Open post

The Simplicity Paradox: Simple > Complex > Simplicity

The Simplicity Paradox: Simple > Complex > Simplicity

“When you start looking at a problem and it seems really simple, you don’t really understand the complexity of the problem. Then you get into the problem, and you see that it’s really complicated, and you come up with all these convoluted solutions. That’s sort of the middle, and that’s where most people stop. But the really great person will keep on going and find the key, the underlying principle of the problem — and come up with an elegant, really beautiful solution that works.” – Steve Jobs.

That’s one of my favorite Steve Jobs quotes. It does a great job in capturing the counter-intuitive nature of how simplicity is harder than complexity.
I’ll go one step further and create the following framework:
Simple > Complex > Simplicity

This framework applies to many facets of life, certainly beyond product development.

Simplicity is elegant. It’s easy to work towards. It’s easy to understand and to sell. But there’s a big difference between simplicity and simple in this case. Once you get over the complexity hump, the solution seems obvious, and seems like it could have been achieved without the work of going through the complexity hump. In practice that’s not the way it works. Truly elegant solutions are the result of fighting through complexity, and are rarely single insights.

Here are a few examples:

The iPhone is perhaps the greatest example of this framework. After the original simple telephone, which let you simply dial and speak, companies went on a feature arms race to continually add new features. Address book, speed dial, calendar, etc. And phones got to be very very complex. And technology companies relished in that complexity and tripped over each other to add features and more complexity. Then the iPhone came along, the result of pushing through the complexity limitation, and ushered in a new era of phones.

In politics we’ve recently seen a similar situation. Romney’s Tax Plan is in the “simple” category. Easy to sell and explain and put into sound bites. But it doesn’t have details because it hasn’t started actually trying to solve the problem. On the next step is Obama, who is so mired in details that he can’t convey a vision for his plan. His terrible performance in the first presidential debate last week showed this. He was trapped in the complexity of the details, trying to explain and resolve real deep challenges to make this all work. Now compare this to the polished arguments of Clinton’s DNC speech, where he accurately portrayed the key issues on both sides, in a simple way that people could understand and relate to.

The same is true in architecture, one of my passions. All architecture started as simple housing and shelter. Then continued to get more complex and decorative. This decorative trend culminated in the Beaux Arts School (literally and figuratively) of architecture. Architects tripped over themselves to add more details and ornamentation the way Samsung and Acer add faceless “specs” to tech products. Then came the modern movement, led by the Bauhaus school, which focused on simplicity. Elegant solutions to challenging problems. And this style and philosophy remains the most elegant in the world. Have you wondered why an Eames chair designed in the 50’s looks more contemporary and futuristic than an Aeron chair designed in the 90’s?

And finally, we see this same trend in the growth of knowledge. Things start as simple: thy sky is blue. Then get complex: because light refracts at that particular wavelength due to the molecular composition of our atmosphere. Then as you really understand things, a new simplicity emerges. The understanding that chemistry, physics, and biology are not separate disciplines, but different ways of looking at the same physical world. As Einstein put it so elegantly: “If you can’t explain it simply, you don’t understand it well enough”.

So whether you’re explaining something, designing something, or building something, keep this in mind. Don’t be naive and think things are simple. Don’t be lazy and let yourself stay in the complex. Work towards simplicity.

The Four Internet Ages

The Four Internet Ages

Now that I’m between startups, I’ve spent a lot of time thinking about what I want to work on next.  I like thinking in frameworks, which makes it easier to evaluate new opportunities and create my own theses around where technology and business is heading.  My previous analysis on the Product vs Distribution Framework focused on the core traits a company needs to ultimately succeed. This post looks instead at the macro trends of the Internet itself. 
Each phase of the Internet’s growth was enabled by key platforms, and had key successes that were able to level those platforms and other trends to create huge businesses.  Each phase also built on the previous phases, and would not have been possible without the foundation created before it. And because of this improving foundation and growing user base, each phase enables bigger opportunities that can be reached faster than before. 

The diagram below shows the four key phases of the Internet, as the platforms and successes in each, as well as the worldwide desktop and mobile Internet user base. Note that the timeline for each is a general approximation of that trait as the defining leading edge of the web, and there exists considerable overlap in the general use cases and impact of each.

1 – The Early Internet

The first Internet phase saw the early growth of users.  This growth (and perceived future growth) was so explosive that it led to the great Dot-com boom and bust. It was built on the proliferation of the web browser (Netscape and later Internet Explorer), increasing penetration of desktop computers with Internet access, and standardization of web protocols.  In the early days, there were not a lot of users or content, so the most logical way or organizing available content for a limited user base was simply an editorial listing of interesting content, thus the birth and proliferation of portals like Yahoo.
2 – The Searchable Web
The second phase of the Internet saw an explosion of content, which was built on the key enablers of the first age (desktops, browsers, web standards). Once the volume of content exploded beyond the early manageable levels, lists and portals were no longer sufficient to access this content. This led to search becoming the only feasible way to harness the incredibly expanding volume of content. 
Google emerged in this era as the key platform because of it’s simple, powerful, and fast search capabilities. This dominance led to a web that was organized through the principles of PageRank, and created the SEO and SEM industries. It also opened up a new world of long tail e-commerce, where small companies could have a much bigger footprint than ever before, and cheaply reach customers across the world.

3 – The Social Web
The third phase of the web was based on saturation of online penetration among the general population. A true social web can only existing once online penetration begins to saturate the population, reaching 70% adoption or more. Interestingly, most developed countries started reaching this penetration level around 2003-2005, coinciding with the birth and growth of social networks. This is in stark contrast to the Searchable Web, which is valuable to users even with low adoption rates as long as there is enough content to search. 
In this phase, Facebook has emerged as the defacto social platform. This platform has facilitated the creation and growth of two of the fastest companies ever to reach a $1B revenue run rate: Zynga and Groupon.  Both of them used the Facebook platform to grow faster and more cheaply than ever before due to the incredible low friction that social networks have to get new users. Zynga reached 100 million users faster than Facebook itself, mainly because most of the users were ALREADY on FAcebook, making them much easier to acquire.
4 – The Mobile Web
The fourth phase if the one we are just entering, which is being enabled by ubiquitous mobile computers (e.g. smartphones). Inherent to this phase is not just the mobile computer itself, but the fact that they are increasingly always with us, and are location aware (know where we are).  The combination of ever-present mobile computing, plus a rich location layer, seamlessly integrated with the social graph and the searchable web creates enormous opportunities. 
The key platforms in this space have are clearly iPhone and Android, but there is still room for winners on the customer side.  Foursquare appears to be one of the early companies to succeed while specifically leveraging mobile + location, but there will certainly be more. 

What does this mean?
What’s interesting about these trends as I pointed out earlier is that each phase gets progressively bigger than the one before, and leads to faster growth for companies that are well positioned.  You can bet that new companies will emerge will get to $1B in revenue faster than Groupon and Zynga. 

The Product vs Distribution Framework

The Product vs Distribution Framework

In the startup world there is an ongoing debate about the importance of having a very solid product versus having great distribution. A few weeks ago at a dinner hosted by David Skok and Antonio Rodriguez we had a very active and interesting debate about the merits of each. Antonio posed a question about which of the two we would prefer. Since then I’ve continued to think about this issue, and created a framework to help me think about it and evaluate the tradeoffs.  

The Product-Distribution Framework

Given that the product vs distribution debate is not just an either or, I wanted to create a framework to help me think and evaluate the tradeoffs of each, and take into account the strength of each. So I created a simple 2×2 matrix comparing weak and strong product vs weak and strong distribution, using a fun theme of plants to characterize the various types of companies (trust me, it’s much safer for work than the initial images).

The Oak Trees – Great distribution and product feed on each other to build great companies
At the top right are the truly great companies, that have created solid products and achieve successful distribution to reach enough people.  These great companies are household names, including Facebook, Apple, and Google.  There are various paths to get there, but most of these companies have a fanatical devotion to creating a solid product, coupled with a very smart, disciplined, and powerful distribution. Throughout the growth, these two strengths continually feed on each other creating a positive feedback loop. 
Example: Apple in 2010 – Apple is firing on all cylinders, led by Steve Jobs’ literally maniacal focus on creating a perfect product. But it’s easy to forget how strong the Apple marketing engine is, creating hundreds of millions of dollars of free publicity with every product launch. They have also created one of the world’s strongest brands, and use their huge profit margins to build a wildly successful retail distribution channel.
Example: Google – Google is the undisputed search leader with about 70% of the market (and holding strong). Early on they built a fundamentally superior search product, perfected it through continual execution and improvement, and leveraged the strength of the product to grow distribution. Now the two build on each other, with their market share in search enabling them to continue to improve their product through more data and more servers, and that continually improving product keeps people coming back to use their search engine.

The Weeds – Potentially profitable by eventually doomed
At the top left are companies that have created great distribution despite a weak product. Contrary to conventional wisdom, these companies can be very profitable despite the weak product because of various channels to reach customers and user habits.
Example: MySpace – MySpace leveraged huge email lists to spam users and build a large social network. They grew quickly and sold for $580 million, but continual lack of execution on the product led to their downfall and imminent irrelevance.
Example: Microsoft – A monopoly is about as strong a distribution strategy you can have, but even that is not enough to sustain a company that continues to make inferior products. Despite being wildly profitable, their stock has gone nowhere in 10 years, and their lack of product execution has put them in weak positions in the next computing wave (mobile and tablet).
Example: American auto makers – The analogy even applies outside consumer technology, as evident by the downfall of the American automotive industry, which for too long relied on patriotism, customer loyalty, and a saturated dealer network to distribute an inferior product. Eventually it caught up with them, and now even though they are building better products, decades of perception still linger with consumers and they face an uphill battle.  

The Desert Flowers – Potentially great but need some breaks
At the bottom right are companies that have created a truly interesting and innovative product, but have struggled to get it out to the people that care about it. Many companies founded by techies have a strong risk to go down this path, continuing to focus on building the perfect product and not spending enough time getting it out there. And make no mistake, Facebook was not one of these companies in the early days, aggressively crawling online groups to get initial profile pictures and emailing whomever they could.
Example: original Mac-only iPod – Given the rapid success of the iPod franchise, it’s easy to forget that it originally launched exclusively for the Mac as a way to drive Mac sales, and took almost 2 years to sell 1 millions units (vs. 74 days for the original iPad, 28 days for the iPad, and about 1 day for the iPhone 4). It was only when they began offering a PC version and leveraging their iTunes distribution platform that sales began really accelerating.

The Tumbleweeds – Also Rans – Lots of challenges, but could work in certain offline niches
In the lower left quadrant are the companies that don’t have a particular strength in either product or distribution, and are thus left facing many challenges. In certain commodity local businesses with limited competition, this may not be an issue (for example the only dry-cleaner in a neighborhood), but for online businesses this can be a killer.
Examples: Most clones that lack distribution – Clones often try and replicate a successful product but lack many of the soft traits and methodologies that made those products successful in the first place, and thus often create weak imitations. If those are coupled with distribution advantages, they can be profitable, but often they are mired in the bottom left quadrant to hang on and pray.

What this means on a practical level
Coming back to Antonio’s question from that dinner, if I had a choice of having either great distribution or great product, I would choose great distribution, turn that into profitability, and leverage that to build a great product. Practically speaking, however, those choices are rarely presented.  Therefore, when starting a new company, the important point is to clearly understand the tradeoffs as well as the strengths of the early team and company.  Then use that to built a great company, whether it means leveraging distribution to get to a great product, or building a great product and figuring out the distribution. 
And the truly great companies are constantly focused on both, like an oak seed going deep into the ground for a solid foundation (the product), and reaching for the skies to get sun and spread the branches (the distribution).
I’d love to hear other thoughts about this framework as well as other good examples. I’m also planning a follow up post about how to move between areas in the quadrant, and specific industry dynamics. 

YouCastr – A Post-Mortem

Introduction: Why I’m writing this

There are many post-mortems from failed startups out there, mainly because there are a lot of failed startups, and the people that start them tend to be very introspective and public about their successes and failures. I’m no different.  This post-mortem will serve to get things off my chest, organize my thoughts, get the most out of the experience, and share my experience with others.

I’m also writing this to be able to point to a single, detailed, lengthy answer to the inevitable questions I’ll be getting from friends and colleagues about what happened with YouCastr. Now people can read to their heart’s content. 

As I write about my experiences, people who know startups will see many common themes.  Most of the conclusions here are consistent with Paul Graham’s analysis and understanding of startups, what’s challenging, and why they fail (if you’re planning on starting a company, especially in tech, read every single one of his essays). But no matter how many times you read or hear certain lessons, you have to do it yourself to really understand. Here are my specific and personal experiences to add color to that advice.

A Brief History

YouCastr is Born

The idea for YouCastr was hatched about three and a half years ago, during a car ride down from Jay Peak with my good friend Jeff Dwyer.  Throughout the entire ride we were throwing ideas back and forth about cool companies, fun ideas, and stuff that bothered us.  One in particular stuck with me, which was the idea of bringing Mystery Science Theater 3000 to the modern age, and let anyone provide their own comments.  After exploring this some more, I ended up settling on sports as the logical market. This would mean the commentary would have to be live, because otherwise it would have been just a podcast platform, which was not very interesting technologically.  A couple of days later I shared the idea with my soon to be co-founder Jeff Hebert, and we continued thinking about it.  Within a few months we put together a team of 4, created our initial plans, and began building an alpha version. 

Initial progress

As we were getting off the ground, three of the four founders (including myself) were mainly working on building the actual product, since we couldn’t do much of anything without that. We learned Ruby on Rails, did our own designing, worked nights and weekend on the startup as a complement to our management consulting day-jobs, and clawed our way to our initial alpha and private beta. 

Building the company

We then spent the next three years building the company, quitting our jobs, raising money,  opening our office, hiring people, firing people, going back to bootstrap mode, and finally, pulling the plug.  

Pivoting

Startups are all about pivoting. It’s like a word game where you start with HURT, change one letter at a time, then end with LOVE, with many other four-letter words along the way (btw, you really can do that in 6 moves). In our case, we started as a virtual sports bar where people could chime in audio commentary and ended up as a do-it-yourself pay-per-view video platform.  Seems like a radical change, but each pivot was the equivalent of changing one letter.

Our first pivot was to focus on live audio sports broadcasting of games that weren’t being televised (instead of adding commentary to nationally televised games), in response to what customers were doing. Then we added video broadcasting to that same market, focusing on broadcasting sports that weren’t covered (primarily high school and college).  And finally we expanded beyond sports, mainly by de-emphasizing the sports branding on our platform, and adding a few features more geared towards video producers than to schools and teams. All of these were natural pivots, not jumps.

Shutting Down – The Reasons and Need

1 – Ran out of cash

The single biggest reason we are closing down (a common one) is running out of cash. Despite putting the company in an EXTREMELY lean position, generating revenue, and holding out as long as we could, we didn’t have the cash to keep going. The next few reasons shed more light as to why we chose to shut down instead of finding more cash.

2 – The market was not there

The thesis of our current business model (startups are all about testing theses) was that there was a need for video producers and content owners to make money from their videos, and that they could do that by charging their audience. We found both sides of that equation didn’t really work. I validated this in my conversations with companies with more market reach than us, that had tried similar products (ppv video platform), but pulled the plug because they didn’t see the demand for it. 

Video producers are afraid of charging for content, because they don’t think people will pay. And they’re largely right. Consumers still don’t like paying for stuff, period. We did find some specific industry verticals where the model works (some high schools, some boxing and mixed martial arts events, some exclusive conferences), but not enough to warrant a large market and an independent company.

3 – The team was ready to move on

The core team of 5 of us has already made significant personal, financial, and emotional investment over the past three years. We have had our share of tough breaks, not to mention almost two years without salary.  

4 – No light at the end of the tunnel

Given the industry trends we were seeing, we just didn’t see a light at the end of the tunnel worth surviving for.  This market may get slightly better, but it’s not going to be a big company.  

5 – We need to put a stake in the ground

We’re survivors. We hate giving up, and have tenaciously survived for a long time.  We survived the worst economic environment in generations (during which we also tried to raise money), and knew how not to die.  But it was time to shut down and move on. 

A Retrospective – Why we failed

In most of the following points I say we, but as CEO, many of the mistakes below can be placed directly on my shoulders, which I will take as a valuable experience as I move on.

Didn’t pivot fast enough

We did a pretty good job of pivoting, but we didn’t make the hard choices quickly enough. Our first pivot, towards broadcasting of original sports content, took us 6 months.  We had legacy users and web traffic that we were afraid of losing. We had a brand in one area. We had a story for investors, customers, and the community. All of that made it harder for us to quickly change our focus and apply all resources to the new area, which we knew was the long term future. We iterated our product quickly, but didn’t pivot fast enough

Didn’t love it
We started the company because we liked the idea and wanted to do something entrepreneurial. We weren’t in love with the idea or market we were going after, and weren’t core users of our product. We worked really hard getting it off the ground despite this, but it made it more difficult to sustain the energy and to understand the best product choices.

Made hiring mistakes

Hire slow, fire fast. That’s how the saying goes. We made a mistake when hiring. We hired too early (before product-market fit), and we rushed into it by foolishly setting an internal deadline to make the hiring decision, which drove us to ignore some gut reactions by some of the core team members. Eric Ries has a great overview of how not to make these mistakes in his Lean Hiring Tips post.

Spent too much time raising money

Almost immediately after our private beta we started trying to raise money, and saw many of the typical challenges when raising money for the first time.  This was late 2007 remember, when plenty of Web 2.0 companies were getting funded, and we got caught up in the spirit.  We spent 6 months raising our angel round, and finally closed it after launching our public beta in February 2008.  We raised less than we should have (Chris Dixon notes that the worst thing a seed-stage company can do is raise too little money and only reach part way to a milestone, which is great advice). We spent 3 months getting ready for raising our next round, then 6 months beating our head against a wall trying to raise money during the financial collapse. Finally we regrouped, but had serious short term cash issues that we had to resolve with additional investment from current investors and cutting of costs. We then focused on the product and made some solid progress, with the plan on giving it another go. By the time we got there, though, we realized we didn’t have the metrics we wanted, so pulled back our fundraising efforts and decided to go back to bootstrap mode. 

Never quite figured out customer acquisition

We never seriously figured out customer acquisition, and had trouble growing throughout our various pivots. Customer acquisition is hard and more expensive than most people realize when starting a company. There has been a lot of good analysis, including Andrew Chen’s overview of calculating cost per customer acquisition and David Skok’s warning about how it can be a startup killer, but it takes firsthand experience to understand that. 

Too many founders

We started the company with four co-founders. I believe Dharmesh Shah nailed it in his analysis of the optimal number of co-founders. In our case, the advantage of having 4 co-founders was that had more people and resources willing to work for sweat equity. But it also presented challenges, including slowing down decisions, as everyone naturally wants to chime in with their opinion.  As CEO, a lot of this falls on me, as I should have been stronger in making decisions. I tend to be a consensus builder, but in my next startup I will have to improve my ability to make bold unpopular decisions while still motivating everyone and building consensus after the decision is made.  On the plus side, the entire founding team was incredibly committed through the entire time, and we never had any of the fundamental founder issues that many people complain about become a major problem. 

Counted on big partnerships to close (when they didn’t) and to help more than they did

We initiated some very interesting high level partnership and business development discussions with large companies. We put some hope into these deals, but large companies move much slower than startups, and every deal took much longer than we needed. Marc Andreesen even draws a parallel that going after large partnerships is like chasing Moby Dick

We put too much stock and hope into closing some of these. And for the ones we did close, we expected more results than we got. It’s an interesting dilemma, because partners are often the best acquirers, but the downfalls can even lead people to suggest to NOT partner with big and powerful companies

What I Learned and some of the positives

I’m an optimist by nature, which you have to be to start a company given the odds and challenges. Given that, it’s good to look back and think about the positives of the entire experience. It’s incredibly satisfying to create a revenue-generating company that is flirting with cash-flow positive, especially having started with a crazy idea. The experience of building a company and tasting success was incredible, and though we didn’t quite get over the hump, we have tasted it, and it will only motivate me more to get there again, through all the hardships.

On a personal level, I’ve learned more about business, people, life, and myself over the past three years than I ever have. I have made life-long friends with whom I can share these experiences with. I have come closer to understanding how to reach my limits, which is one of my fundamentals goals in life.  Startups let you try and find those limits  in a way that most jobs wouldn’t, and I’ll keep pushing. And best of all, I have realized that my true passion is entrepreneurship. I have a very addictive personality, and therefore only do things where I am comfortable with the extreme case, like starting companies or running ultra-marathons

We built a great culture, including some great parties for the local tech-scene, competitive ping-pong matches, and late night coding sessions with house music blasting. We converted three lifelong PC people to Mac, all of whom are happier for it.  And most satisfying, we helped kick start two young and promising careers, those of our designer and community manager.

Moving Forward

My long term goal is to continue starting companies.  There’s no question I’m in a better position now to start another company than I was when I first started YouCastr.  It’s almost like thinking back at how much more fun high school would have been had you known what you knew when you graduated back when you started. Practically speaking, though, after having gone almost 2 years without a salary, I’m not in a financial position to bootstrap another company, which is the way I would really want to start a company.  

In the short term I’m working on various projects for my consulting company, Scenario4. We are working with some great people on some fun projects across the tech world.  I am also pursuing several side projects, including the recently launched BooksforBits, a non-profit company designed to help accelerate the transition to ebooks while getting books to communities in need.

It’s an incredibly exciting time in technology. We are on the cusp of the post-pc era, which includes a revolution in the mobile space, and in how we interact with and perceive computers. We are in a world where startups can be more capital efficient than ever before. And most interestingly for me, as a society we are going to be facing challenges unlike any we have ever seen. 

With every ending is a new beginning, and I’m excited about what lies ahead.