brian’s was much better than mine).

In a series of conversations with investors, Brian and I confided that we weren’t terribly excited about the problem we were solving.  I made it clear that I didn’t think I was the right CEO for the business we were trying to build. That wasn’t setting us up for success, and we discussed four possible ways to keep things going:

  1. Keep going as is.  We suck it up and push things forward — at least to a stage where we could be acquired for a meaningful amount of money.
  2. Switch our product direction.  With a year of capital in the bank, we could build something better suited to the team’s skills and interests.
  3. Find a new CEO.  We find someone who’d be better leading this business than I would be.
  4. Put ourselves up for sale.  With a solid team and some impressive technology, we could put ourselves out there to larger companies who might want to acquire us.

We talked through each of those options.

Keep going as is?
Continuing as is would mean continuing something that wasn’t working, and unlikely to position us for success. There are plenty of entrepreneurs who can thrive while working on something they aren’t interested in, as long as they have a clear goal to go after. Brian and I are nerds and want to work on problems we find interesting and important.

Switch our product direction?
If there had been something we were dying to build, I could have seen trying something new with the team we had. But Brian and I were both pretty drained of both energy and ideas after 2.5 years, and the idea of starting over had very little appeal.

Find a new CEO?
If our product had been mature and ready to be scaled, we might have tried to find a new CEO. As it was, we had three meaningful trial customers, none using the product heavily. We were still figuring out what to build, meaning we’d need a scrappy founder CEO and not a professional growth CEO.

Put ourselves up for sale?
Our five person team was solid and it’s possible we could have found an acquirer. Selling the company would return a bit more money to our investors, it would allow us to claim our startup was a successful one (Bonafide was acquired by ____! Woo-hoo, we won!), and it might lead to a slightly better financial outcome for our employees.

However, an acquisition was uncertain and (if successful) would take time (so employees might leave, while we’d spend more of our capital), so we didn’t feel the ends would justify the means:

  • Investors were largely indifferent to an acquisition: even if we sold the company, they would likely lose money because their success is largely driven by one or two 100x returns, not by getting $0.75 on the dollar instead of $0.25.
  • Employees might have received marginally better equity packages, but the differences would likely be small, and they would entail being locked into an employer chosen by Brian and me.
  • The financial rewards might have been stronger for Brian and me as founders — but we have both been through acquisitions and are wary of the downsides of golden handcuffs. Those financial rewards would have to justify the opportunity cost of not doing something else, and in the event of a small acquisition we felt that was unlikely.
  • Brian and I would both rather admit defeat than lay false claim to victory, so we didn’t attach much value to checking the box that said our company was acquired.

That left a fifth option: shut down the company and the website, help the team find jobs, and return the modest amount of capital we still had in our bank account.

Though not the option we’d hoped for, shutting everything down felt right.  It was a strong decision: we would cut losses, and get on with our next things.

Question 2: The shared address book product held little appeal for our team — but in December 2015 it seemed like it would be better to build it than to continue what we were doing. How did we get to such a desperate position?

To answer this question, I go back about nine months to early 2015.  At that time, we had decided to focus our team on a product called Free Agents. A member would post about a Free Agent — someone they knew (member or not) who was starting to look for a new job. Other members could ask the poster for an introduction to the candidate. The dynamics were great: members’ recommendations were seen as quite valuable, the member was helping a friend/former colleague they held in high regard, hiring managers saw impressive candidates in their inboxes, and Free Agents were getting interest from great companies and hiring managers.

Things looked promising: after we walked through the metrics on Free Agents at a board meeting, our lead investor encouraged us to start preparing a list of VCs who would lead a monster Series A.  I was feeling confident.

There was just one more thing I wanted to show: that we could generate a significant number of Free Agent postings.  Unfortunately, this wound up being the downfall: the key actor (the connector) lacked motivation to post about Free Agents, and our attempts to encourage them were unsuccessful.

Four or five months working on Free Agents with no improvement on metrics put me into a bit of a tailspin. I felt like I was flailing as a leader and I didn’t really know what to do next. And not knowing what to do next is a very dangerous position to be in as a founder/CEO.

As an investor/adviser/friend to dozens of startups, I’ve observed the behavior of many founders. Only a select few are able to combine both a clear sense of purpose with humility: they know where they want to go but are wise in asking others to help get there. Some founders fail because they’re too stubborn to receive and process feedback; others lack a confident vision and wind up being pushed around like a kite in the wind.

In the summer of 2015, a year and a half in to Bonafide, with our Free Agent product not taking off, I’d become kite-like. I didn’t know which way to go and I hoped that a gust of wind would come along and swiftly point us in the right direction.

We moved offices, a change I hoped would magically lead to a reset and yield a surge in energy and creativity. We brought in several people to help us think through problems, and I held out the unrealistic hope that one would be our savior.

Meanwhile, we were working through several very complex projects that would never see the light of day. I couldn’t pinpoint one good thing to build, and so we built everything. Several people came to the table with interesting ideas, and rather than picking one we combined them into a smorgasbord of features that didn’t belong together.

All of that changed at a board meeting in December, when our lead investor rightly called us out for working toward a product that was far too complicated and unlikely to succeed.

Instead, he suggested, why not build a shared address book for teams? Every day, sales people at his portfolio companies were trying to figure out who he knew at Ebay, at Target, at Apple — either to initiate a conversation or to close a deal. It would be extraordinarily valuable to map out all of the people your colleagues know — by company, by job title and skill, by name.

The product he suggested was many times simpler than the other products we’d built and diagrammed. There was a simple privacy model — everyone on a team could see everything shared by their colleagues — and it took advantage of a lot of our intellectual property.

On the other hand, it sounded like a product for sales people — a departure from what we’d been doing, and not an area of interest or expertise for anyone on our team. Had I been presented with the idea two years prior, I would have quickly declined: a sales tool was not something I could envision spending years of my life on.

But that December day, the team and I were flailing, and the grass on the other side of the fence looked more attractive than the messy pile of junk on our side. Eager to leave the junk pile, we quickly climbed the fence and we jumped.

Question 3: What went wrong in the lead up to Free Agents?  Could we have done things early on that would have improved our likelihood of success?

When we started the company, we had a compelling vision and were working on some really important and interesting problems. We had a strong team and Brian and I were experienced entrepreneurs.  That meant we were able to raise a substantial seed round with a fairly vague idea of what we we’d be building.

Having that money in the bank seemed great: we wouldn’t have to worry about fundraising for a long time.  However, it meant we had little urgency to get to the next stage.  A lack of urgency and a lack of focus isn’t a great combination, and in our case it led to us building a large number of experimental features:

  • several iterations on best-of lists — best engineers I’ve worked with, best Series A VCs, top visual designers
  • a feature that parsed introduction emails you’d written to pull out quotes about people you were introducing (“John is the best engineer I’ve ever worked with”)
  • global lists of top people in various fields, largely using signals from Twitter
  • a feature to recommend someone looking for a new job (a “Free Agent”) to other Bonafide members and manage the process of introducing them to hiring managers
  • a dashboard of your standing as a connector
  • a profile with the nice things people had said about you in email introductions over the years
  • a feature that let you tell others what kind of people you were looking to find
  • mobile apps and a Chrome extension to let you write introductions more quickly
  • a way to recommend service providers to others with whom you share an affinity (e.g., lawyers to companies with the same investors)

That long list of features should tell you two things: first, our engineering team was talented and moved quickly; second, we were all over the place, and as CEO I was doing a poor job focusing the team’s efforts.

By the time we decided to focus on the Free Agent product, we’d been working on Bonafide for well over a year.  That meant our team (myself included) had less patience to get things right with Free Agents.  When our early attempts to make Free Agents weren’t fruitful, we got frustrated: it felt like we’d been at this for nearly two years without a lot of progress.  Instead of stubbornly persisting, we got impatient and flailed.

Had we focused earlier on, we could have attacked Free Agents (or another problem) more patiently and persistently.  I wish we had.

Parting Thoughts

When we first officially became employees of the company, each of us signed an offer of employment.  The offer letter, drafted by our lawyers, referenced the “company employee handbook,” a document which (as far as I knew) didn’t exist.  When my colleague asked me about it, I figured I should put together a very formal document (our company started as Laserlike).

That “handbook” followed us across four offices and now sits in my attic office.

I leave with fond memories and strong relationships, and I’m proud that my colleagues and our investors maintained a mature “treat people like adults” approach throughout.

I learned quite a bit in the past 2.5 years — though maybe not as much as a child does — and I’d like to think my colleagues did too.

I certainly made some mistakes — I’m not sure what qualifies as too stupid — but I’m really glad I got to work with some amazingly talented people who screwed up a lot less than I did.

This is my first time shutting down a company.  It’s a painful experience to go through, but in our case it was the right thing to do.  I walk away humbled, but also grateful for a terrific team and the chance to build something meaningful.

Mike Greenfield founded Bonafide, Circle of Moms, and Team Rankings, led LinkedIn's analytics team, and built much of PayPal's early fraud detection technology. Ping him at [first_name] at mikegreenfield.com.