The Other Stuff That's Not Product That You Need To Build Early

One of the things that I didn't think about before I started my first startup were all the parts of the infrastructure that weren't product.  They've also consistently been the things that other entrepreneurs forget about when they come to me with their "ohmygodIhavethisidea" ideas.  They're also vitally important because they allow the founding team to actually execute the pivots you'll need to execute quickly, intelligently, and without sacrificing development of the actual product that you're trying to pivot.

The first part of this infrastructure is the admin panel.  This admin panel is what will allow you to do troubleshoot for customer support, engage in administrative actions, and audit your payments in and out of your system without having to mess around with your production database.  I cannot impress upon the first-time web entrepreneur just how important it is to have a robust and well-functioning administration system.  When you're building your administration system, it is important to make sure that you can disable and delete users, view and change payment regimes, and reset user passwords (because they will invariably not be able to find your reset password form and they *will* e-mail you).  If your product takes off and you find yourself bringing on board a business person, they're going to be tasked with support, and you do not want whoever is doing support interrupting engineering to complete basic Level 1 and 2 customer support.  If your product really takes off, you may find yourself bringing on board a dedicated support person.  If you wait until traction to build out your administration panel, you're going to be ceding ground to the competitors that spring up when they see your traction.  Django's admin panel is so well developed that you may not have to build your own, and is a reason to strongly consider the framework if you're familiar with (or interested in) Python.  Even as your product pivots, the administrative support tasks generally don't.  

Make sure you've built a functional admin panel before you launch.

The second part of the infrastructure you're going to want to build (or implement with WordPress, Melody, or some simple bare-bones CMS plugin that someone has already released for your framework) is a content management system (CMS).  It should not take a push of new code to change the static content on your website.  Holding up basic changes and updates to your about section, marketing pages, terms of service, privacy policy, and help section because you need to wait on some guy to merge his branch into master (or check into trunk, or whatever your VCS calls it) is silly.  You don't want to have your marketing person or intern create some hot new copy then have to hand it off to engineering to copy/paste it into code and markup each and every link, line break, and image to get it live on the site.  It makes the business side frustrated with the pace of changes and the engineering side of the house gets frustrated at having to do such menial tasks when they could be building some crazy awesome technology that is your core long-term competitive advantage that you brought them on board to do.  Again, some web frameworks have bare-bones CMSs or templating engines built into the system out of the box.

Make sure you've built (or implemented) a content management system before you launch.

The third part of the infrastructure you need to build is a good metrics dashboard and logger.  The standard piece of advice is "pick a few metrics, but log everything".  The idea of logging everything when you've decided to focus on just a few Key Performance Indicators (KPIs) can often cause heartburn.  But it's relatively simple; build a separate table (or database) that automatically logs all the various things that users do.  Don't worry about duplicating writes that you're doing elsewhere in your database.  Having a separate table (or better, database) will save ou on reads and JOINs and whatever else when you need to dig in.  The one thing the logger needs to do is be *auditable* - that is, you need to be able to say that user_id=8273 did x on page one, then y on page two, then z on page three.  These loggers are so important that there are venture-backed startups that exist solely to help other startups collect these logs and make sense of the data they're collecting.  The KPIs are, of course, actionable metrics that exist in the aggregate, but the ability to see what any individual user has done will help you troubleshoot, audit, and make hypotheses for new features and functions.  Also, one quick note: Google Analytics aren't auditable at the user level nor actionable in helping you make forward-looking hypotheses.  Everything in Google Analytics is backwards looking.  The product is free as in beer and all-but-free from an engineering cost perspective, which is great.  GA does give you the ability to find exception cases and the basic funnel visualizations can be useful for getting conversion metrics up and running quickly, but don't kid yourself - Google Analytics is mostly data porn.  

Make sure you build a real metrics dashboard and implement a real logger before you launch.

Product Development Starts With A Hypothesis, Not A List Of Requirements

For a startup to be successful, the founders (or whoever owns product) need to have strong opinions about both the market problem and the product solution.  Sometimes - maybe even often - this means ignoring your users.

That doesn't mean that you don't listen to your users - even companies as insular as Apple and Nintendo do some work reaching out to others.  Apple designs for others; they just don't do formal market research. Nintendo has "random employee kidnappings".  But if you don't have the luxury of having the singular, strong editorial voice that Apple gets from Steve Jobs or that Nintendo gets from Shigeru Miyamoto (or that 37signals gets from Jason Fried), you probably have to set yourself up to conduct lots of experiments.

Do you remember your scientific method?  If not, here's a really, really good primer on the scientific method.  Never mind that it's written for children - it's good.  Here are the steps:

Ask a Question
Do Background Research
Construct a Hypothesis
Test Your Hypothesis by Doing an Experiment
Analyze Your Data and Draw a Conclusion
Communicate Your Results

If you've noticed, we've actually done the first two steps.  Asking a question is finding a market you're interested in.  Doing the background research is figuring out the problem statement.  The next step, where we are now, is to construct a product hypothesis.  Remember, when you test hypotheses, they're falsifiable - you only can find out what *doesn't* work for sure.

What this means is that you have to keep pivoting until you find yourself in product/market balance.  Too bad that every product hypthosis and every pivot costs you in time, money, blood, sweat, tears, and all the other various inputs.  You can't iterate forever; you're going to run out of one of those inputs (most likely money).  

The only way to reduce your time in this cycle is to make a really good hypothesis from the get-go.  If you're not playing with other people's money, then you better make sure that you don't have to pivot at all.

Listen to your users on the problem statement - they can feel that pain.  But listen to your judgment on the first product hypothesis - there's a reason you want to solve this problem, and it's not because you're a stenographer.  It's because you see something everyone else doesn't.  Asking better questions allows to more clearly see what others - even your users - may not.  Here are some better questions to ask:
  • Are your customers time poor or just frustrated with the time they have to spend on a task?  
  • Are your customers cash poor or not seeing value for their money?  
  • Are your customers status poor or are they embarrassed to use your product?  
Customers often will tell you one answer, but they often mean another.  It's your job to 1) ask better questions, 2) read between the lines of the answers, and 3) use your judgment to make a really good product hypothesis.

Dropbox doesn't save a large amount of time; but it makes saving/retrieving files thoughtless.  RC Cola just didn't have the status of Coke (or Pepsi); you can always get store brand pop if cash is king.  

If you better understand the true pain, you'll make a better product hypothesis.  Dropbox is great because it "just works", but it took Drew to define what "just works" meant above and beyond Mozy, Carbonite, XDrive, and every other storage in the cloud provider.  RC Cola "tasted better" (they won every blind taste test they sponsored, supposedly), but it turned out that people didn't pick their pop based on taste.  They picked it based on status, nostalgia, inertia, (and as the supermarkets found out) price as a tiebreaker between Coke and Pepsi.  Remember, your users may lie to you.  

Users may not know what they want in a product until you show them.  But they always - always - know their problem.  

(That's what Steve Jobs is so damn good at.  He keeps using Apple's prototypes until the stabbing pain of the problem Apple's identified goes away.  Then and only then does he bless a new device/program/feature.  See copy/paste on the iPhone.)  

Your judgment on how to solve the (real, maybe unstated) problem of your potential customers is your initial competitive advantage.  Your ability to get a product hypothesis out there that your users don't immediately reject (and hopefully would be terribly disappointed if you took it away) is what gets you paid the big bucks (when you exit).

Because You Hate Your Family, or, Stuff To Read Over The Holidays

Let's face it - you love your family, but that's today, before you see them.  By Boxing Day, you're going to want to curl up in the fetal position as your uncle yammers on about the Cleveland Browns.  As long as you're hiding out in the basement, or drinking alone at the depressing corner bar, you might as well do something productive.  Here's a special "must read" list of books, blogs, and presentations while you stew about in the hellhole your parents call home.

Steve Blank - the granddaddy of them all.  Steve coined the term "customer development" as a parallel process to product development.  Steve's blog is fascinating (the man has done some very interesting stuff), but the must-read is his book, Four Steps to the Epiphany.  In particular, it specifically talks about how to work when you're disrupting an existing market versus when you're building a new market for your product.  You have to read Four Steps first.

Eric Ries - the new hotness.  Eric took Steve Blank's class, learned about customer development and had the insight to pair the customer development process with extreme programming to create a continuous improvement/feedback loop.  Eric coined this process "Lean Startup".  Eric's blog is great, but he has a MVP beta of his posts in an e-book, which may be easier to read.  You should buy it now, since Lulu might take a while to ship.

Dave McClure - the pirate.  Dave's AARRR model is the best compilation of every driver you need to measure.  You can't just slap on Google Analytics and think you're done, folks.  He's kind enough to post new versions of his Startup Metrics for Pirates presentation as he revises it (video of an older presentation), but Dave's blog (while a bit all over the place) is chock-full of great insights.

Sean Ellis - the Glengarry leads guy.  You've built it; now what?  Sean only works with companies that have raised venture money and already achieved product/market balance (see why I don't call it product/market fit).  You don't even have a product yet.  Why is he a must read?  Because his approach to measuring product/market balance is better than anyone else's.  His startup pyramid tells you each necessary step before you can turn on the growth spigot.  Sean's blog is updated less frequently, but every post is great.

Andrew Chen - the savant.  Andrew's spent a lot of time thinking about viral loops.  Virality isn't something that just happens to great products - although that does happen from time to time - virality is something that can be baked into your product's DNA.  Andrew's blog is full of posts that go into viral loops in detail, and it's much easier to read now that he made a special categorized list of posts for you.  You can't possibly read all of this before you come home, so I'll stop there.

Requirements Gathering Is Not Customer Development - But It Does Define The Problem Statement

Customer development is all about testing your hypotheses.  This is the Lean Startup diagram by Eric Ries that you (should have) seen before:

Let's start at the beginning and focus on the Customer Development Process, and in particular, the Customer Discovery loop at the very beginning:

Steve Blank has defined the Customer Discovery loop as four parts: Author creates Hypotheses; Author Tests Problem Hypothosis; Author Tests Product Hypothesis; and Verify, Iterate, and Expand (slide 11):  

The problem that I've seen time and again is that people fall in love with the process and get so caught up in the notion that they can just iterate through their hypotheses that they create terrible hypotheses.

This is dumb.

People - especially smart people - love the idealized notion of "iteration" but, in practice, it turns into "throw shit against the wall and see what sticks".  Then they realize the hypothesis testing iteration process takes time and soon, they start to believe they're making progress merely because time has progressed.  Then they start product development too early.  Then they... well, you can guess what happens from there.  Don't do that.  Spend just a bit more time up front.  Understand the problem you're solving first, so you can make better product hypotheses from the get-go.

Simplify the Customer Discovery loop into two parts: requirements gathering to define the problem statement, and hypothesis creation that you test with actual potential customers in the Customer Validation step.  Here's the difference between me and Steve: you don't iterate on the problem hypothesis - the problem statement is a fact that you have to suss out from your requirements gathering.  Your job is to suss out 1) whether or not there is a problem, 2) how big of a problem it is, and 3) how big the market is that's affected by the problem.  (You don't want to develop a product to solve a problem that won't return your investment of time and money.  This is the result of jumping into the "test product hypotheses" too early.)  These are all verifiable facts; the problem statement is not a hypothesis.  Anyone else going out to the same customer base and asking the same questions will come up with the same problem statement that you do.  

Defining a problem statement really isn't all that hard.  Sure, you can read research reports and get market sizes and all that jazz, but the one prerequisite you must do is really quite simple: talk to enough potential customers until you've reached some point of diminishing returns.  (You'll know it when you get there.)  Often, this doesn't take more than a dozen.  If you really want to be rigorous and pretend you've gotten to statistical significance, talk to 30 people.  Here's a simple list of questions:

* How do you do [x]?
* What's the worst thing about doing [x]?
* How much extra time/money/energy does it take to deal with the worst thing about [x]?
* If I could solve the worst thing about [x], how much money would you pay me?  (Note that if it costs money to deal with the worst thing about [x], theoretically you could charge up to 99.9999% of that amount.  Theoretically.)

Everything else is optional.  My trick is to keep them talking with one simple phrase: "that's really interesting.  Tell me more."  You may even (read: you will) get product insights from your problem statement questions.  

Eventually, you should be able to hone in on what exactly is the jabbing pain in your customers' eyes.  For salespeople, it was that they had to input all their information into ACT and then do extra work when they got to the office to share it with their manager (  For runners, it was that the shoes they had weren't really comfortable for running long distances (Nike).  

The iteration comes only with the product hypotheses - and I use the plural because you want to consider all the different product approaches that can solve the problem.

Let me finish with this: you don't define the Minimum Viable Product.  The market does.  At the beginning, your job is to form good product hypotheses that you test.  These product hypotheses spring forth from spending the time in requirements gathering to sharply define the problem statement that you are going to solve with whatever your product solution ends up being.  

If you've done the work to define a robust and accurate problem statement as a result of a rigorous requirements gathering process, you'll receive much better market feedback when testing your product hypotheses.  Eventually, when you start market testing your alpha product, you'll have a much shorter process getting those orders/eyeballs/whatever to determine you've hit your MVP.  

Next time, I'll talk about when to follow and when to ignore your customers when defining and testing your product hypotheses.

Bring Back The Evening Paper!

Commodity national news is dead.  Newspapers are dying.  The AP wire on Yahoo News (or Google's more heterogenous and more cluttered version) and are "good enough" that all other services providing "just the facts, ma'am" provide no incremental value.  Most observers recognize that this leaves a void in the local space, and predictably you see newspapers retrenching into their neighborhoods, fending off competitors like, EveryBlock, and ESPN's local sites.

But newspapers have so much more than just news, and that's why I love to read them when I visit my family in St. Louis.  I find value from the comics, the circulars, the coupons, the crossword, and, yes, the columns.

But why do I only read the paper when I'm visiting my mom and grandparents?  Why, the NYT and Tribune executives plaintively scream, why oh why don't I subscribe where I live?  

My biggest issue is delivery.  At my former apartment in Chicago's Logan Square, I had no faith that any paper I paid for would still my at my apartment building's doorway when I woke up in the morning.  I even subscribed to the free Saturday Redeye and it never showed up.  My new apartment is one of those old houses in a better neighborhood near Somerville's Davis Square, and I'm considering getting the Sunday Globe and/or Times.  But only on Sundays.

But here's the other delivery problem beyond just getting what I paid for: when the paper comes in the morning, I don't have time to read it.  When it comes in the morning, half of it is stale the minute it hits the stoop because I read that information online the previous day and the other half is stale by the time I do have time in the evening.  It's completely useless to me (again, except on the weekends, where my mornings are leisurely and my evenings are packed).

But this can be solved with a change to the content and a switch in delivery time.  An evening paper that focused on analysis and columns, rather than the stenography that passes for reporting, would be fantastic.  I could indulge with 15-20 minutes attempting the crossword; I could read the comics over a cup of tea (or a glass of bourbon or whatever your racially- and temporally-appropriate stereotype is); and I could browse through columns and analysis of new and interesting topics that aren't top of mind during the workday.  

Giving up all pretense of presenting the bare facts of news would free an evening newspaper from the tyranny of the mid-day deadline.  An evening paper as I envision it wouldn't compete with the evening news hosted by Brian Williams, Katie Couric, or Diane Sawyer.  It wouldn't compete for the advertising dollars of incontinence products and life insurance.  It would be more alive and it would be a better vehicle for television, movie, and other entertainment advertising than the morning paper or the evening news.  

A new evening paper would require shunting aside any pretense of being balanced in favor of presenting a wide range of opinions (I think the Atlantic does the best job of this on the national blogging/magazine side), but it could easily be done given the newspaper companies' existing delivery infrastructures and brands.  

And that's the way it is.

One related note, since this is being passed around this morning: I don't in any way believe we are nearing the end of hand-crafted content.  Just read Ezra Klein, Matt Yglesias, Felix Salmon, Ta-Nehisi Coates, or any of the other brilliant writers who pump out fantastic free content with real, actual reporting on a near-daily basis, and you'll realize that at the national level, there's a plethora of great, handcrafted content.  (BTW, are there any female writers - aside from Digby - who I should be subscribing to?  Megan McArdle lost me a while ago.)

Maybe it's different if you're checking TechMeme every 15 minutes to see who gets "credit" for "breaking" some "story" about some "gadget", but the stuff that I read is great.  Hell, just follow Paul Kedrosky; it's like the dude was put on this earth to create and share interesting shit that's perfectly up my wheelhouse.

Forget Product/Market Fit; It's All About Product/Market Balance

In my time watching and working with startups as an investment banker, VC, entrepreneur, and consultant to startup teams, the notion of product/market fit has gone from a new insight to conventional wisdom. 

However, product/market fit never seemed like the right notion to me. 

After doing some thinking and working very closely with a handful of startup teams in the last year or so, I realized that a team isn't looking for product/market fit; it's looking for product/market balance.  The notion of balance highlights the delicate task of getting a product to meet a market's needs:

Even if a team builds a robust, stable product offering, the product can fail to meet the market's needs and the product lands with a thud:

The notion of balance serves another purpose; balance provides a basis to extend the metaphor into the other important parts of a product experience: positioning and pricing.  Here, you can see how the product supports positioning:

Likewise, the product + the positioning supports the price structure:

If the product itself balances with the market's needs, poor positioning can upset the balance and lead to everything tumbling down:

And a poor price structure can lead to defeat even with the right product and positioning strategy:

You can't decide to position a product as premium if the product is shoddy.  Likewise, the price you choose reinforces and strengthens your positioning - you can't have a "value" product that is priced higher than your competition.

Startup teams must make sure that everything about their product, positioning, and price structure maintains the delicate balance necessary for market success.

Which States Have the Best Startup Environments?

In my last post, I crunched the data to see where Illinois ranked on a variety of metrics related to venture funding for startups. Turned out that no matter how you sliced the data, Illinois came out pretty poorly. This poor showing was a pretty strong indictment of the infrastructure, leadership, and general environment for startups here in Illinois - as far as I'm concerned, whether or not you want funding, being able to get funded is an indication of quality infrastructure and a strong startup environment.

So the obvious question is, if Illinois is so terrible, which states are good?

In my analysis of the data last time, I made two claims: 1) seed/startup funding is the best stage to evaluate, because those are the first dollars a new team is looking for. I believe that new dollars for a new team is the best indication for general environment, given the small dollars involved and the less company-specific nature of seed investing. (Nevada ties for dead last in the startup/seed data, no matter how you slice it, despite the success of Zappos.)  2) per-capita metrics, while a bit flawed due to network effects, are the best option we have, better than aggregate dollars or aggregate deals because per-capita metrics control for the "California is the biggest state, so it should see the most dollars" objection.

So, I re-ran the numbers to see what states had the highest per-capita dollars invested in startup/seed deals[1]. Here are the results for the one-, three-, and five-year spans ending 2007:

As you can see, there's a clear "step" between Massachusetts and California and the next tier. There's more than double the amount of money per-capita in Massachusetts and California in 2007 than there was in Minnesota and Washington state, the 3rd and 4th best on the list.

For the three year period of 2005-2007, just as for 2007 alone, Massachusetts and California dominate, with significantly higher per-capita dollars flowing to seed and startup deals than the next tier of states (from New York to Virginia, which has a range just slightly larger than the difference between New York and California).

Lastly, here's the data for the five-year period of 2003-2007. Delaware is skewed by very high per-capita amounts in 2003 and 2004 (the state had $23.60 invested per capita in 2003 and $49.05(!) in 2004 [2] ), but the rest of the chart is generally in line with the one- and three-year charts. Again, Massachusetts and California are heads and shoulders above the rest of the states.

In all three charts, you see strong showings from New York and Washington, which I wasn't surprised by. What I didn't expect was the strong showing by states such as Maryland and Utah. I know that Maryland may benefit some from federal money, but if that was so strong, we'd expect a stronger showing from Virginia and DC as well, which we don't. As for Utah, I know the Computer Science program in Utah is strong, and there are a number of technology companies in Utah (Omniture immediately comes to me), but still - it's astonishing. There must be something cultural in Utah that makes startups there more palatable than, say, Illinois.

When I talk to entreprenuers that want to start their own technology companies, the first question I ask is "do you have a co-founder?" The second I ask is "how committed are you?" The third I ask is "if you're so serious, why haven't you moved to California yet?" Turns out the data says that I should amend that last question to "why haven't you moved to California or Massachusetts yet?" (This, of course, assumes that you can syndicate a seed deal among angels in Massachusetts as easily as you could syndicate a deal in California. Again, I don't have angel data - see [1].)

Again, comments welcome on the data, methodology, and requests for other cuts of the data.

[1] Now, of course, the data I had was only about VC firms' funding - it was silent on angels. Knowing this was an issue with the PWC MoneyTree data, I've spent a fair amount of time since my last post begging for angel data. [1a] But, in the end, no one could find angel data to send to me. So, at this point, I've given up on angel data. (If you have some, please send it to me.)

[1a] I even managed to get promised an invite to the psuedo-TechStars demo day here in Chicago (it was actually an independent event, put together because so many of the Boston summer cohort had Chicago ties), but I never received it, despite a flurry of e-mails the day of. Many thanks to Brad Feld for trying to get me in; boo to whoever was supposed to send me the official invite. (I had time/location, but I didn't want to crash without an official invite.) I really wanted to get a headcount versus the Mountain View headcount and see who showed up. I was going to take that list and try to determine which Chicago-area angels actually had done deals in the last 12 months. As far as I know, Apex's seed investment in Appolicious is the only one in the last twelve months.

[2] Anyone have an explanation for these two outlier years? Delaware comes in at $0.00 per capita for 2005, 2006, and 2007 seed/startup deals, so I'm guessing just a handful of deals - maybe even just one or two deals - skew this chart, especially as Delaware is a relatively low-population state. The PWC data I got from SSTI is aggregated; I don't have access to the Thomson database with the deals themselves.

Illinois' So-Called Tech Leaders Have Failed The State

So, Chicago's lacking in tech leadership, says a columnist in the Chicago Tribune.  No, it's not, respond the heads of various tech organizations in Chicago and the State.  As someone who's founded a startup here in Chicago, I'm certainly more inclined to agree with the Trib columnist than with the panel.  However, I'm also a policy wonk - so I decided to take a look at the data to see who's right. 

After examining the data, it is clear that the leadership at DCEO, the ITA, World Business Chicago, IBio, the NanoBusiness Alliance, and the ISTC has failed.  While the organizations that signed the rebuttal may do many things, as they list in their bullet points, it is clear that the outcomes one would find from successful leadership just are not there.  What they're doing is not working, and there are no signs of improvement in Illinois' technology competitiveness, no matter how you slice the data. 

The best independent, objective data I could think of was venture funding data for startup companies[1].  The assumption, of course, is that venture funding is very highly correlated with technology leadership and serves as a legitimate proxy.  The focus on startups is because organizational leadership can most help the smallest of companies and early entrepreneurs as they jump into the deep end.  (However, I've presented the overall data as well to avoid charges of bias.)  Employment figures are fuzzy - you probably don't want to include CDW salespeople in a list of technology employees, for example.  I gathered state-by-state data from PricewaterhouseCooper's MoneyTree Report.  My source of data was the State Science and Technology Institute (SSTI).  SSTI's Venture Capital data site further breaks down the MoneyTree into easy-to-use state-by-state data[2]. 

You can see the data for Illinois at the SSTI page for Illinois.  The data show that Illinois companies have raised roughly $440 million per year since the last bubble year of 2000, which sounds decent, until you realize that Illinois companies only attract, on average, 1.59% of all VC dollars - across all stages - raised in the US.  The average number of Illinois VC funding deals per year for the 2001-2008 period is 70.5, which is just 1.99% of all deals in the States.  (Here's the page for the US data.)

These numbers aren't any prettier for Illinois startup companies.  For the 2001-2007 period (there is no by-stage data for 2008 available on the SSTI state-by-state breakdown pages), Illinois startups raised a total of just $56.7 million, a pittance of just $8.1 million per year.  It's even worse on the per-deal statistics: Illinois had a total of just 32 startup fundings for the entire seven-year 2001-2007 period.  That's an average of just 4.6 startup fundings per year.  The total startup funding across the entire US for the 2001-2007 period was $5.6 billion, with a total of 1,845 deals over the seven-year span.  That's an average amount of $740 million per year and average number of 264 startup fundings per annum across the entire United States. 

That means that Illinois companies, on average over the seven years, raised just 1.74% of the seed money and did just 1.09% of the startup deals done nationwide.  This, despite the fact that Illinois has 4.24% of the US population.  For just the smallest startups that organizations like DCEO and the ITA should be helping the most, Illinois is getting less than half of the dollars and doing less than a third of the deals that you would expect for a state of its size.  (It's still less than half for dollars and deals overall, as well.)

The numbers are just as embarrassing when you look at Illinois relative to the other states in the nation rather than just the United States as a whole:

(Slide 4)

(Slide 2)

Illinois ranks an average of 16th among the 50 states, DC, and Puerto Rico in startup dollars for the 2001-2007 time period.  Illinois did a bit better and averaged a rank of 13th in startup funding deals for the period. (Illinois' median for dollars was 14th and its median for deals was 11th.)

However, again, these figures favor the larger states over the smaller states; i.e. you'd expect California to be the leader in both categories (and it was) since California has the most people.  To really compare, we must look at the states on an apples-to-apples basis by looking at the per capita figures: 

(Slide 3)

(Slide 1)

It turns out that the per-capita figures are even worse for Illinois than the non-normalized figures.  Illinois averaged just 20th amongst the 50 states, DC, and Puerto Rico in startup dollars per capita for the 2001-2007 period.  Illinois averaged a pitiful 23rd in startup deals per capita for the 2001-2007 period.  (Illinois' median for per-capita dollars was 18th and its median for per-capita deals was 24th.)

States that consistently beat Illinois in the rankings include Colorado, Texas, and Washington, to say nothing of California and Massachusetts.  However, even states like Maryland and Pennsylvania consistently beat Illinois' rankings on all four measures - startup dollars, startup deals, startup dollars per capita, and startup deals per capita - year after year after year.  On a per-capita basis, states like Connecticut and even New Mexico consistently outrank Illinois in dollars and deals.

By the way, Illinois' average rank in dollars for 2001-2008 (remember, SSTI has total figures for the 2008 year) - regardless of stage - was 13th, and its rank in deals was also 13th.  Again, the average per-capita rankings - the better apples-to-apples comparison - was much worse.  Illinois' average per-capita rank in dollars raised for the 2001-2008 period was 22nd, and its per-capita rank for deals done (again, for all VC deals regardless of stage) was also 22nd, both barely over the median - a median artifically low due to states like Wyoming and Alaska that see no venture activity, year over year.  Illinois is actually below the median when you exclude the inactive states - shameful.

That's just not leadership.

The data are clear - our technology organizations have failed to do their jobs and changes need to be made to make Illinois more hospitable to technology startups if the state has any chance at competing in a 21st century economy.  


[1] Let me be clear: I fully acknowledge that one can grow a startup without raising outside funds.  However, I believe that those are exception cases, not the rule.  More importantly, I believe the environment that allows for funding is also the best environment for a bootstrapped startup.  In other words, the plural of anecdote is not data.

[2] Data geeks: I fully acknowledge that the MoneyTree data excludes deals done solely by individual investors (angels); however, 1) this is the best objective data I could find and 2) my guess is that individual investors would only further demonstrate the conclusions I'm about to make.  Feel free to grab your own source MoneyTree data and run it yourself.  I couldn't find industry breakdowns by state, and I believe that the overall data is plenty clear about the conclusions to be drawn.

This wasn't even a black swan - how the non-TARP Chrysler creditors made an elementary mistake

The goings-on on various econoblogs regarding Rattner supposedly threatening the creditors who didn't go along with the Chrysler debt cramdown is one of the more generally well-written "you people are missing the pot" things I've read in a long while. 

Chrysler, of course, filed for a Chapter 11 bankruptcy reorganization after all the creditors could not come to an agreement.  The Obama administration, by way of the Treasury Department and Steve Rattner (the "car czar"), managed to get the four largest lenders (who had no choice, as they're being propped up by TARP funds), the UAW, and Fiat on board.  But the agreement failed to get enough support from all stakeholders, and President Obama all but blamed the non-TARP debtholders for the bankruptcy filing during a news conference the morning of Thursday, April 30. Of course, the non-TARP guys objected to this characterization.  Later in the late afternoon April 30, Perella Weinberg, one of the leaders of the non-TARP group, came around and supported the administration's plan, but it was too late.   The bankruptcy was set.  A lawyer representing the non-TARP lenders said that Perella Weinberg only came around because of threats made by the administration to bring negative PR to bear by way of their supposed influence over the White House press corps.  Perella Weinberg, of course, denied this.

On Finem Respice, Equ Privat wrote that there was a more sinister threat - that the administration did not just threaten bad PR, but would bring the IRS and SEC to bear upon the non-TARP creditors, their employees, families, alma maters, dogs, and favorite sports teams.  Of course, this could be seen as an abuse of power (as we are all terrified of the IRS and all investment managers are scared of the SEC's ability to "slow them down").  Equ Privat called the purported Rattner threats "fascist".  For this, she was mocked, mercilessly.  The author of The Epicurean Dealmaker, used Looney Tunes to jab at the non-TARP creditors, saying "[i]t's called politics, you fucking morons. Stop being such a bunch of whiny pansies."  The point TED made was that even if such a PR threat was made, there were a lot of people in the government under under a lot of stress, and the threat was just a negotiating tactic anyway.

As Felix Salmon points out, the non-TARP creditors don't have much of a leg to stand on for their whining, coerced or not into a supplicant position at the mercy of the bankruptcy judge.  Their position in Chrysler was a moral hazard one - "of course the Obama administration won't let Chrysler go bankrupt".  Thus, if it won't go bankrupt, as senior secured creditors, they stood the most to gain.  Pretty simple arithmetic.  Problem was that they didn't think the administration would really let Chrysler go bankrupt - albeit a Chapter 11 reorg rather than a Chapter 7 liquidation - and they did, because they had all the post-event actors on their side: Chrysler management, the large banks, and a private actor, Fiat, to act as a savior. 

This was a completely predicatable outcome, and the non-TARP guys treated it as impossible.  They assumed that the sanctity of American bankruptcy law and precedent would strengthen their hands well enough to force a positive outcome for their vulture investment.  The non-TARP creditors are morons, not for whining, but for not seeing this play out the way that it did.

It's called political risk.

Look, we generally think of political risk when making investments in, say, Pakistan or Russia.  But to completely discount the political risk of a vulture investment in Chrysler debt is completely insane.   The notion that a Democratic administration was going to stand idly by as vulture investors were going to cram down the UAW and TARP banks is foolish.  One, there's a legitimate policy argument that the United States needs a Detroit-based domestic automobile manufacturing enterprise for national security.  You can agree or disagree, but it's a legitimate policy position.  Two, the UAW is a powerful union in a swing state.  To put it another way, the UAW ain't UNITE HERE.  Three, it was completely forseeable that the Obama administration would take advantage of the troubles at GM and Chrysler to kickstart their environmental policy goals. 

The idea that the administration would put all of that aside based on fealty to the well-established order of creditors that you'd have in any traditional bankruptcy, as the non-TARP creditors believed, is naive at best and a violation of the prudent man standard at worst. 

Look, in most vulture situations, the non-TARP creditors would be right.  You'd have an orderly liquidation, and the secured creditors would be in the front of the line to receive proceeds.  But not here - the political risk wasn't even close to zero.  It'd be like making a bunch of greentech investments with Dick Cheney and Phil Gramm as President and Vice President.  You have to, at the very least, merely consider that maybe, just maybe, future policy decisions will make your current investment thesis weaker.

The non-TARP creditors were insane to think their hand was as strong as they played it.  It wasn't even a bluff - they really thought they were playing Omaha high low and that they were holding a Wheel.  In the end, it turned out they were playing 52 card pickup and they lost.

How To Legally Incorporate Your Startup (Quick Answer: Get a Good Lawyer) - A Conversation with Yokum Taku

One of the best things about Silicon Valley is that almost anyone will meet with you for a cup of coffee provided that you're reasonably intelligent and not an axe murderer.  My intelligence aside, I'm not an axe murderer, so Yokum Taku, partner at Wilson Sonsini Goodrich & Rosati, was kind enough to meet with me on Friday. 

As people who read Hacker News know, I consider Yokum's Startup Company Lawyer blog indispensable for budding and active entrepreneurs.  There is a wealth of information on all stages of a startup company's legal needs, free for the reading from one of the Valley's most esteemed lawyers.  Yokum most recently helped bring WSGR's termsheet generator to life and drafted the publicly available Series F documents for Adeo Rossi's Founders Institute.  Wilson also drafted YCombinator's "open sourced" Series AA angel financing documents.

In addition to Wilson Sonsini's work, Cooley Godward drafted a set of angel funding documents released by  TechStars and the NVCA also has publicly available venture funding documents, which were led by Sarah Reed of Lowenstein Sadler.

While we did discuss Dawdle, Yokum was kind enough to talk to me about a question I've been asked plenty of times: "how do I get started if I don't get into YCombinator/TechStars/LaunchBox/pick the clone of your choice?"  This actually ended up being a rather fascinating discussion, and with Yokum's permission, allow me to share with you what we talked about.  (Quotes are from handwritten notes I took during the conversation.  No recording was made.)

I presented the following hypothetical to Yokum: you have two people who want to create a startup.  They're convinced it has the potential to be a billion dollar company, and one of the founders was able to convince her parents to seed the company with $50,000.  The other founder has an uncle who's a small-town lawyer who's willing to create and file the paperwork to set up the company for free.  Given this situation, and incorporating Yokum's knowledge of financings and exits backwards and forwards, the question was "what set of documents should the founders give the uncle as a basis to form the company?"

After thinking about this hypothetical, Yokum responded "I don't think those docs are out there."  He volunteered that it would be "better to use [the YCombinator Series AA] docs than anything else [he] could think of", since there's "not much [the lawyer] could really screw up if he was reasonably intelligent."  But Yokum did volunteer that the ideal would be to have an experienced startup attorney at an established firm draft documents for the company given the firm's standard documents.

Surprised, I asked Yokum if he knew of any firm using the open sourced documents since they were available for free for anyone to use.  He said there was not a lot of actual use that he knew of since every investor already has their own set of documents with their preferred terms.  Yokum stated that the documents serve to "give entrepreneurs a chance to look at [sample docs] before getting educated [by] their attorneys," who ostensibly charge for the privilege.  The documents effectively serve to educate startup founders for free rather than actually replace seasoned counsel doing incorporation work.

Given that Yokum's advice was to forgo the free attorney and seek out experienced counsel, I asked "where does a startup founder find this counsel?"  Yokum responded with a list of regions: Silicon Valley, Seattle, Austin, Boston, New York City, and DC.  I joked, "not Chicago?" and he responded "well, I was just giving you a list of cities where [WSGR] has offices."  (A quick one, that Yokum.)  In fairness, Wilson Sonsini does not have a Boston-area office.  But upon discussion, I agreed that experienced counsel would naturally arise in areas that had active levels of investment, which Yokum characterized with two things: venture investors, and "companies that 'throw off' rich people to be angels."  Yokum volunteered that he only knew of one or two venture firms in Chicago, and I named a couple more in town.  He then asked about angels, and I had to concede that I knew of no active Internet angels in Chicago or the Midwest as a whole.

However, the list of areas that did have counsel that Yokum termed appropriate was higher than I thought it would be, meaning that there are a number of lawyers at a number of firms in a number of cities from which to choose.  I asked "well, how does someone find such a lawyer?" and Yokum responded, simply, with "an introduction."  He elaborated, saying that even the largest firms still do startup incorporation work, but that not every "bakery and restaurant" needs this sort of counsel.  Introductions provide a filter for hard-working partners to signal that a company is worth representing.  That said, Yokum was very clear: "any [WSGR] partner [in any city] is happy to sit down for coffee and chat" with founders.  Yokum made the further point that any startup "[has] to network to be a successful startup company" anyway, so it wasn't unreasonable for partners to want some sort of filter to meet for coffee.

I then asked "well, where should founders network to get this introduction?"  Yokum, being extremely patient with me, explained his thinking: in any startup hub, there are numerous events where startup founders meet.  Anyone can stick out their hand, introduce themselves to another founder, and strike up a conversation.  Startup founders are generally thrilled to recommend their lawyers to another founder (if they like their counsel), so it's actually rather easy to get a name or three.  Then the new company can do a little internet research and ask the founders they met at the event to do a quick e-mail introduction to the lawyer(s) they might want to represent them in their new venture.  Founders should meet for coffee with those partners before they make a final decision on representation - again, any Wilson Sonsini partner will meet with you as long as you have a warm introduction.  (And although it may sound like just talk, since WSGR is "Google's lawyer and Sun's lawyer", they still do startup incorporation work.  As Yokum quipped, "once upon a time, we did Sun's incorporation paperwork".)

I found this conversation extremely valuable: Yokum acknowledged that any set of open sourced documents wasn't appropriate to use as a "fill in the blank" template, but he also didn't state that only a small handful of Valley firms were qualified.  There are a number of firms in a large number of cities that have partners that not only "do" this work, but have the requisite experience to not make the simple errors that prove costly down the road. 

I personally would recommend Wilson Sonsini to any potential founder given my high esteem for Yokum and his colleagues, but there are partners at other firms in all these regions that may be a better fit for your given situation.  Please note that you are looking for a good fit with a particular partner, not just the "name" of the firm.  Even though each firm's documents are "off the shelf" and an associate - or paralegal - will do most of the work, having a good relationship with the partner is critical for quick and accurate answers to your well-researched questions.

[EDIT: Apparently YC may make their incorporation documents public, which may specifically address this issue by giving the "lawyer with a shingle" a fill-in-the-blanks template that would work: ]