Author Archives: pooneetkant

Equity for early employees: Opinions from the startup community

One of the biggest questions for any early stage company (or anyone looking to work for an early stage company) is how much equity to give early employees (including, in many cases, people ultimately categorized as “co-founders”). This is a topic I’ve thought a lot about, and I’ll get into my detailed thoughts in a later post. However, the starting point for anyone thinking about these issues should be to see what others in the startup community have said about the topic. Here is a collection of some of these approaches / opinions from respected parts of the startup community. 

Paul Graham, founder of Y Combinator – Paul Graham, the respected founder of Y Combinator, has a very simple approach to evaluating how much equity to give a potential employee: “If i is the average outcome for the company with the addition of some new person, then they’re worth n such that i = 1/(1 – n). Which means n = (i – 1)/i.  For example, suppose you’re just two founders and you want to hire an additional hacker who’s so good you feel he’ll increase the average outcome of the whole company by 20%. n = (1.2 – 1)/1.2 = .167. So you’ll break even if you trade 16.7% of the company for him.”

VentureHacks post Series A Compensation Ranges – This post at VentureHacks has some range data on what sort of equity compensation people get at post Series A startups. Slightly less applicable for pre-Series A companies, but still some data to think about. The post says that a COO could expect 2-5%; a VP, 1-2%; and a Director, 0.4-1.25%. One important note is that all of these ranges are after dilution from the Series A financing, so the pre-Series A numbers should be correspondingly higher for that and of course general risk reasons.

Hacker News, Y Combinator’s engineering-heavy news thread – Hacker News focuses on engineering topics, but the discussion about employee equity on this thread is heated and has a variety of opinions.  Some people said that 1% for a technical role at a funded startup was quite generous; others said that early employees should be getting much more.

Fred Wilson, Union Square Ventures – Fred Wilson has some useful thoughts about deciding how to grant equity. He says that for the first few people you bring on board to fill out the team, there is no formula. They should get what you think they need to accept the offer and be properly motivated – they are extremely important employees for everything from setting the culture of the organization to executing on the product vision, and so their equity has to be decided on a case-by-case basis. After these team members are in place, Fred advocates moving to a strict formula based on the cash/stock split future employees receive.

Mark Suster, respected entrepreneur turned VC – Mark writes an excellent startup / VC focused blog, Both Sides of the Table.  Although he does not have much in the way of hard numbers, his general framework for thinking about how much equity to seek is interesting. He breaks employees into two categories – those who are in the “learn” stages of their career and those who are in the “earn” stages of their career, and argues that if you are in the learn stage, the amount of equity is not as important as what you might be able to learn from the experience.

The Pyramid Approach Although this is probably not a “respected member” of the startup community, this post discusses a formula based approach that I think mirrors how a lot of companies think about this question. Basically, the post argues that founders should reserve 20% of the company to pay the first 100 employees. The first 10 employees get 1% each; the next ten get 0.5% each; employees 21-30 get 0.25% each, and so on.

Joel Splosky on Dharmesh Shah’s – Joel has another “layered” approach to dividing startup compensation.  In his view, the founders should end up with 50% of the company, and then there are four layers of employees.  The truly early employees, of which there will probably only be two or three, should split 10%; the next set of employees, perhaps ten, split another 10%; and so on.

David Beisel, Genuine VC – David Beisel is a former entrepreneur turned respected early-stage investor. David’s approach is to distinguish between two types of hires – senior level hires who could be expected to grow with the company, and junior level people brought in to fill a specific and smaller need.  For the senior level people, focus on equity; for the junior level ones, focus on some reasonable amount of cash (if possible with the funding situation).

These resources should provide a good starting point for a company or potential employee thinking about the tough issue of determining how much equity is “right” or “fair” to give or seek. If you have come across other good resources or thoughts on this question, please let me know in the comments below!


Startup funding: Convertible notes primer

I’ve been working in the early-stage (pre-Series A) startup world for over two years now, and have been part of many discussions about how to fund companies at that stage. The most common funding source at this stage tends to be the general “friends and family” pool, and the most common mechanism for actually raising those funds tends to be convertible notes (often abbreviated as “converts”).

There are other options – this post talks argues for raising early-stage through straight equity rather than converts and even provides the framework legal docs to get started. I will discuss my experiences with converts and my thoughts on raising via converts vs. equity in a different post. In this post, I aim to provide a simple primer on how converts work and what the key terms are. The real reason for this post is because while at Winestyr, I’ve had to explain converts to a number of people that I’ve hired…and it can get complicated. Techcrunch also has a great guide to converts, but I think it’s a bit involved for someone just getting started. To keep things simple, I’ll also avoid discussing tax implications of converts vs. equity financing – if you want more details on that, the Techcrunch piece has some good starting info.

Converts exist for one main reason: they save both the founders and the investors time by allowing them to defer important decisions about the company until a later date. The key question they defer is valuation – if an investor wants to invest $100,000 in a company, what % of the company should that investor receive for that $100,000? For publicly traded companies, this is never a question – if you buy $100,000 of Facebook stock, for example, that means you now own roughly 0.0001% of the company – because investors have determined that Facebook overall is worth about $100 billion. As an investor, you know this when you purchase the stock because all of that information is publicly available and governed by Facebook’s trading activity on the public stock market.

However, private companies like startups do not trade on any sort of public market, so there is no outside source to consult when you want to know how much of the company your $100,000 will give you. In a world without converts, you’d have to come to an agreement with the founders about what the company was worth, and then document that agreement as part of your investment. For example, you and founders could agree that the company was $900,000 without the $100,000 investment, meaning that after the investment it would be worth $1,000,000 and you’d own 10%. However, coming to an agreement on the value of a company that is unlikely to have much in the way of revenue and certainly won’t have years or even months of operating results is a hard process! The founders of the company will want that valuation to be as high as possible to they give up less of the company to you the investor, and vice versa. Converts allow you to defer this difficult (and time-consuming) discussion.

The way converts accomplish this is by making the assumption that the company will raise additional funding in the future from more experienced investors than the initial friends and family group. This could be from venture capitalists or from sophisticated private investors often called “angels”. The assumption is that these sophisticated investors will have an easier time establishing a value for the company, partially due to their experience but also just because the company will have more data with regard to how it is performing that will help arrive at a value.

Mechanically, the way this works is that the convertible note simply converts into equity at whatever valuation that later round of sophisticated-investor financing sets. For example, if a company raises $500,000 from friends and family in the form of a convert, and then one year later raises money from VC at a $3,000,000 valuation, the convertible note would convert into an ownership percentage equal to $500,000 / $3,000,000 or 16.67% of the company.

There are of course many complicating terms and factors involved with converts. The three main ones are interest, triggers, and caps/discounts. One big reason for these terms is to provide the convert investor with some extra value for putting money into the company at a riskier time than the subsequent sophisticated investor – it just wouldn’t be fair to that earlier investor to give them the same terms as the later investment.

  1. Interest – most convertible notes have an interest rate attached to them (usually in the 8% per year range). This gives the investor some additional value when their note converts into equity. In the example above, if the $500,000 convert had an interest rate of 8%, after one year when the company raised VC money, the convert would have a total value of $540,000 instead of the $500,000 that was actually invested. This would in turn mean that the convertible note would actually be worth 18% of the company instead of the 16.67% above ($540,000 / $3,000,000 instead of $500,000 / $3,000,000). Interest rates are very standard in converts and rarely a source of significant contention during negotiations.
  2. Trigger – Convertible notes have to define some “trigger” event – that is, what actually causes you to undergo the conversion? This is usually either a funding amount, for example, the “first additional financing of at least $1,000,000.”  It could also be a valuation threshold, such as “the first additional round at a valuation of $2,000,000 or above” or just something as simple as “the first additional round of financing that sets a valuation.”
  3. Cap/discount – One outcome a convert investor might be worried about is since they put money in without establishing what that investment is worth, what happens if the company really blows up so that when they raise their next round, the valuation is very high? Caps/discounts solve for this issue. A cap is the maximum reference valuation when calculating how much equity the note converts into – in the example above, if the note had a $2,000,000 cap, then even though the company raised money at a $3,000,000 valuation the convert holder would actually own 25% of the company post conversion ($500,000 investment / $2,000,000 cap).  A discount is similar in concept – it gives the convert investor a discount to the valuation of the equity round when calculating the post-conversion ownership.  Sticking with the example above, if there was a 25% discount with the note, the note would convert using a reference price of 0.75 * $3,000,000, or $2,250,000, yielding an ownership percentage of $500,000 / $2,250,000 = 22.2%. These terms are often used together, and are more complicated to negotiate and resolve than either the trigger event or interest rate. There is no wide consensus on what is “standard” or “market”, but from my experience and reading, I’d say a discount of 15-30% and a cap in the $5-7 million range seems reasonable.

Hopefully this post helps provide some background on what converts are and how they work. If you have any questions or think something is unclear, let me know in the comments and I’ll do my best to rectify.

Review of Udacity CS 101 – Progress through Lesson 3

As mentioned in my coding plan, my first step in learning to code has been working through Udacity’s CS 101 free online course. I read a number of reviews (most of the available reviews suck, though – this is the best one I found) before I started working through it. Most of the complaints seemed to be from people who had literally no background in any concepts of CS, and since I do have a bit of a background I figured I’d be okay.

I’m now well into Lesson 3, so I’ve gone through Lessons 1, 2, and 2.5, which was added after the initial course was put together. I’ve done the homework for each as well.

Overall, I think the course is great! The instructor does a great job of being clear, the forums quickly answer the typical questions that might come up, and the exercises build upon each other well. One thing I did do is to follow along with the course running my own Python interpreter locally on my Mac. I used IDLE (for more details than you want about how to get everything all set up, check out this post). That allowed me to play around while Dave was talking and generally gave me more flexibility.

Lessons 1 and 2 were a good review for me, and I imagine they will be a good though slightly challenging introduction for someone completely new to the world of CS. So, I was able to get through those pretty quickly. The course does a good job of making it seem like your progress is tangible because everything is being taught with a goal in mind – to build a rudimentary search engine. That really helps anchor the material and make it much more concrete. Plus, the goal is not some BS goal of building a simple program – building a search engine is serious business, and I imagine it serves as a good incentive to keep going.

Over the coming weeks I’m excited to continue onward, and will do another post on my progress after a couple more lessons!

Why I’m Learning to Code

There has been a lot written on why people should or should not learn to code. My reasons for wanting to learn to code are simple:

1) I find it intellectually stimulating to learn a new skill like this. (After I get to some level of coding proficiency, I want to learn how to play the piano, which I think will be much harder)

2) I like understanding how things work on a deeper level, especially something as fundamental to my life as “technology”

3) I’ve been working for web startups for over 2 years now, and I know that learning to code will make me a better employee, manager, or founder at any web company

4) Relatedly, I’ve been repeatedly frustrated by my inability to make even small changes to our site or help out with some of more trivial coding tasks we have

5) I’d like to eventually be able to build my own MVPs for ideas I want to test

As I said in my intro coding post, my goal is not to become a coding master – at the start, I just want to learn enough to be able to converse more fluidly with my CTO and to hopefully help out on smaller projects or tasks. Eventually I’d like to build some rudimentary web apps of my own…but I’m not expecting that to happen overnight.

Learning to code: The Plan

I’ll get into just why I want to learn to code in another post, but here is what I’m planning to do. Let me know if you have any suggestions!

Note: I’m not a total novice when it comes to coding. I did play around with Basic a bit in middle school, and HTML and C++ in high school. I also am reasonably proficient with R, which is my favorite data analysis software (and it’s free, unlike Stata / others), and I’m good with building functions in Excel (courtesy of my years at J.P. Morgan). So, it’s not much, but I do have a bit of a background which definitely helps.

This progression is geared toward ultimately developing in Ruby on Rails using tools like Heroku, Git, and AWS, since that is what Winestyr (and of course, many other sites) are built in.

Step 1: Take Udacity’s CS 101 course – this is a very entry-level course that uses Python to teach principles of computer programming while building a rudimentary web-crawler. I hope it’ll refresh what I learned long ago and set me up to progress slightly more quickly through subsequent courses.

Step 2: Go through “Learn Ruby the Hard Way” – I want to have some fundamentals of Ruby before I dive into Rails.

Step 3: Take the Hartl tutorial to Rails – I know there are lots of options out there, but I like Hartl’s approach, and it comes highly recommended.

Step 4: Take Coursera’s Startup Engineering class – A high-level class on turning academic CS knowledge into something actionable…covers some front-end stuff, databases, testing, deployment, etc.

From there I’ll re-assess and see where to go next!

Note: thanks to Kapil Kale, Mark Glenn, and Elliot Garms for some of the thoughts here.