Rafael Corrales @rafaelcorrales - Tumblr Blog

My very early (and ongoing) involvement with Instacart

The hardest part about starting a company is actually starting it.

I remember showing a couple buddies my early product screenshots and pitch for LearnBoost, and many were skeptical - they were looking for the 30 reasons something might not work, as opposed to the one or two reasons it might work.

Fast forward a few years: I'm running LearnBoost, we've raised from some great investors, and among other things we're creating the landmark libraries in Node.js. Then one day I get an ominous email from my friend Apoorva (I looked it up and it was on May 22nd, 2012) saying he had a major update and wanted to meet for coffee before work the next day.

What Apoorva showed me were some pretty crude mockups of what Instacart could become - he had some wireframes with grocery products and that’s it. You couldn’t make an order, and the data was a bit messy, but his vision was enormous. As I asked questions, I realized Apoorva had several unique insights into how to actually build a big business. It felt like I had been let in on some incredible secrets. And what really struck me was that it was the same feeling I felt when I was starting my previous company.

We both knew there was a need and a way for me to help as a founder who had been there before. That meant strategic help with fundraising, go-to-market, when/where to expand, etc. but it also meant tactical help like the employees/execs I have referred into the company. Heck, Apoorva didn't even have co-founders at the time :) but funny enough one of the candidates I referred then ended up being someone I once again referred in later and is now an executive at the company. A match made in grocery heaven!

I have loved being involved with Instacart. Not just because I love helping startups, because I do love that and it's my reason for being at CRV, but also because being the first call (or text) that a founder makes for any bit of help is personally rewarding. I love the ambiguity of the earliest stage startups and trying to provide clarity and sanity from my own experiences in the trenches. And I’m on a broader mission to be more helpful, and at an earlier stage, than most of the investors I had broadly gotten to know through my own experiences.

So while it’s early days for Instacart, and for the other companies I am lucky to help out (e.g. Pebble, as one example), it’s certainly fun to be on this journey from the other side of the table. And while I’m usually in my favorite spot of helping out ambitious founders behind-the-scenes, lately I just can’t help but be proud like a family member of how far founders like Apoorva have come in such a short amount of time!

Of course I'm on the lookout for the next Instacart or the next Pebble, which means that I'm always ready to be helpful to founders at the earliest possible stage. The goal is to remain the first call (or text) that great founders make when they need help.

The Fundraise Overplay

In poker, beginners tend to make a consistent mistake: they overplay their hand. Overplaying a hand is when a player bets big because they think they have an amazing hand, when in reality something like two pair is just a decent hand. In other words, the bet size gets way past the reality of the player's hand.

I have seen something like that happen in fundraising. And I almost fell for it myself when I was a founder.

The fundraise overplay is when a founder tries to raise a round that is way disproportionate to traction, the opportunity, the team, or some combination of those variables. Like in poker, sometimes a weaker hand with a big bet will flop a great result - but most of the time, the overplay means the person betting just gets burned.

I remember with one of my raises that I tried to raise $5M on crazy terms... when we had 6,000 users who weren't paying. Yep, totally nuts.

Yet in the end, we had a stellar team that was one of the top open source companies on Github, and traction was starting to ramp up, so we got a few termsheets for lower amounts and went with one of them.

So initially I had done a fundraise overplay, but corrected myself and realized that a $2M additional raise was exactly where we deserved to be - instead of getting the raise and reality even further apart.

The mistake I see founders make, especially in an over-heated environment that is analogous to playing with a table of very loose players, is they overplay their hand. From there, they try to raise way more than the stage they are at (or at a way higher valuation), continue to overplay their hand, and then try to come back when no one is bidding at that high price.

The problem is this: in poker, you just lose a hand and go to the next one. In startups, if you overplay your hand and get called bluffing, you might have just sunk your company.

Tactical advice for seed stage budgeting

When I started to write this post I realized I had a lot of value locked up in old emails between my fellow founders. So what follows are direct quotes that I have made anonymous, and they are from a few years ago. Not surprisingly, I find these nuggets of wisdom still hold true:

I justified it with a budget. [Investors] don't really read those at the time (nobody objected that I had engineers budgeted at $Xk) but they do expect you to hold to it, so try to put some real thought into it. They also definitely want to see 18 months of runway.

I can share my budget with you if you keep it deep dark confidential and realize that it's proved way wrong.

Don't be over-optimistic on payroll costs. Yes I am the lowest paid guy in the company.

Do not attempt to hold your breath while running a company - pay yourself tight but pay yourself enough to live your life without learning how to cook in your non-existent hours.

You cannot get away from offering healthcare to your employees. I think I pay roughly $500 / month / employee can't remember precisely.

Assume $1500 a month for "other." You always spend it. Travel, conferences, pizza, lunch, whatever.

Don't fuck it up ;)

That last one surprisingly gets sent among founders very often. And yet the reality is that a budget is going to always be precisely wrong (if you can model the future perfectly, you're in the wrong business). So think about your seed stage budget as guidelines, not rules.

Today's innovations are tomorrow's table stakes

At LearnBoost, we built one of the top open source companies in the world. Part of how we did that was an "innovation" at the time (early 2010), in terms of being prolific in the world of open source, creating much more value than we captured, and then using Github to extend our recruiting reach. Our team members created and open sourced many major Node.js libraries (socket.io, ExpressJS, mongoose, etc), which in turn caused many top developers to want to work on our team. We had a backlog of very high class developers who wanted to work for us!

It's no surprise that a few years later it is now table stakes in developer recruiting to open source and mine Github to find great individual talent, teams, and so forth.

It's true across the board: today's innovations and hacks become tomorrow's table stakes.

So what are you going to do? What's your "innovation" as a startup founder?

One thing you can do is act quickly when you've stumbled upon an innovation. You're to tap the well of that innovation as often as possible. I know we did at LearnBoost - by recruiting an "extra" engineer who would be our primary open source presence. I figured that having a talent like TJ on our team (more open source followers than nearly every company on Github) would create more value for the world and for us than anything else. In hindsight it's a no-brainer, but you can't believe the investor pressure I felt to not increase our burn on "frivolous things".

In the end, it worked out and our company was acquired by Automattic, the company behind WordPress. The team we built was a key piece in that acquisition.

The broader lesson: try your best to find a pocket of innovation and tap into it as much as possible before people even realize what you're doing.

Ultimately, chemistry remains, for most teams, baseball's holy spirit: too significant to ignore, too abstract to touch. A gift, by the power of which a team may abound in hope. "Maybe somebody will try to figure out a way to classify personalities and quantify the impact of those," Bill James once said. "I doubt that's going to happen in my career. But maybe the next one."

Fantastic article. And, as with everything I post here, it relates to startups.

The top question to ask when hiring

"Would we be lucky to have this person on board?"

That's the question I always asked when we were hiring someone, at either of my companies. If you set the bar that high, you'll tend to do great hiring. You'll also hire slightly ahead or behind plan, but the point is that you won't be married to some arbitrary plan - instead, you'll make more of the right hires over time.

At my last startup, I remember asking that question rhetorically when we had the opportunity to hire TJ, even though we were staffed up and probably already one engineer over plan when it came to hiring devs. By asking "would we be lucky to have him on board?" it became pretty obvious that we should still hire ahead of plan. [1]

Asking yourself this question with every hire helps you avoid the temptation to just keep filling seats and on the other extreme it helps you avoid the temptation to make it impossible to join your company.

[1] It was pretty obvious, actually, but it still proved the merit of asking this question. He has more open source followers on Github than nearly any company you can name on Github. We were lucky to have him on board... and to be fair, he was lucky as well in joining our company, which was already one of the top open source companies on all of Github :)

Getting sucked in to a patent dispute is no good for any entrepreneur. By the time it’s done, you may be able to recite 40 pages of patenteese by memory, and have learned your Markush from your Jeppson. But if all you need is a quick summary, just cut directly to the independent claims. You’ll be done in a minute.

Investor Distribution Lists

When I was a CEO, I used to send out mass updates to my existing investors. It was a great way to keep everyone on the same page at the same time - in other words, it was fair and it provided leverage.

Now that I'm a venture capitalist, one of the things I see good and great startups do is send out updates to prospective investors. And of course, it provides fairness, leverage, and also helps any founder drum up additional interest among investors. I love this and I wish I did this when I was running my own companies.

The ideal ongoing update is every 6-10 weeks and in the form of an email that is bcc'ed. You want to use bcc because you don't want investors to know who else you are considering - but you also don't want to email every investor you've ever met, just the ones you actually want to work with in the future.

This "text in an email update" tends to have the following:

Team: what key hires have you made recently? How is everyone doing? How are *you* doing? (Believe me, I want to know because... well, I've been there and I know being a founder is really tough!)

Product: what features have you released? What has surprised you with the existing product? What's in the pipeline?

Metrics: what do you find most relevant to your business? Given that, how are you doing on those metrics? Bonus: cohort data! I love cohorts...

Financials: when is "cash out"? What is burn like? When are you raising?

How I can help: I see *great* startups add this to their updates to prospective investors they want to work with. I love it. And I am more than happy to help, especially to win your trust and respect.

If you can keep these updates on a pre-determined schedule, that's even better. With the updates I used to send to existing investors, I sent them on the first of the month, even if there was a holiday - and I did that for +3 years. So if you are sending emails to prospective investors, it's also good to pick a cadence and stick to it.

In summary, it's good to send out updates to existing investors and it's also good to send updates to prospective investors. Much of the content from updates to existing investors can easily be sent to prospective investors, meaning there is only marginally more work to be done to open up the potential to a high upside win for your startup.

Outperform your contract

The kinds of people you want in your organization tend to "outperform their contracts" - typically from day 1.

I'm drawing a connection to sports, where athletes across many sports will sign a contract or get an extension and then surprise on the upside.

This manifests itself in two general ways - either a young player with certain risks tends to "de-risk" themselves and then go on to outperform (in football, this would be Colin Kaepernick) or a superstar goes on to outperform their large contract (in football, this would be Calvin Johnson).

Bringing it back to startups, I've hired and fired many people and almost consistently the best, enduring employees were the ones who came in right away and just blew me away with their contributions - from individual contributions to culture fit to generally picking up slack.

In a startup, if an employee isn't blowing you away out of the gate, you probably made the wrong hire. It's certainly a balancing act, because sometimes employees will take time to ramp up or your organization hasn't done a good/fair job of onboarding.

Otherwise, you will know deep down if an employee (or colleague) is the right fit if they are "outperforming their contract."

Five stages of player development

Joe Maddon is the manager for the Tampa Bay Devil Rays. He is known for maximizing his team’s potential. He partially does it by being a proponent of these five stages of player development:

1) I’m happy to be here. 2) I just need to survive. 3) I belong here. I can do this. 4) I want to make as much money as possible. 5) All I want to do is win.

In any great team I’ve been on, the majority of the team was in that final stage.

Not surprisingly, in business when most people truly want to win, the group or organization tends to outperform.

The only trick, then, is getting the majority of the group to that final stage. That’s the essence of leadership.

Bus factor

I first heard about "bus factor" in 2009 when I was working with a couple developers on the first version of LearnBoost (this was pre-cofounders and pre-pretty-much-everything).

Bus factor is basically how much you would miss someone at work if they were hit by a bus. Yikes, I know. But another way to think of it is how silo'ed is the knowledge of code, information, relationships, etc. within your organization?

The goal would be to reduce your bus factor as much as possible - a lower organizational bus factor means more redundancy and it also means better communication and teamwork, leading to better results. [1]

Startups are generally (but not necessarily) going to initially have a high bus factor by virtue of their early stage and limited headcount. I remember in 2010/2011 we didn't have a lot of redundancy at LearnBoost and I would worry about what would happen if we lost a team member - fortunately because of our culture and how we thought about granting equity, we never did, but that is always something that is top of mind to any founder.

As a founder/CEO, I would think of reducing bus factor two ways - can you either reduce bus factor in the organization or can you make sure the most critical individuals are going to stay at your organization for a long time? That's part of the reason why in both of my previous companies I preferred to give more equity than was "normal", so that my colleagues would have a partner mentality, and it's also an input into why I cared so much about work culture!

Incidentally as I was writing this, I realized it's also part of the same reason I joined CRV, which people don't really know is one of the few top VC firms with a completely flat partnership (read: equal carry and pay). So in addition to already being culturally aligned to work as a team, CRV is also economically aligned - all of which reduces bus factor, leading to better communication and teamwork, which leads to better support and results for founders. And that was something I was subconsciously drawn to, given I had believed in it and practiced it as a founder and now get to practice it as a venture capitalist.

[1] That's true in larger organizations, unless you're trying to keep an initiative or project confidential, in which case you're going to have to trade off and accept a higher bus factor.

Faster Than I've Ever Seen

I hear "we do X faster than I've ever seen" pretty often. It's a statement that falls on deaf ears.

Typically, what an entrepreneur has seen is usually their network. So while I also used to say "our developers are the fastest I've ever seen at shipping high quality product" it was limited by everything I had seen. Of course, some founders are also angels and have really great visibility. And of course I understand why I was saying it - you have to *really sell* your team when there often isn't much else to base a raise off of :)

Now take an investor's point of view. I "see" (aka interact with and prepare for) about 5 high quality startups a day. And I hope through continued hard work and reputation that number continues to grow over time. So after seeing 25 high quality startups a week, and having officially been at CRV for 2 months working with George, I've seen about 200 high quality startups.

My point is if you're the founder who employs the fastest developers you've ever seen, that's true; but it's also true that you're not likely to be the fastest devs that an investor has ever seen.

And that's why a bold statement starts to fall on deaf ears. Aside from often not mattering if you're the absolute fastest, it's a statement that can't be backed up to the person being sold - which means it's likely an ineffective tactic when raising money as a founder.

Age doesn't matter, backgrounds don't matter, nothing matters. You have four teammates, they can be anybody, and you either know how to click with them or you don't.

That's what makes Marc Gasol special, and that's why you can't totally measure him with stats. I voted for him as my first-team All-NBA center. Why? Because he can blend in with any four guys in the league. He'll anchor your defense, grab some rebounds, make every smart pass, post up when you need him. He's a phenomenal leader and chemistry guy. And he's malleable.

In startups, bottlenecks shift but worrying is a constant

Over the past few years, I became friends with many startup CEO friends (read: a loosely held support group). Actually, most of my friends are startup CEO's or founders. In any case, this group ranges from the "I don't have a concept" stage to the growing wildly and likely to IPO stage.

There's one thing all startup CEO's have in common:

Every startup CEO, regardless of stage, is always battling something - it's like everyone always has their hair on fire. In other words, bottlenecks shift but worrying is a constant.

From the outside, several of my startup founder friends and acquaintances look like they have everything put together, almost as if they are invincible. [1] Some share problems more than others, but everyone has challenges. I really know this because I'd see those challenges, and I'd share my own challenges with my fellow founders - we'd always help each other out.

These are the things employees never see and are unlikely to understand, simply because it's more difficult for employees to see the entire picture of the company historically or even day to day. [2]

In startups, uncertainty is constant, and it's the severity of uncertainty that is the variable.

[1] This is one of the biggest perpetual myths in Silicon Valley.

[2] Last month I (privately) surveyed startup employees with questions about how they view their management and their CEO's - generally, they're surprised when I tell them that most startup CEO's work out issues with other startup CEO's, rather than on their own or with their investors.

Winning as a Team

There are many things I care about, but when it comes to work I care deeply about winning, particularly as a team.

The New England Patrios are an example of winning with a team mentality. If you play for that team, the focus is not about individual statistics. The way the Pats make this work is that they do everything possible to make sure every individual buys into their team mentality.

With the Pats, they don't put up with the people who fall out of line. A player can be Hall of Fame talented (example: Randy Moss) but if they don't buy into the team mentality, they are gone.

No one is bigger than the collective team.

It turns out that the team approach works - the Pats are consistently one of the best teams in the NFL. If you look at their roster, year after year, they don't have the talent on paper that most teams have in the NFL. But what they also don't have is egotistical "superstars" running around the locker room, causing problems, and just playing for themselves.

It means that individuals have to sacrifice some pride to get their hands dirty and play a certain role within the team.

There are small, unnoticed moments where an individual has a chance to complain, or disturb the team dynamic. That's why even a seemingly innocent yet selfish moment can be the "straw that broke the camel's back" and get someone kicked off the team. The coaches notice these things, and it's not hard to tell who is buying into the team approach and who is "starting to go rogue" -- and if they go rogue, you can bet they'll be gone soon.

In a successful organization, it's about team members being the glue and caring about the team more than they care about themselves. In a world class organization, even the smallest contributions are noticed.

In the case of the Pats, not everyone can score the touchdowns, but everyone can pitch in. Across the board, the best coaches make sure to praise someone executing to perfection within their role.

This ultimately means trusting that it'll work out better in the end if the team succeeds, because then the glory of winning shines on every member of the team.

This post comes lightly edited from my old blog.

Are we aiming at the right planet?

A startup just getting going is similar to a grounded, pre-mission rocketship in one sense: anywhere in the entire universe is a potential destination.

The key differences from that point onwards between a startup and a rocketship:

While a startup is on a mission plan, that plan is subject to change midflight; meanwhile, a rocketship is on a set mission plan.

While a rocketship knows what planet it's going to end up on, it more importantly knows exactly how to get there; a startup merely has a thesis about which planet to visit and the way to reach that planet.

A rocketship isn't competing to get to the right planet, vaguely defined, against other rocketships; a startup is in a fierce competition.

I've now seen it firsthand many times - too often a startup's founders confuse themselves with a rocketship, thinking that their mission plan is set in stone or that there isn't a better, adjacent planet to visit or that they can take their time getting to the right planet.

Asking simple questions before the journey (e.g. are we aiming at the right planet?) and along the way (e.g. is this the best way to get to the best planet?) are useful techniques for founders to avoid confusing themselves with something that they're not.

Fortunately, asking questions and being willing to face hard truths before they become big problems turns out to be a low cost, relatively quick way to avoid the pain of ending up on the wrong planet or running out of fuel.

Forced to Innovate

After World War II, the Japanese factories were so devastated and resource constrained that they were forced to innovate. They had to be lean and mean, because they didn't have the resources that the U.S.A. and other relatively strong industrial economies had at the time.

Under a period of great duress, they came up with many modern operations techniques. These constraints on resources fueled a period of incredible innovation. Toyota and other Japanese companies came up with JIT production and many other innovations as a result of the various limitations facing the country.

That's the point - being put under stress will make you dig deep and can lead to wonderful results often bordering on the impossible.

From an old post I wrote many years ago on my old blog.

Trending Blogs

Recently Viewed Blogs

Rafael Corrales