May 2010 Archives

Gmail has become unusably slow

 
When I switched to Gmail in 2004, I believed the hype. Never delete a message again--no need. We have tons of space, and you can search it all really fast like Google.

That time has passed. Gmail has gotten slower and slower for me, and as of the last few weeks it has become unusably slow. Before you ask, yes, I've tried it across lots of browsers and computers.

It can take 20sec to switch labels, and even longer to search for something. But here's the worst part--it takes just as long to send a simple message!?! Why? What does sending have to do with anything?

It's become the bottleneck in my day, and I don't know what to do about it. And I'm not alone.

A few days ago I decided to start taking action. First I emailed support. OK, first, I tried to email support. 

Have you ever tried to email Google support? It's almost impossible to find the contact form. Here's the support home page. I dare you to find out where to report this slowness issue.

You get to this page on slowness. After going through the wizard, you click on 'report your issue' at the bottom, and it takes you here. Wait, that's not a contact form, and you can't get to one from that page! Anyway, here is a contact form; I found it going through another problem wizard.

Needless to say, I haven't heard a response :)

Next step: I disabled chat, buzz & tried the older versions of Gmail. No luck. Then I disabled all labs, after which I perceived a very modest improvement, but still unusable.

Next I removed most of my labels. I have four now (down from 32). This seemed to help a bit as well, but still not much.

So this morning I went drastic. I deleted all my contacts and started deleting mail. Ridiculous huh? That totally breaks the original selling point of Gmail, but like I said I'm at wits end here.

Deleting stuff has resulted in the biggest improvement so far, but it's still slow. Perhaps a bit better than unusable now, but still terrible.

You are currently using 4247 MB (56%) of your 7459 MB.

In a last ditch effort, I bought some extra storage from Google thinking maybe I'd get some kind of premium level service. So far, no.

Google's been recently launching lots of cloud products, most recently a storage product to compete with Amazon's S3.

In other words, they obviously have the resources to make Gmail fast. So what's the deal? They must know about the slowness. The only reasonable explanation is that they are consciously under-resourcing it. Again, why?


Update: there are also a lot of good comments on HN.

Update 2: after a bunch of testing with my account, I'm confident at least my slowness involves something around having more than 4GB of mail. I deleted a lot of messages and got down to 3.6GB. It was then relatively fast again. I then sent myself a 25MB file (the limit) repeatedly until I got back up to 4GB. Right after 4GB, it got slow. Go figure.

Update 3: Google reached out to me and "fixed" my account. Here is what they said

When can startups be called successful, e.g. reddit, dropbox?

 
My last post on startup company failure rates engendered discussion on what constitutes success in the context of a startup company. I specifically avoided this subject because it was ancillary to my point that some startup sub-groups have inherently different failure rates than others. 

Common metrics for failure are going out of business or not having a Web site (without a known liquidity event of some magnitude). Thinking about these it seems clear that success metrics are not the opposite of failure metrics. I think this comment on my last post gives this perspective well.

I think the failure rate is better than the oft quoted 90%, but the success rate is not better than 10%.  Of course, what failure and success mean drives the answer.  In a pool of smart technology based startups, if failure means losing all the invested capital, I suspect the failure rate is 60%-ish.  If success means an annual IRR of 100+% and liquidity within 5 years, I suspect the success rate is much less than 10%.  These are investor centric definitions.  If you used more entrepreneur focused definitions, the rates would change somewhat.

So yeah, a lot is in the eye of the beholder. However, I think we can do better than that. In a lot of cases, most people will agree x company was a success. Where is that line?

Let's take this other comment about reddit as an example. Reddit sold to Conde Nast for an estimated $12M after having raised ~100K (according to crunchbase). Even if those aren't the true facts, let's just suppose they are for the sake of argument.

Yes, it definitely does depend on your definition of success. For a lot of young entrepreneurs, getting funded, an early exit, or a buyout from Conde Nast are all success. That makes sense in a way, especially because these outcomes are all popular and well-publicized. 

If you think of a business as something that benefits society, makes sustainable jobs and can outlast a founder, I think the bar is very different. Put another way, I am more proud to have started sustainable companies than I am to have sold them, and I would like young and new entrepreneurs to have the same perspective. 

I think the vast majority of people would say reddit is/was successful. Their site is used by millions of people, and they returned good money to both the founders and the investors, though seemingly not yet to CN. 

It wasn't a 100M+ exit for sure, but I think it certainly passed the bar of success. Clearly you can keep being more and more successful, but there is a line somewhere and they passed it. 

I'd say the same thing about the sustainability and jobs reference. Those are more ways (like more money) to be even more successful, but aren't necessary conditions of success.

Let's tweak the original outcome though. What if they had raised $12M and as a result the investors got their money back and the founders got nothing? It would still be used millions of people, but I find it hard to call it a success from either the entrepreneur or investor perspective. Maybe you'd say it is a successful site, but wasn't a successful company.

What if they had raised $10M, and the founders ended up with $1M each and investors the rest, i.e. a meager return? That would definitely not be successful for the investors, but I think most people would still say successful for the founders. However, I don't think that crosses the general line because a big subgroup, i.e. investors, would not look at it as a successful outcome.

What if it was the original financial outcome but CN shut down the site immediately? I think you'd still call it a success for both the entrepreneurs and investors, and a success in general. That is, I don't think what the acquiring company does matters that much in terms of evaluating the success of original entity. It's a sad fact that most acquisitions screw up the company, but it is what it is. 

In that case, I think you'd say reddit was a success, but the acquisition was not successful. Again, I take the commenter's point that creating a sustainable company might be considered more successful, but that doesn't mean it wouldn't have crossed the bar of general success.

Here are some more:

  • Is Dopbox successful? It's used by millions and making money. I think if you asked anyone at this point they'd say it is successful except maybe the investors because they haven't exited yet. But if Dropbox tanked tomorrow, would it still be a success? Probably not. So then...

  • Do you need an exit to be successful? I don't think so. When my last company exited, it had a run-rate of about $1M/year with ~85% gross margins and was trending upward on both. If we never exited and that had stayed for a few more years and then died, I would still call it a success. It would have returned the founders a decent amount of cash. We had no investors though. So then...

  • Does success depend on the return of your investors? Certainly it does in their eyes. You could walk away with a decent exit and they could get a measly return. In that case, you might think it is successful, but I think from the outside there will always be an asterisk that goes along with the story about your company.

    That is, to be universally thought of as a success, I think you need a good return for both the investors and the founders. Since my last company took no investment, I think it can be called successful in the no exit scenario, but if we had taken money and not made a good return for the investors, I think that would make it not successful. So in a real sense, the bar is perhaps lower if you don't take money. With that in mind...
  • Is ramen profitability successful? Taking the above into account, if you don't raise any outside money, and it enables you to live how you want indefinitely, then I think so. If you do take money, then no; it's a good milestone along the way, but you aren't successful yet. But...

  • Is YC successful? I think it's in a similar situation to dropbox, perhaps even further along. In an anything's possible world it could tank tomorrow, but that is rather unlikely. They seem to be about break-even, though a few 100M+ exits could change that quickly. And there are a lot of possibilities on that front: justin.tv, posterous, dropbox, etc. 

    But from another perspective, they're probably already successful because they've changed the startup landscape, right? So...

  • Can a company be successful without a good return for anyone? I think no, though I think the people involved can get a lot of respect from the whole ordeal. If YC tanked tomorrow they'd still probably be thought of as visionary the way companies who enter new markets are thought of visionary even if they aren't the ones that end up capturing those markets. In other words, they would made a recognizable difference in the world, even if they didn't return much to the founders or investors.

Another way way to look at this question is would a given scenario result in future investors being more or less likely to fund the founders. I hesitate to tie anything to funding, but I think it puts an interesting twist on it.

Suppose the founders choose an exit that makes them money but not the investors, e.g. a talent acquisition. Future investors may be wary of those entrepreneurs. Does that make them any less successful? What about their companies?

The amount of money you have also seems to drive a lot of the eye of the beholder stuff. A $1M personal exit looks very different depending on where you sit.

Finally, another nuance seems to be tied up in expectation and potential. If Dropbox had taken no money (not true, but go with me) and tomorrow sold for $12M would it be considered successful? I think people expect more since the potential seems way higher.

Similarly, if a famous entrepreneur has an early exit, is that any less successful? If Marc Andreesen had taken no money for Ning (also very not true, but go with me) and sold it for $12M, would it be considered successful? Again, I think not because people expect more.

Startup company failure rates are grossly misleading

 
Every startup founder has heard that 90% of startups fail within the first five years. I'm sure that rate is very wrong, but let's assume for the moment that it's true.

Even if it's true, it's probably still wrong for you because you are probably in a subgroup with a much lower failure rate. If you're a startup founder and reading this blog the following is likely.
  • Your business has low initial capital costs.
  • You're doing a software/Internet startup with low marginal costs.
  • You're not doing it blind, i.e. you're getting a lot of good advice from experienced people.
If you add to that the right personality traits, e.g. determination, and some savings to give you a decent run-way, then your expected failure rate is certainly much lower than 90%, perhaps 50-60%.

At that rate, you might as well call it a 40% success rate.

Duck Duck Go Searches Are Now Externally Anonymous

 

duckduckgo.pngDuck Duck Go searches have been internally anonymous for a while. IP addresses are not stored. Neither are full user agents.[1] 

When you search at Duck Duck Go, there is no way for me to know who you are, or tie your searches together. For more info, check out the privacy policy. It's short--206 words; compare that to Google's 2,137 words.

When you search on Google, not only is your info stored, but also when you click on a link, your search terms are passed on to that site via the Referer header. A lot of sites use this information to tailor content and advertising to you specifically. Your searches also show up in analytics tools, which people use for SEO and other tracking purposes. This information leakage creates legitimate privacy concerns.

If you use the encrypted version of Duck Duck Go, the Referer header is not sent as per the HTTP standard. However, not everyone wants to use the encrypted version because it is slower to initially connect.

As of today, the Referer header is also not sent when you use the normal http version of Duck Duck Go. In other words, your search terms are not leaked to the sites that you click on, regardless if you use the encrypted version or not.

Referer headers are sent by your browser (client side) automatically, so I can't control it from my servers.[2] As a result, I'm currently using a meta refresh to force a client side redirect, and if meta redirects are turned off, a JS location.replace from that redirect page.[3]

After a lot of testing, I've determined there is negligible slowdown with this process due to the way I've implemented it. (In fact, most search engines make a background request for click tracking already--I don't.) The meta page is sent in memory via the nginx echo module. As you're already in a keepalive state, this happens essentially instantly. 

However, I realize that some people may want to turn this off, e.g. if you want your search term leaked. In conjunction with this release, I've also added a redirect setting to do just that. You can also force it one way or another using URL parameters.

A related issue was brought to my attention a few months ago on reddit, which has to do with the use of images served from Amazon's S3 service. When those images (mainly favicons, but sometimes in the 0-click box) are requested, your browser sends to Amazon the Referer header, which includes your search. In response, I had made four changes to address this issue.
  1. I added a setting to use POST requests, which solves the issue completely.
  2. I added a setting to disable favicons.
  3. I added a setting to disable the 0-click box.
  4. I started making all calls to S3 over https, so the headers would not be sent in plain text.
There are two issues with these. First, they all impact usability. POST requests break the back button and URL copying; you may want to see the images; and https slows things down a bit. Second, a couple days ago it was pointed out that despite these changes, the Referer header is still sent to Amazon, albeit encrypted, and they could be storing it. 

I decided to ask Amazon about their logging policy. They have a setting called Server Access Logging, which I do not have turned on, and so their logging policy in this case was unclear. Apparently they do log even if Server Access Logging is turned off.

All of the information exposed via Server Access Logs is in our internal logs - including referrer strings.

There were a lot of good suggestions on Hacker News on how to address this issue, but they all similarly impacted usability in one way or another.

I have now solved this problem by setting up a reverse proxy between me and S3. This costs me more bandwidth and server resources, but it is worth it to solve the privacy problem for you. Additionally, it actually improves usability because a) I set up a cache on my end and b) I can now turn off https to S3.

Furthermore, it is even more private than simply dropping the Referer string. Since you are no longer making the request on your side, your IP address isn't being sent to them at all. I can also explicitly set the Referer string (using the nginx more headers module), which I set to 'http://duckduckgo.com/';

I welcome feedback on these new processes. As they are new, I'm sure there are bugs to work out and further optimizations to work in. I already have a few in mind myself.


[1] Actually, user agents currently are not stored at all. In the future, however, I may compress user agent strings to short codes, e.g. FF for FireFox. For reference, the current nginx logformat is as follows.

logformat  main  '127.0.0.2 [$timelocal] "$request" $status $requesttime $bodybytessent "$http_referer"';

[2] As I noted, if you click on an http link from an https page, the Referer is not supposed to be sent. However, if you have the server redirect you from an https link to an http link, clients will pass the Referer header through. Annoying!

[3] Note that this client-side, so if you have a client that doesn't behave, it may not work. I've tested it on most modern browsers, including Chrome, Safari (including iPhone/iPad), Konqueror, FireFox, Avant, Opera, and IE (including IE6).  

My angel investing strategy vs Joshua Schachter

 
Joshua Schacter has a great AMA thread on HN about angel investing. He shares a lot of interesting info about his strategy, and it's about as diametrically opposite to my strategy as you can get.

I think his strategy is essentially the Ron Conway strategy, and at his rate ("34-ish investments" in 1-2yr?), he'll be the next Ron Conway in short order. 

The strategy is pretty simple. 
  • Get access to all major deals (super high deal flow);
  • Invest in anything that seems like it may be big (lots of investments);
  • Don't worry much about terms.
For the record, I think this is a great strategy if you can do it, but I can't do it for two reasons. 

First, I don't live in Silicon Valley, which is essential to getting super high deal flow. Most deals are there, and to get access to everything you seem to have to (at least occasionally) have in-person meetings, go to parties/social events, sit on panels, etc. Bottom line, without being there, you won't get into all the deals you need to be in to make this strategy work. And as I additionally hate short trips and airplanes, I'm even less able to compensate for my home in Valley Forge, PA.

Second, I don't have enough money. Even at his small investment size (10-25K), he's still done "34-ish" deals. Quite frankly, until I show some modicum of success, that is too much outlay for me.

As a result, I'm looking for very different investments. Here are some major differences I found intriguing from the thread.


I've done 34-ish investments.

I've done three :). If you're doing the Ron Conway strategy, you're trying to get in relatively early in the next billion dollar company, or at least in ones that will sell for $100M+. That's why terms don't matter that much. When you get 100-1000x time returns, getting access to that big deal is the most important thing.

The problem is, it's really hard to tell which deals are going to be the big ones. And there are only like 15 100M companies founded each year (reference from Marc Andreessen that I can't find right now). When you cut out ones you won't do because of product area (life sciences, clean tech, etc.), the number dwindles to just a few or less.

As a consequence, you have to do a ton of deals each year, just like he's doing. Given that I can't do a ton, I've taken a different approach.

I'm not looking for the billion dollar company, or even the $100M+ exit. I view it as entering a raffle where I can't buy enough tickets to make the odds decent.

Instead, I'm looking for companies that can go big in the $20-50M acquisition range. Sure, if everything went 100% perfect, they could IPO or $100M+ exit, but the more likely "medium-high" scenario (and the one I look to) is a decent sized acquisition.


I can't or won't negotiate terms, lead a round, syndicate a deal, or sit on a board.

I'm on two boards (of my three investments). I led one and half-led another. I'll negotiate terms--you sort of have to if you're leading. Note that I'm also willing to take terms as is if it makes sense; I've done that once already.

This also follows from this standpoint of doing less deals. Less deals means each one counts more, so I'm putting a lot of effort into each deal, both before and after investment.


Use of capital is a typical open question. But no, I do not ask for a budget. They should spend most of it on people and the rest on other constraining resources.

This is an example of effort I spend before deals. I do a lot of due diligence. I do want to know about use of capital, for example, and everything else for that matter. I'm not the guy who is going to write you a check after one conversation, which you should be wary of btw. 


Startups don't generally tell me about their decisions often enough for me to be able to benchmark this, I think.

Post-investment, I'd like to help as much as possible. I don't nag or even am very pro-active since I don't want to bother founders. But I'd love to help move the investment forward, even technically. For example, I've helped with scaling/downtime/dns/email/testing issues in addition to the normal intros/strategy type help.


What portion of your investments are outside of the bay area? None.

I've done one in Philly (though they just moved to SV), one in SV, and one in BOS. Location (as long as it is US) is not that important to me.


Are you doing it for financial reward or for the enjoyment of it? Probably the second, currently. I haven't seen a lot of return in real dollars.

I have enjoyed it so far immensely, though I am definitely in angel investing to make money. As such, I'm looking to get a decent equity % so that when an exit does happen, I can make some return and do more angel investments. As I've tried to articulate above, if you're not expecting a 100M+ exit, then the deal terms matter more.


Valuations are often in the $2mm-$5mm range. Sometimes much higher. Never lower. 

I'm looking for $750K-2.5M. For deals I'm leading, the target is <=$1M. For deals I'm joining, it's above that. So there is a bit of overlap, but not much.

I think that means I'm looking at earlier deals and never "hot" deals. You don't want to go over 20% dilution, so we're talking about a minimum $200K round, and scaling up as the valuation goes up.

7/10 companies coming out of accelerators aren't getting funding. There are a lot of reasons for that, e.g. solo entrepreneurs, ideas that need major pivots, etc.  However, I think there are some great potential companies in there if they had the right mentors.


In my experience, companies raising less than 500k are making a huge mistake ... The point of raising funds is to hire people who can accelerate the growth of the business. $200k gets you a person or two, which doesn't do very much. So you find yourself raising again pretty much immediately. The fundraising process is a colossal waste of your time. An entrepreneur going this route is not making good trade-offs. In my experience, most of these companies fail.

In my opinion/experience, this is not the case if the company is already making revenue (even if just $1K total) and has some good user acquisition ideas. That's really what I'm looking for. And in those cases, I think a $200K round makes perfect sense because the money goes to acquire customers and not to head count. If the process works, then the earnings that ensue can fund head count, if and when needed.


If $200k is good but $500k is not as good, it smells like a lifestyle business. That's not bad, but it's not a good investment for me.

There are two issues here. First, that's why use of capital is so important. If the $200K is for a specific marketing purpose that fuels a viral engine, then it makes perfect sense and is not a lifestyle business. That is, if you have a real customer acquisition channel that you can pour money into, then you shouldn't need more than $200K to test that assumption. Or even less. 

Second, dilution/valuation is directly related to round size. If you're talking really early where I want to be (at lower valuations), round size is essentially capped at 20% dilution.

In any case, I agree that fundraising is a major time sink, but raising a small amount can go really fast if you have a group of co-investors that believe in what you're doing. That's what I'm trying to be part of in Philly.


I expect it will take 5-10 years to get a return in many cases.

I expect a much shorter time horizon, avg 5 years. I think the diff here is related VC exit times

Again, it comes down to strategy. If you're looking to 100M+ exits, you basically need VC. I'd fund a deal that may not need to take VC, ever.

***

Just to show that I'm not a complete contrarian (crazy person?), here are some things he said that I completely agree with.

Communication is key. If you want people to do things, you must communicate in ways that facilitate the action. Make it easier for me as I am bandwidth constrained.

I don't understand b2b and am terrified of it. [bc of] Sales. Seems very hard. Not that that route is bad but I understand b2c better.

I really want to see a working prototype first. Rarely do i invest without that. ... You aren't learning anything until you are interacting with users, and learning is the most important thing you can possibly be doing. Through friction the apple is polished, I think.


Duck Duck Go Traffic & Sponsorship

 

sponsor.png

Duck Duck Go served 1,182,204 result pages last month. You can now track the progress (or lack thereof ) on the new traffic page. As you can see, this month is on track for a bit more. 

Thank you to everyone who's been spreading DDG for helping to make these traffic #s happen. And a special thanks to the reddit & Hacker News communities for being so supportive.

I'm also now accepting DDG sponsorship. I just put a square 85 pixel sponsor banner on the right bar

For those users who hate ads, note that there is already a setting to hide the right bar, and if enough people request it, I'll create another one just to hide the sponsor banner (within the right bar). However, I plan to only accept high quality sponsors and banners.

Sponsorship is exclusive. That is, there will only be one sponsor at a time, on a weekly or monthly basis. The sponsor banner will display on all search result pages.

The new traffic page is meant to count exactly those (sponsored) page views. Pages without search results, e.g. the homepage, about page, etc., are not counted. Also, if someone clicks 'More links...' on a search results page, that is not counted as another page. For example, if someone clicked 'More links...' five times that would still just be one page.

I have no idea what to expect so this program may fail fast. (This program is modeled after Daring Fireball's sponsorship program.) I will keep the traffic page up regardless though. 

The initial sponsorship rate is $1.5K/week or $5K/month, subject to increases based on demand. To kick things off, any sponsor who pays before the end of this month can get a discounted rate of $1K/week or $3.5K/month for up to two months (pre-paid). Of course, please let me know if you have any questions.

A FB ad targeted at one person (my wife)

 
ad.png
The other day I gave a presentation with Steve Welch on the use of social media in politics. Steve was walking through (live) the process of creating a Facebook ad. 

He started targeting the ad by location and interest, and the number of potential people he was reaching began decreasing on screen (Facebook tells you dynamically). Then I got to thinking--could you target an ad at literally one person? 

In theory, it wouldn't seem that difficult, given you can target by a lot of different things: interests, school, location, workplace, etc. If you could actually see their profile it would be super easy, but I don't think you even need that much. With just basic facts about them, e.g. their LinkedIn, you could target an ad sufficiently narrowly to reach essentially just him/her.

So of course the next step was to actually try it.

My wife goes on Facebook a lot to look at pictures and stuff. She's also mentioned many times how she actually likes the ads because they're targeted pretty well at her interests. So I thought she'd be the perfect subject.
reach.png

First I made the ad (above). Toast is a name we sometimes call our son Eli. Yes, I even messed up grammar in the title of my ad...

Then I started targeting. It proved just as easy as I thought it would be. First I  targeted to literally just her by using the stuff on the right plus her major and gender. Btw, I couldn't get FB to say it would target any lower a number than 20 people (although it does say fewer.)

But when I logged into her account, and it wasn't showing all the time on top.

So I increased my CPM bid. But it still wasn't showing all the time on top (it showed at least occasionally then though). Side note: you can see a bunch of ads by going to the ad board page.

Then I backed off a bit (to the targeting on the right) so I'd be included as well. I logged in and found the ad on my account and "Liked" it as well as clicked on it. My thought was maybe their ad system would then perceive it as a better ad and show it more. This seemed to work, but it is hard to tell whether that really had the effect or not. In any case, it started showing up a lot more. Not all the time, but when it did show up it would stay on top for a number of page views.

Then I waited until the night and subtly prodded her to check out Facebook. We went through old photos of Eli and it was just sitting there on the right on many of the pages, but she didn't notice!

I saw first hand why CTR is so low on FB. I steered us towards the album with the picture I used for ad and literally there was the big version of the picture and then the ad on the right (below). 

And then she noticed it. 

She immediately got what was going on, she looked at me and we broke out laughing pretty hard for a while.


ad2.png
This was of course all in good fun, but I also think there could be some good business cases for this technique :)

To pivot or not to pivot

 
There was a great post last week by Blake Jennelle that said you should pivot if your startup feels that you are constantly fighting an uphill battle. The premise is that if you have a really good idea+execution you'll get product/market fit, and "gravity" will do the rest. If you don't, the uphill battle will be endless and demoralizing.

I generally agree with this advice, but I feel it is often applied incorrectly. I see startups pivot (or even give up) too early, and visa-versa. In this context, I'm talking about major pivots (as opposed to tweaks).

Don't give up before your launch. 

People get launch jitters, which often stem from fear of rejection, humiliation, failure, etc. Ignore all that. Embrace the MVP concept and launch something already!

Even if you think your idea won't work and you think you have better ones, still launch. You never know what will happen and what pivot thoughts may come from real customer interaction.

And consider this--investors want to see determination. Having a history of switching between ideas without launching anything pretty much shows the opposite. OK, so I hope I convinced you to launch :).

When you do launch, here are some things to look for that you are on the right track.

  • Emotional reactions, love or hate. If you put it out on Hacker News and you get really positive or negative comments, that's good. You've hit a nerve with some sub-group of the population. You can work with that.

  • Requests for changes. If you're getting unsolicited feature requests that's a sign that someone cares about what you're doing. Remember each piece of feedback probably represents a lot of people who have that issue.

  • Interaction with your site. If your analytics say that some people are spending a while on your site, that means that sub-group is intrigued enough to give you some of their attention. That's great.

  • Repeat visitors. Same.

If you launch and you see any of these things, don't give up yet. Instead, you need to do customer development. You have the potential for product/market fit. Now you need to talk to these people who are interacting with your site, and figure out how to build on what is working.

A lot of people give up in this stage when they shouldn't. Yes, some of the above happened, but it still feels like an uphill battle to get more. You have visitors trickling in that you may be able to count on two hands.

What you have to realize though is that you probably don't have a powder-keg idea (and never will), because they're super rare. It's more of a movement where you need to cultivate your followers and pay attention to minute details. 

At my last company, I saw the above signs, but it is was another six months of little tweaks before it started to really take off. At Duck Duck Go, it's been a slow and steady climb from the beginning. I think it would have been easy to give up on either of these projects if not looking out for these signs and using them as motivation and validation.

When you're farther along, here are some signs you're really on to something.
  • People offer to help you for free. 
  • People make add-ons to your site on their own.
  • People ask if you're hiring. 
  • Pople ask to invest in you.
All of these are signs that either you've made a real connection with customers or that people believe in your vision and market potential. 

Yes, you should be vying for product/market fit and doing anything to get there. However, many people give up because they don't immediately get there. Don't do that. Look for the markers that you are on the right trail.

There are east coast (below Boston) hackers--draw us out...

 
I just read Matt Mireles's new post, The City of Founders Without Hackers, where he calls out NYC as having a startup-friendly developer shortage. I also read Tobin Schwaiger's response, A City of Technical Founders Without Non-Technical Talent, where he says the problem lies more with low quality non-technical founders.

I have a lot of experience with these issues in Philadelphia, and I'd like to offer a third perspective. I moved to the Philly area four years ago and knew literally no one in the entire city. I came from MIT, where I was part of a very startup/hacker-centric community for the previous nine years.

It doesn't take that many people to make a difference. The biggest thing I've learned is that we're talking small numbers here. Early startups need one or two hackers, and so with a hundred good hackers you've actually got a lot of potentially good startups. 

NYC metro is 19M people; Philly metro is 6M. When I first moved here my premise was that in a city that big there has to be a hundred good hackers. All we have to do is draw them out. I think Matt is onto this:

On one hand you have what is becoming a relatively tight network of business-minded founders & investors; on the other you have a loose federation of hackers who seem to be not only disconnected from the business people in the community but also weakly networked amongst themselves. This is a HUGE problem. 

Hackers don't want to go to traditional networking events. We're a largely introverted bunch (or more specifically, INTJs & INTPs). These personality types hate bullshit and phoniness and a lot of stuff you see coming out of wannabe entrepreneurs at traditional networking events. And in general we don't like to "network" for the sake of networking anyway--we like to talk about hacking. 

So really the problem is not that we don't exist or that good non-technical founders don't exist. It's that both groups are small in number and we're not meeting.

As a hacker myself who didn't know anyone here, I made it my goal to draw out fellow hackers. After some fits and starts, I finally settled on a monthly startup hackathon group. We've met 26 times, and last month was widely regarded as the best yet.

The group and its events have no agenda--no scheduled speakers or whatever. It's just a time to get together and hack or talk about hacking/startups or to get feedback/help on projects. 

The first Philly startup hackathon had 4 or 5 people that I got from posting on news.yc. It stayed that way for a while. I looked on craigslist and emailed some people from there (startups looking for hackers or co-founders). And then slowly it grew and grew to where it is now. Currently there is 160 people in the group and 10-40 show up at each event (who shows up varies by location, time of day, etc.).

And this is not the only group or place in Philly to meet hackers, though it is the only one I know focused on startups. You can also find hackers at technology meetups and at co-working space. As NYC is a much bigger city, I'd be really surprised if there were not equivalent groups. I know a little about Baltimore and Atlanta, and I know there are similar groups there as well.

You can't just show up and snag a hacker. It's a relationship building exercise. Instead, you show up month after month and get to know everyone. You keep tabs on what everyone is doing. And eventually, and naturally, it all plays out. 

I've personally hooked up a bunch hackers and startups over the past couple of years. Once you're in the hacker network, people like myself will connect you to other hackers. But you have to have a presence. And patience.

If there is no good hacker group to get such a presence, then start one. I recommend a format like the one I described above, i.e. no agenda monthly hackathons.

Twitter RT Test Results

 
Test: I asked @duckduckgo followers to RT this tweet.

tweet1.png

I also RTd it from @yegg (my personal account) with slightly different text.

tweet2.png

Hypothesis: I wasn't sure what to expect, but figured I would get a bunch of RTs because my followers seem pretty solid (not spam, auto follows or other non-sense).  After that, I thought maybe I'd get some 2nd level RTs. I wasn't even holding out hope it would go viral, and of course it didn't.

Results: I tallied up the RTs using twitter search. The @duckducko tweet was RTd by 18 people using Twitter's RT system. The @yegg tweet was RTd by 6 people. Then there were 11 people who RTd it on their own, 4 of which got RTd 1 time each. This totals 39 RTs.

All of these people for the most part aren't spammy either, i.e. their accounts look real, with real followers. Counting them up they had 4,406 followers (avg 126, min 6, max 408). I threw out one outlier who had 22,150 followers but was also following 24,308.
 
As you can see from the tweet, I linked to a special URL, dukgo.com, that hardly anyone uses, and so I used it to track clicks. All told, 73 clicks. So that's ~2 clicks per RT--not too good.

I also considered RTs/followers. @duckduckgo has 848 followers, so that's 3.4% who RTd it. At the second level you have the 4,406 followers tied to those RTs plus my 911 followers for 5,317 followers. At 10 RTs, that's 1.8%. So you can see the drop off.

Here are my takeaways:

  • You need a lot of followers. I'd say you'd need two orders of magnitude more, i.e. 100K real followers, to make this at all worthwhile. Then we're talking on the order of 10K clicks.

  • To go viral, you'd need RTs by important people. I'm really grateful for all the RT support, but no one had a ton of meaningful followers. I think you'd need that celebrity push to get it out there, which may kick off other celebrity RTs.

  • To go viral, you'd probably need more compelling content. Of course this test was business related, but you'd probably need it tied to either more of a fad/news story or have more of a hook, e.g. a super interesting Web page on the other side.

  • Viral coefficient is not 0. There were second level RTs. If the content also lent itself to RTing, i.e. it was a game or something that involved tweeting, you might be able to bring that up and keep the chain going.

I also tried Fiverr, a service where people say what they'll do for $5.  I spent $15 on 3 people who seemed legit and said they would retweet to all of their x thousand followers. Only one has done it so far, and that yielded a RT by 1 person (none of which I counted above). So I'm guessing that is not going to be a good advertising channel.

Will the real standard terms please stand up?

 
I got a lot of feedback from my post on liquidation preference (some constructive, some very critical). This post essentially addresses those criticisms and makes some new points about seed financing.

First off, I'm not a so-called super angel, which I've made abundantly clear from the beginning. I'm just an entrepreneur who is starting to get into angel investing. And even then I'm an outsider/outlier. I've never taken investment beyond friends and family for any of my companies. I don't live in the valley or Boston or NYC (Philly actually). And I'm very-technical hacker (the good kind).

I'm opening up my new angel thought processes in real time as I get into it. I think this approach is very rare and that there needs to be more of it, so I figured I'd start in hope others would follow! I realized from the get-go this approach would generate a lot of flack, but that is par for the course. I grew a think skin a while ago, mainly from stuff like this. Needless to say, however, personal attacks, condescending tones, and snide remarks still annoy me. I'd appreciate it if you kept those things off of my blog (and everywhere for that matter). 

Anyway, in effort to continue my thinking and engender some more discussion, I want to address the various criticisms that have come up in response to my first post. The biggest point, which Nivi made best, was that I was proposing a non-standard term, and non-standard terms should be rejected out of hand for process reasons (as they will hurt your deal flow, among other things). 

I want to make clear I was never suggesting I would insist on this term in mass syndication deals. In fact, I'm not insisting on it at all. I thought it was obvious I was talking in the context of a term sheet negotiation where I was leading, i.e. playing off of valuation, etc. I realize now that wasn't clear and I should have qualified my post from the get-go. You live, you learn. 

That said, the bigger point I want to make is...

  • Would the real standard terms please stand up? Everyone's talking about "standard terms" without naming them. I've personally been collecting angel financing agreements over the past few years and they're *all different*. When people continually make reference to this magical standard they're implying that 90+% of deals look the same, and I'm proposing something in that wacky third-rail 10%. 

    There is no 90%. For each term, there are certain leading types people use. But often the leading one within those types doesn't even get a majority. And that's not even addressing specific language (implementation of the term type), which definitely matters btw.  When you vary each term across the whole term sheet, each final term sheet looks pretty different.

    Every angel or angel group has their own lawyer, and in turn, their own docs they'd like to push. And those docs are all different. And not just slightly different; widely different. As I said, they not only vary on terms, but also on specific language for those terms. They also vary on side-agreements they want to you sign, e.g. employment/IP/vesting agreements, etc.

    Only in the past few years have so-called standard docs appeared on the seed scene. By my count there are four of them that people take seriously--Techstars, YC, Series Seed & Founders Institute--and they vary widely too! There's a great post on some of the macro differences between YC, Techstars & Series Seed. 

    For example, Techstars has broad-based weighted average anti-dilution whereas YC and series seed have none. Each has different protective provisions. Series seed has a right of first refusal, whereas YC and Techstars are silent on that point. Series Seed also has drag-along rights while the others don't. YC has no future rights, whereas both Techstars and Series Seed have different future rights. etc. etc. Bottom line is they're different. And none of those differences even addresses the macro convertible debt trend. (Even when they agree on a macro level, again, the micro language varies.) 

    Recently, Brad Feld tried to get these to converge into one set and then quickly gave up because basically no one could agree and had no compelling reason to do so. So bottom line, I reject the premise that there are standard terms in seed deals.

    I've tried to find hard data on it, and I'd welcome/encourage/love more experienced people to provide some. All I can find on liquidation preference in particular is WSGR's market data (part of their Term Sheet Generator). As you can see, it varies a lot. Yes, I realize these are Series A rounds, but it's all I have at the moment (besides the deals I've collected personally, which I said are all over the map). In any case, the VC world is being disrupted. VCs/angel groups who do Series A rounds are also doing seed rounds, and as far as I can tell they're bringing their terms with them, or at least some of them. 

  • There are good reasons there are no standard terms. Why can't anyone agree on magical standard terms? There are a lot of reasons, e.g. people have varied experiences (edge-cases they want to cover), people see different types of deals (even seed deals), and it's hard to get rich lawyer types to agree in the first place :).  Brad has another good post on why it's hard to agree to particular language.

    But there is an important underlying point here that all seed deals and their terms shouldn't look alike. That is, every deal is indeed different. Three examples are 1) each deal has a unique set of risks and sometimes you want specific terms to account for those risks; 2) some entrepreneurs and investors want specific terms, e.g. certain valuations, and so other terms need to crop up to come to an agreement (if both sides really want to make a deal); and 3) deals come in different shapes and sizes.

    On this third example, consider my case. I'm ideally looking to deals really early in really small rounds (<=250K) where, if everything goes well, there will be no need for follow-on VC. Those specifics change a lot of things with regards to terms, and I hope put my original post into a bit more context. This is another thing I should have made abundantly clear in my first post, but failed to do so.

    In that context, convertible debt doesn't make sense since there will ideally be no follow-on round. Additionally, a pile-on preference is likely to matter less, since if they do end up taking VC, it's likely to be a smallish Series A (or even another angel round).

    For the record though, I welcome standardization as much as it is possible. I don't want to pay any legal costs like anyone else (well, mostly everyone), and I'd love to use these converged standard docs, which is why I think...

  • My proposed docs are arguably more standard than you're going to find elsewhere. Like I said in my first post, I'd like to use the Series Seed docs pretty much *verbatim*. In fact, my proposal was literally to change one word in them, from a one to a two. I'm guessing other people using Series Seed are changing things here or there and adding side agreements on top of them. But more to the point, if you go out and try to get financing, you're, for the most part, not going to get open source docs that you can read ahead of time. I'd personally like to be able to offer those.

  • Liquidation preference is a negotiation wrt to the whole term sheet. As I said above, if you want to insist on particular terms, all you're doing is pushing all the negotiation onto valuation (and option pool size). That's fine I suppose, but you can arrive at better compromises in certain situations by opening up more terms to negotiation. That's what I heard from Fred in his post. He negotiates in various types of liquidation preference in certain situations, mainly when there is disagreement on valuation.

    OK that's all well and good you say, but 2x liquidation is still non-standard and no one wants it, to which I reply.

  • I'm not wed to a 2x non-participating preference. Several well known angels commented to me that they prefer other liquidation preferences (other than 1x, non-participating) including 1.5 participating (with a cap) or dividends. George Grellas (startup lawyer) said on Hacker News "in my experience, in all normal investment environments, founders will resit it for 1x only (usually going for participation with a cap)." First of all, this further goes to the point of there not being a universal standard. Secondly, I don't so much mind any of these choices. They're all basically the same if you run the different scenarios on them.

    I picked 2x non-participating as an alternative to participation, in response to Fred's comment (and my limited experience) about entrepreneurs hating so-called "double-dipping." I have also read that dividends can cause accounting headaches and in some cases on-the-books insolvency. So my response was basically summed up in: OK, people hate participation and dividends, and you can get to the same place using 2x non-participating, so why not go there? The insistence that it is somehow worse than those seems a bit silly to me. Is it for no other reason than you see that less nowadays? Btw, in Brad's post on participation he notes how the so-called "standard" has evolved and is different to begin with across coasts.

    Going further, why is that all these other things are allowed to vary across the docs, but people have some problem with liquidation preference in particular? Or is it just that I brought it up, everyone has different opinions, and people just react to things they're not used to within their experience? 

    But like I said, I'm pretty much indifferent across these economic terms. At this point I'm just intrigued.

  • I don't want the entrepreneur to take all the risk. A bunch of entrepreneurs were pretty angry that I had the audacity to even suggest making a term more investor friendly. Again, I want to be clear that this is in the context of an overall term sheet (including valuation) discussion.

    My point was that I'd prefer to give on valuation a bit to make it more likely to make a modest return on a liquidation event such that I can continue to make angel investments, which would of course be allocated to fund more entrepreneurs. I'm putting up my hard-earned money (I got from entrepreneurship btw), and, if five years down the road, the company liquidates assets for some small amount, I don't think it is unreasonable to get a small return on that investment. 

    Let's consider two cases. I was essentially advocating for approximately a 15% nominal return. The real return would be less due to inflation. A 1x straight preference (non-participating) would be a 0% nominal return, and it's really significantly negative, again because of inflation.

    The second case is if you just got common, which a bunch of people said they would insist on. In that case, I would essentially lose most of my money whereas the entrepreneur would get most of the sale price (in this liquidation scenario). I'm not saying I know for sure what is fair, but that doesn't seem fair to me. And more to the point, I'm not going to invest in anything for just common. As Fred said in his post's comments, "i think everyone who has cash at risk in a deal should have preferred stock. i have no idea why angels do these investments in common.".

  • I hope the liquidation preference never triggers. Contrary to what some people implied, I never want the preference to trigger. People seem to think at a 2x, nonparticipating preference (in a small round) skews incentives so much that I would want the entrepreneur to take a small exit. Nothing could be further from the truth. I only want to invest in things I think will have a decent chance of a medium sized return, i.e. one where this preference would never come into play because everyone involved will have done very well.

    That was sort of my point about choosing this value. Sure, weird preferences in certain situations can skew incentives, but I don't think this is one of them, especially in context of the deals I want to do. I'm considering very small rounds where the preference amount wouldn't exceed 500K total.

  • I'm looking for investments that, if all goes well, won't need VC. Of course, this could be a terrible strategy and I'm open to changing it, but that's what I'm going for right now. The most insightful argument I think was from Mark Suster, who said in the long-run this is bad because later investors will pile-on preference. Other people suggested every one should just be using convertible debt. I'm not saying any of that isn't true or great, but it doesn't work for this current strategy. Like I said above, when there is no VC, convertible debt doesn't make sense. And also, there is less of a risk of screwing up an eventual VC round.

    The point for the preference discussion in the first place, which no one really commented on, was that new angel investors could make more deals, and that's good for the ecosystem. Not that people should make personal decisions based on externalities, but I do think on a macro level more angel investments should be encouraged. So in small deals where there is no need for follow-on VC, why shouldn't the standard be for a modest preference to allow for those angels to do more deals?
All that being said, I'm really not that obstinate about any of this. I'm thinking out loud so of course my thinking is likely to evolve over time. Thank you all for the continued constructive discussion.

What I installed (and uninstalled) on my new computer

 
I just bought a new desktop. Right before it arrived I was listening to Chris Dixon on Mixergy opine that Skype probably couldn't happen today because people don't trust downloads anymore. Somewhat tangentially it got me thinking that people probably download less software now as well because of the ascendence of the cloud, and I probably don't need to install/download much software on this new computer. 

I'll let you decide whether that was the case or not. I wrote down everything I installed (and uninstalled), in order.

  1. Windows updates. I had to go through a few rounds of rebooting to get them all installed. That's pretty amazing (and annoying) since Windows 7 just came out, but whatever.

  2. Google Chrome. My Web browser of choice (at the moment). I love how it syncs my bookmarks now too, which is one of the main reasons I installed it first.

  3. Adobe Acrobat Reader. First thing I did was check my email and someone sent me a PDF. Really, you can't pre-install a PDF reader? There might be a better one to install, but I just went with what I know.

  4. Adobe Flash Player. Next thing I did was go to a Web site that required flash...

  5. Skype. I use Skype all the time, especially to enable video chat between my son and my parents.

  6. Vodburner. This is a Skype add-on that I pay for to help me record Skype video chats for my traction interviews. While I was installing Skype I figured I might as well get this set up too.

  7. Nvidia GeForce GTX 260 drivers. I have two 28" HannsG monitors (HG281) at 1900x1200 resolution (1080p). Yet they were rendering at 1920x1080 and everything was blurry. First it took me a long time to figure out the resolution was wrong. Then it was was really annoying to fix because it wouldn't let me set a custom resolution. So I went to the nvidia site, which sent me to the hp site and their their download said it wasn't compatible with my computer. So I went back to the nvidia site and found the latest drivers. After install, everything worked fine. Side note--now Windows wants to install an "update" of the old drivers. I told it to "hide" that update :).

  8. Putty. Once the text was clear, I wanted to check something on my servers. I use putty for that.

  9. Sonos Desktop Controller. This whole time I was listening to Pandora over my Sonos system. A sucky song came on and it was clear it was also too loud. So I installed the controller that lets me control the music in the house from my desktop.

  10. PGP Desktop. I keep my passwords and other important docs on an encrypted virtual drive that gets mounted as a regular drive by this software. I needed my passwords for facebook, twitter, etc. (I use random passwords), so this was next.

  11. Adobe Illustrator CS2. I have a folder for software to install with this and partition magic in it as I purchased both of them via downloads a long time ago. I saw it next to my PGP folder, so I went ahead and installed it next as I know I'll need it soon enough.

  12. ITunes. Eli (my son) watches videos through here, and it syncs my iPad and iPod Touch, which I use for development.

  13. Firefox. I debug stuff in Firefox, and got an email about a bug, so I decided to download it next.

  14. ForecastFox, Web Developer, Firebug, YSlow. These are the Firefox add-ons I use regularly. While I was installing Firefox I thought I'd go ahead and add these.

  15. Safari. I use it just for testing. But while I was doing Firefox I thought it would be good to just do now.

  16. Opera. Same story.

  17. Uninstall Norton Stuff. Ugh, I hate this stuff and wish I had an option not to have it pre-installed in the first place.

  18. ClamAV. This is my replacement for the virus part of Norton. The other parts I'm fine with the pre-installed Windows firewall and Windows Defender. I have a smoothwall setup in my house for more firewall protection.

  19. WinSCP. This is the other piece of software I use to routinely connect with my servers (for transferring files). I needed to transfer an image, so this was next.

  20. Quickbooks 2008. I went down to the basement to get CDs. This was one of them. I use it to do company accounting.

  21. Adobe Photoshop Elements 5.0. On CD. I use it to do image manipulation. Great deal actually--I've found I've never needed more than this "elements" version.

  22. Picasa. I manage my photos in picasa. Photoshop made me think of it.

  23. Adobe Premier Elements 2.0. On CD. I use it to edit video sometimes, though I don't recommend this program. I just don't have a good alternative at the moment.

  24. Vmware Server. I use it to develop with. I have a FreeBSD image that mimics my servers. I wanted to fix some bugs so this was next.

  25. Uninstall Microsoft Works, Office Home and Student Trial, PowerPoint Viewer, Compatibility Pack for the 2007 Office System. I wanted to install Office 2010 (from BizSpark), but it wouldn't let me install the x64 version before installing all remnants of x32 versions (this stuff). I find this odd since I have an x64 version of Windows--so why would they pre-install x32 versions.

  26. CutePDF. While the uninstalling was going on, I needed to PDF something (I save receipts this way).

  27. WinRAR. Someone emailed me a giziped file, and this is my decompressor of choice.

  28. Microsoft Office 2010. Once that other office crap finished uninstalling, I installed this.

  29. Gmail notifier. Alerts me of new emails.

  30. Gmail notifier https patch. Come on--are you ever going to update the notifier to include this natively? I use https gmail and it doesn't work with notifier without this patch.

  31. VNC. I use this to connect to my desktop from my laptop (usually to get a password). When I went back to my laptop I noticed it was missing :).

I could be an outlier, but that sure seems like a lot of software to me! All in all it was ~20 downloads and 3 CDs. If I wasn't a developer at all, I think I'd still have done ~10 of them.

Liquidation Preference (my case for 2x, non-participating)

 
I recently closed my third angel investment and the first one I led from beginning to end. Leading made me think a lot about deal terms. We ended up using slightly modified Series Seed docs, which I plan (and would love) to continue to use. 

The one term we slightly modified from the model term sheet was the liquidation preference. As I seem to be at odds in this change with people much more experienced than myself, e.g. Fred Wilson and Chris Dixon, I want to lay out my thinking and hopefully get some feedback.

First, why would I ever differ with these guys? I think it's simply because I don't have as much money to invest, and as a consequence, my angel investing strategy is different. 

These guys are so-called super angels. And if I had more money I'd probably end up being one too. But right now, I have a limited pool of angel money. It's enough to do a bunch of deals and I'm fine with losing it all, but it isn't enough to do the YC/Conway (super angel) strategy.

Here's my basic argument. Suppose you have a "hit" rate of 10%. I realize it's more complicated than that and I've developed a spreadsheet to model it better, but for sake of simplification just go with me for a moment. 

At this rate, every deal you do has a 90% chance of not being a hit. People then think intuitively, OK, I need to do 10 deals then. Nope--the default intuition is wrong. What's the probability that you do 10 deals and don't get a hit? (9/10)^(10) =~ 35%!!! That's a massive %. At 20 deals you're down to 12%, but that's still pretty high.

So basically you have to do a lot of deals if you're at all serious about making money off of angel investing. A recent Monte Carlo simulation by Jeff Miller confirms this thinking. He concludes 20 deals is a sweet spot. 

If you're a new/small angel like myself with a fixed investment pool, what can you do to maximize the # of deals? Two things really. First, make smaller investments--though this isn't great because there's a lower limit where you'll get shut out of deals ($25K is probably the realistic minimum) and, when you do hit, you won't get that much back.

Second, liquidation preference. In my previous post on angel investment scenario planning, I noted that small returns (1-3x) don't influence overall deal pool IRR that much. And that's true in a static scenario. But the real world is dynamic and if you can do more deals you can increase the chances of hits, as noted above. Getting money back through liquidation preference won't move the IRR needle, but it will enable you to do more deals. And to the small angel (me) that is the most important thing.

If you have no idea what liquidation preference is (and you want to know!), check out Brad Feld's two great intro posts on the subject. The default term from the Series Seed Term Sheet is essentially "One times the Original Issue Price." That means in an exit event, the investors have a choice to a) get their money back or b) get their equity %. It's an economic choice. If it's a small exit and the equity % doesn't yield as much money as was put in, you'd opt to get your money back; otherwise, you'd opt to get your %.

Tweaks on this model are:
  • Get multiple times your money back (e.g. 1.5, 2 or 3 times) in choice a.
  • Get your money back first, and then also participate in the distribution of what's left based on your %, which is called participating preferred.
  • Participation, but with a cap. In this one you participate up to a limit, like 3x. So you get your money back, and then your % allocation, but if that exceeds 3x your money you have the choice whether to get that or your straight %.
Anyway, entrepreneurs seem to hate this stuff, especially the participation part. Some people call it "double dipping" because you get money back and then you double dip and get more money back out of what's left. I think this is somewhat of an understandable knee-jerk reaction, but everyone should look into different realistic scenarios to see what the actual implications might be. 

I woke up today to read Fred Wilson's evolved view of liquidation preference, which is that he used to insist on participating preferred, but now only does (and with a cap) in certain specific situations. 

Apparently Chris Dixon agrees with this view, and goes further saying "[a]nything more than that (participating preferred, multiple liquidation preferences) divide incentives of investors and the entrepreneurs. Also, this sort of crud tends to get amplified in follow on rounds."

My current thinking is that a 2x liquidation preference is a sweet spot. Let's look at a concrete example. Here is a spreadsheet I made where you can play around with it and other scenarios at your leisure.

Suppose you put in 50K in a 500K round on a pre-money of 2.5M (post-money 3M). You end up with 1.67%. Let's say the company doesn't take any more investment (for simplicity sake) and in year five it exits/liquidates.

In scenario 1, you got the normal 1x liquidation preference. In scenario 2, you get 2x. And in scenario 3, you get 1x, participating. Here's how those scenarios play out in terms of IRR at different exit sizes.

 1M - 0.%, 15%, 6.%
 2M - 0.%, 15%, 13%
 5M - 11%, 15%, 22%
10M - 27%, 27%, 34%
20M - 46%, 46%, 50%
30M - 58%, 58%, 61%

At high exits, the term doesn't matter much--they converge pretty quickly. It clearly matters <=5M (at least in terms of IRR). With 2x liquidation preference, your IRR jumps from 0 to 15%, a healthy % that you can use to re-invest (and do two more deals at the same size). And interestingly, you can see the participation right does worse than 2x at small exits, does better in the middle, and smooths out at the high end.

Now let's take the cons in turn.

  • Splits incentives. When you get crazy preferences, this can certainly happen. And if you have participating with a cap, as Brad notes in his post there is a plateau range in the middle at which the investors are indifferent to exit size, which is weird. But at 2x, non-participating, in a smallish round (<=500K) I don't see much issue with regards to incentives. The preference only kicks in for low exits (in this case sub 6M). The entrepreneur should not be getting financing if they would be happy with an exit lower than that.

  • Mucks up later rounds. There is a risk here in that, at the very least, you can expect later investors to get similar preferences. Brad points out scenarios where you can actually lose out as an investor by having these rights (if the next investors get them too). Again though, if they get limited to 2x, non-participating, I think that is a manageable risk.

  • They'll get thrown out anyway. I've heard this, but don't have enough experience to know if this happens or not. Ideally you'd also get some control rights on the next financing so at least you can be sure you won't get screwed in that next round.

  • IRR doesn't matter for low exits. It matters to me :). In particular, as stated above, if it did return the preference amount, this would yield the ability to do 2 more deals instead of 1. A few of those, and an initial pool of money allocated to 10 investments may easily stretch to 20.
Update: some good comments on HN

Update 2: I made a new post explaining more of the background on this post and addressing the comments. 

About Me

RSS.