It's Insanely Hard to Make a Kick-Ass Mobile App

It would be hard to find a self-respecting geek and technologist who disagrees with the phrase "the future is mobile." All the coolest new gadgets are tablets and smartphones and all the biggest plays in content are on those devices. The market dominance of the biggest tech companies depends on how you use that device in your pocket or backpack.

So we get it. Mobile is hot. Mobile is irreplaceable. And most importantly, mobile is the future. But this mobile future takes software for granted.

At both Skritter and CodeCombat, we learned that because we built mobile versions of our apps for iOS. We found that there are so many apps, they have been devalued to the point of monetary irrelevance. That's sad because a good app is a piece of art. The user flows, the interface, the streamlined backend, all the pieces of a finely-tuned app take so much time and energy to perfect that I wanted to write a post to call attention to the level of software perfection that most people have acclimated to without even knowing it. Here are three reasons why it's insanely hard to make a kick-ass Mobile app:

System Constraints

You could say that Objective C is the problem, or put another way, the lack of >insert language< or >insert framework<. But the reason why developers are shackled to Objective C is to make sure processes finish in human time, and this is fundamentally a performance bottleneck in mobile hardware. Swift is a great step forward, but it's immature and for complex backend work, Objective C is still necessary. 

Skritter's iPhone app is great, but at core it is just an app to learn Chinese characters. Skritter doesn't pilot your car or allow you to edit movies or do any other mind blowingly complex task. But when it first launched, it wouldn't run on all but the n-1 generation of devices. When we tried to launch CodeCombat's tablet app, on the iPad 2 Air could run it at decent speeds. Why? Because older devices have limited memory, CPU cycles, and bandwidth, newer versions of iOS brick older devices, and news features are only available to developers on newer versions of iOS.

For a generation of coders accustomed to limitless managed memory, high-level programming abstractions, and thousands of deeply functional opensource libraries, mobile is a step back to the Byzantine world of Commodore 64s.

If you have built an app that doesn't choke on graphics and feels responsive, you have accomplished an incredible challenge and I tip my hat to you.

No Features

User-software interaction paradigms are changing and simplifying. Photoshop's submenus, window management, and setting screens afford it a level of feature depth I don't believe can be achieved in a mobile environment no matter how sophisticated and intuitive UI conventions become.

Put another way, there is a terminal threshold of feature depth to an app and this limits the scope and usefulness of any one app. That's not an intractable problem and the solution that has been embraced to date is app fragmentation. There are separate Skritter Chinese and Japanese apps for this exact reason. But consider the logical extreme of this design trend: if you were to make a fully-featured Photoshop app work on a mobile device, you would end up with several hundred micro-apps to do one thing very simply. Just download Instagram to see what I mean.

So, if you are attempting to make a kick-ass mobile app, you must make a lot of decisions that ultimately come down to "how little will my customers allow me to integrate into this thing?" Err on the side of simplicity and you might end up with a fart app. Go the other direction and you have a pile of unusable menus and sliders.

Extreme Polish

The app market is cruel; even moreso for the need to do a Hollywood launch. Because nobody has solved the discoverability problem in the app store, you need to launch big and magically figure out exactly what users want ahead of time. That last point is especially troublesome for people used to iteratively designing for the web. The road to app store success is littered with the corpses of apps that launched big but hadn't fine-tuned their interfaces, didn't have time to include that key feature users wanted, or just failed to make the app novel in some primal way.

Even the basic business sense of not overbuilding is turned on its head for mobile development. If you somehow manage to land that sweet TechCrunch article or New York Times acknowledgment, the app has to be perfect not just from an implementation point of view, but from a feature and context perspective as well. A perfectly implemented, compelling, novel app isn't enough; you have to know that your customers will agree about your choices.

There's always another app for whatever you are doing, and that makes customers more picky. The pressure to get it right the first time can be crushing.

So Get Insane and Start Building

I write all this not to discourage, but to uplift all those developers struggling with their mobile projects. It's tough to make something good, let alone make something good that also happens to make enough money to warrant the effort. For those out there building apps of consequence, I want this to read as a note of enouragement: stick to it and overcome the difficulties I've written about above. The mobile future is depending on you.

It's Religion All the Way Down

It was a hot day and as Rebecca and I drove north on i70. I was looking straight ahead as I pulled up behind a beat up SUV with a bumper sticker extolling the virtues of Yoga. A hand poked out of the front window and flicked a cigarette.

In the passenger seat Rebecca craned to confirm the cigarette siting, "Is that person smoking?" she asked. "Doesn't seem to fit with their bumper sticker."

"True," I said as the light turned green and I accelerated to pass the SUV, my underpowered Suzuki Aereo struggling to hit 40mpg, "it's religion all the way down."

"What?" Rebecca asked.

"Oh, it's just something I've been thinking about recently. Everyone always says that there's no accounting for taste, but until recently it was hard for me to admit there were chain smoking Yoga enthusiasts. It just seemed too contradictory. The best model I've found for respecting those kinds of differences are to treat them like religions."

"Yoga isn't really a religion," Rebecca said. "And what's with all the way down? Is that a rip on the turtles?"

"Yeah, I just tack on the turtles bit because it's funny."

With the SUV slightly behind us, I quickly changed the topic, "Quick, can you see what the driver looked like? I'm curious what a chain smoking Yoga enthusiast looks like."

Rebecca swiveled around in her seat to look, "I can't see, there's too much reflection in the windshield."

"Darn, I guess it will have to remain a mystery."

Marketing is Just Sales Without Accountability

Over the course of the last 7 years, I've spent a considerable amount of time managing sales and marketing for my own 2 software companies as well as a web consultancy. I've attended conferences, made sales calls, printed flyers, and schmoozed at investor gatherings. And I've come to believe that marketing is just sales without accountability. Bear with me for a moment.

Marketing is Sales

Both marketing and sales are intended to produce the same end result: make more sales. Marketing is the top of the sales conversion funnel; the first interaction with a potential customer in which good salespeople provide value and make a lasting impression. But let's not loose sight of the end goal: that first contact with a potential customer is fundamentally about sales.

Companies don't sponsor trade conferences and staff booths because they want to spend tens of thousands of dollars on snazzy signs and airfare, they want to make more sales. Companies don't maintain blogs because they want employees to become excellent writers, they want to be noticed for their contributions to their niche and make more sales. Companies don't buy television, print, and radio advertisements because they want to support actors and graphic artists, they (you guessed it) want to make more sales.

Marketing Isn't Measured

The problem with marketing is that it's so far towards the top of the sales funnel that it's difficult to measure. Sales calls either result in a sale or not, so it's natural and easy to measure their effectiveness. It's a lot harder to measure the value of meeting someone new at a trade show. Will the person become a paying customer? Will they refer someone to your business? Will they write about what great work you are doing? Or will they trash you to a potential customer? It's just not something that's obvious or easy to measure.

As a result, businesses and business owners (myself included) simply throw up their hands at some point and start generalizing. "We have to get our name out there somehow, and people watch television, so why don't we try a TV ad?" Or perhaps "We know some of our customers attend the BigConf Trade Show, so let's exhibit this year."

The unspoken rule of marketing is that most of it is wasted effort, but for a myriad of reasons, it's too costly to figure out which part is wasteful. My suspicion is that for companies in established markets, what most folks would recognize as marketing probably is worth the time and energy. Most companies can't prove that, or course, but the fact that they are still around means that they are either spending little enough on marketing  that their failures don't sink the ship, or the marketing efforts are actually paying off.

Stop Marketing and Start Selling

If you work for a tiny company or are building a new product, you can't afford to waste resources on unprofitable marketing or sales. You have to start measuring marketing efforts in the same way that you might measure sales. At this point you might be saying "Yeah, well, that's all fine and dandy, but how the heck do you propose I do that?"

Simple: pick some metrics and start recording data.

Here are some examples of things that I've personally tried:

1. Enter people you meet at trade shows in a CRM and track sales as they come in later.
2. Attach a coupon code to print materials and track how many times the coupon is used.
3. Ask incoming sales leads how they heard about you, be bold and ask for a name if the person is willing to give you that information.

You might now be saying "but I'll never be able to measure activity X!" One striking thing I've noticed about sales is that salesmen and sales departments are loath to invest time or money in anything that can't be tracked. That's a good starting point for marketing too. That doesn't mean you can't do activity X. It just means you have to get creative about tracking it. In cases where you can't find any way conceivable way to track it, that's generally a good indicator that it's not a profitable marketing channel.

Conclusion

You might well choose the wrong metrics to measure the first time around. For instance, the first two years that I attended trade shows for my company, I measured booth traffic. I literally just wrote down how many people talked to me at the booth. It turns out that people visiting the booth didn't predict sales at all. But that data saved us more than more than $10,000 when it came time to exhibit in year three. We had hard numbers that suggested that either 1) it didn't make money for us to attend trade shows or 2) unqualified visits to our booth are not related to making sales to qualified leads. Dodged a bullet there!

Marketing doesn't have to be blind, and when it does, don't do it. Treat marketing like sales, measure everything you can, and in a year or two you should be able to make at least one decision that saves your company $10,000.

Sivers is Right: Customer Service is the New Marketing

This post was originally published on 8/21/2011.

Recently I read an old interview with Derek Sivers about his CDBaby customer support. I've always found his writing educational, but in this interview his closing sentence struck me as particularly true: "customer service is the new marketing." That got me thinking about my own experience with my company and how much we labored to do marketing "right."

We started Skritter on a shoestring to teach Chinese and Japanese. We decided that we would either make money and support ourselves, or stop working on the site. After the first year, the site was generating cash, but not very much. Nick, Scott and I decided we needed to hit marketing hard. We partnered with a great guy in China to help us reach more customers and we set about our marketing.

We began with Adwords. We attended several relevant trade shows. We tried promotions and giveaways on the site. We started Tweeting and developed a strategy for Facebook content. We had two marketing interns in China talking to learners on the streets of Shanghai. We were present at Chinese testing centers, we contacted schools to sell site licenses, we did webinars, distributed print ads, setup an affiliate program, spent serious time on SEO, created a content strategy to improve SERP results, collected quotes from marketing consultants and agencies, redesigned the appearance of the site, and conducted A/B tests.

Throughout all of this, we continued to grow, but our growth curve didn't look a hockey stick, it was the gradual upward movement of a well liked company.

We always made it a priority to bake in tracking to our marketing, and after a year with little to show for our effort, we decided to give it a temporary rest.

That week we watched in amazement as the site kept right on growing, despite negligible marketing outlays. We nervously twiddled our thumbs, waiting for the growth numbers to level out and go into free fall, but that never happened. Instead, Skritter continues growing to this day, and we have strong evidence to suggest that it's because of the way we treat customers.

First, we conducted a poll of our user base and found that more than 35% of our customers had personally been in contact with either Nick, Scott, or I. That meant 1 of every 3 paying customers had traded a personal email or met us face-to-face! Considering the size of the site at the time, that was an astonishing number of customer interactions.

Second, even after months of no marketing, growth continued unabated. Moreover, we continued seeing a trend we hadn't paid much attention to earlier: people arriving at the site and telling us (albeit not in an easily quantifiable way) that they were there because of referrals. This was corroborated by a forgotten poll we had done more than a year before which attributed most new inbound customers overwhelmingly to referrals.

So, while we were trying our best to find scalable marketing solutions, all those hours spent in front of keyboards being sociable and helpful to current customers drove site growth. Our customer service has been our marketing, which directly supports Derek Sivers' opinion.

Marketing doesn't always make sense for companies. Where marketing is concerned there are two schools. The first is the "build it and they will come" school of thought. The second is the "most companies needs marketing" school. We clearly followed the latter and although we are just one data point, when we started my second company,  we took Derek's advice to heart and disregarded marketing at least. When you've only got 5 customers, there's a real temptation to focus all your time and effort on finding more customers. But that's missing the forest for the trees.

In our experience, the key to a larger customer base is the person right in front of you, money in hand, waiting to see how you'll treat them.

Games: The Anti-MVP

This was originally posted on 6/6/2013.

My cofounders and I were sitting in the Mission on a sunny morning, eating crepes and talking with a survey and testing guru about testing business assumptions for our new startup.

"I would be hesitant to sink more than a week or two into the idea without data from customers."

My cofounders and I traded sideways glances. The truth was that we had been working on our new startup for 4 months already. We had been play testing our game for more than a month, but  it would be a while before it was ready for a beta launch .

Our meeting left us feeling uncertain, our previous startup had a minimum viable product ready to ship inside a month; why was this taking so long? It didn't hit us until later on when we were speaking with more experienced game devs.

"You have something ready to play in 4 months? That's great."

We had been operating on the assumption that like a B2B startup website, we could throw something down in a week or two, go to customers, test it, and begin iterating. But games defy MVP release schedules. Nowhere is this better illustrated than in 2D Boy's illuminating blog series about developing World of Goo. It's true that they churned out their first version in a week, but their first game tests (which occurred several months into development) were a mess. At CodeCombat we understand that all too well, unfortunately.

Games must be fun. The primary business assumption of any game is "users will like playing it." It's almost guaranteed that users won't be satisfied playing a two week old MVP. In fact, that's sort of the definition of a minimum viable product: it's rough, unpolished, and easy to change. If you're doing it right, the MVP is difficult to take seriously. By contrast, the quality standard for games has been raised so high that it's often not enough to slap something together to gauge user interest.

The takeaway we've learned is that startups with experience building websites need to step back and redefine their MVP expectations when building a game. It takes longer to make something fun rather than just functional, and unfortunately games are defined by the first term, not the second.

Your Child's Name Won't Limit Their Digital Future

What are you planning to name your children? If you answered with any common first and last name combination, your child may be at a digital disadvantage. They will be condemned to a life of appending numeric sequences to their user names, picking off-brand Twitter handles, and choosing unrelated domains for their websites. Their Facebook and LinkedIn profiles will be difficult for new acquaintances to locate, Google results will misrepresent them to future employers, and their children will have an even harder time of it.

At least, that's what I thought several weeks ago.

It's easy to imagine that any given name you choose for a child will be common enough that somebody out there has grabbed the domain, the Twitter handle, the Facebook username, the Gmail address, and countless other digital identifiers. In this paranoid world-view, your child is relegated to being a second class citizen of the net (or whatever it becomes) simply by dint of having a common first/surname combination.

But how large of a problem is it really? To answer the question, I did some research on full name variation and came away surprised.

It is trivial to find popular baby names for a given period of time, but finding the frequency of full names in the US is another matter. The CDC and census bureau both don't have full name information for confidentiality reasons, and the only place I was able to find a list of unique first and last names was from a company doing greyhat Facebook advertising and data mining [1]. Here is their list of the 100 most popular first and last names on Facebook in 2009 [2].

I then ran these names through a bulk domain availability search (where the name "John Smith" turns into "johnsmith.com"). Not a single .com domain is currently unregistered.

Of course, that test doesn't tell me much except that at the very edges of the full name distribution, most of the domains are taken. The bigger question is how long the tail is. If 98% of first name/last name combinations are not contained within the top 100, then your kid will be fine unless you do name them John Smith.

Since I wasn't able to get my hands on a dataset for full names, I decided to approximate the frequency of firstname + lastname duplicates based on individual surname duplicate frequency in the US. Because surnames have fewer letters than first names and surnames combined, there are fewer possible combinations. This means that there should be a lot more duplicate surnames than duplicate first name/surname combinations. So, if it turned out there was a lot of surname repetition, I wouldn't be able to conclude much about how common first name/surname combinations are, I would just know that they are less common than the data I can see. However, if duplicate surnames turn out to be relatively uncommon, I could conclude that first name/surname combinations are more uncommon still, and disprove my suspicion that the uniqueness of my child's name is important to their digital future.

Among surnames, the top 100 most popular constituted 16.4% of the US population, the top 1000 accounted for 38.9%. That means that the majority of surnames lie in the long tail rather than the head of the distribution. We can therefore assume that the distribution of full names is even more skewed towards the long tail since the domain space is substantially larger. This assertion is supported by the Facebook data above.

This Wikipedia graph suggests there were around 350,000,000 users on Facebook in 2009. The total number of names in the top 100 comes to around 592,000, which represents just .1% of users on the site.

The conclusion I have to draw is that unless you name your son John Smith and your daughter Sarah Smith, they will have perfectly viable digital identities available to them when they graduate from college and need to start looking presentable to the rest of the world. So rest easy; I know that I, for one, want to name my first child Ellis. According to White Pages Names, it seems it is suitably obscure, and hey, I notice the domain "ellissaines.com" isn't registered. I think I'll register that ... just in case.

Special thanks goes out to the staff of the Alden Reference Library at Ohio University for helping me obtain this link and other vital statistics I used in this blog post! Specifically Sherri Saines (@bibliosanity), Tim Smith, Kelly Broughton, and Cary Singer.

[1] The company was later banned from scraping Facebook, which is presumably why their data is so old.

Save Time, Cash, and Carbon with Amazon Prime

I have been a heavy user of Amazon prime for years, and it baffles me why anyone shops offline for anything but perishable grocery items. If shopping is something you enjoy as a recreational pastime, you can stop reading now, but for all the rest of us, read on to see why Amazon Prime is not just good for your budget, time, or carbon footprint.

Assumptions

Let's start by making some assumptions. First, I will assume that you earn exactly the national median for a fully employed person in the US: $32,140 which is approximately $15/hour [1].  I assume you spend money like a typical American as well, meaning that you spend approximately $7,423 per year on personal care, misc, alcohol and tobacco, apparel, and entertainment [2]. I assume you have access to a car [3] that gets the national average of 24.1 mpg [4]. Further, I'm going to assume that you are like 90% of Americans and live within 15 minutes of a Wal-Mart [5] which represents an 8 mile distance [6]. I will make the further simplifying assumption that you purchase all your goods at Wal-Mart. You value your free time at $10/hr or a 2/3 of your working wage. Your primary shopping objective is to save money on the goods you want and need.

I will assume that most of your shopping is done for relatively common goods, which is just to say that you are not trying to find a new Tiffany diamond necklace, you're shopping for things like shoe laces, t-shirts, and garbage bags. I will assume you shop about as much as the average American, or .75 hours/day [7], and that a typical shopping trip requires 2.16 hours for a total of approximately 127 shopping trips per year.

The Cost of Offline Shopping

Let's start by calculating the costs associated with your current shopping habits. 127 shopping trips per year means your car is being driven 127 * 8 = 1016 miles per year getting you to and from the store. The IRS mileage rate is $.56/mi [8], which means you are spending $569 in car depreciation every year to shop. Further, at 24.1 mpg, and an expected fuel cost in 2013 of $2.04/gal (so cheap!) [9], you're spending $83 on gas for those trips. The total cost to you in terms of automotive expenses is therefore $652/year.

Then there's the time cost. Shopping takes time away from doing other enjoyable things, like watching movies, taking walks, and eating with friends. In the above assumptions, we put the cost of your time at approximately 2/3 of your working wage, $10/hr, which means that you are paying yourself 274 hours * 10 = $2,740 to go shopping. So far, the total cost of shopping offline is $3,392 per year and that doesn't count any of the unpleasantness of fighting through weekend traffic, having to visit multiple stores (because remember, in this hypothetical example you only ever have to shop at one store), and finding a parking spot at an already overcrowded mall.

The Cost of Shopping on Amazon Prime

Save Time

With Prime, goods are delivered right to your door and the actual finding and purchasing of goods requires significantly less of your time. Consider a simple example to demonstrate the point: it would likely take you less than 15 minutes to find a common item like a serving ladle (not an affiliate link) on Amazon which is less than the one-way time required to get to a store. If you were able to avoid only 25% of your shopping runs over the course of the year, that would be 68.5  hours of your life back. We're talking about a 2.8 day's worth of time you can could spend playing fetch with your dog, socializing with friends, or reading interesting blog posts like this one. That's 8 days of vacation time from work. If you are aggressive about shopping online, you could easily avoid far more than 25% of your shopping trips.

Save Money

Prime costs $79/year (soon to be $99), restricts buyer choice, and recent studies have shown that Amazon can cost as much as 20% more than Wal-Mart [10]. But it's still a money saver. The average American is spending $7,423 on non-grocery non-medical purchases per year (see the CNN money link in the footnotes below). $7,423 * .2 = $1,484, add the $79 subscription fee, and it costs $1,563  more on average to shop on Amazon compared to Wal-Mart. If you shop at Target, Macys, or any other more expensive retailer, the cost savings is likely to be much smaller. Compared to the cost of offline shopping calculated above ($3,392), however, Amazon will save you approximately $1,829 ($3,392 - $1,563) per year in automotive upkeep, gas, and time.

In addition, the estimated 20% premium is an estimate based upon a relatively small sample size for non-Prime customers. So the above estimate is likely understates the actual savings.

Shrink Your Carbon Footprint

Finally, shopping online is better for the planet because it significantly reduces the energy cost of shipping goods to consumers [11]. The most efficient way to deliver goods to end users is on big trucks, planes, and delivery vehicles like that UPS truck with all the Amazon packages in it. True, such trucks get poor mileage per gallon, but they deliver hundreds of packages per run, and unlike your personal vehicle, UPS and other shippers are spending big money on reducing their fleet mpg [12].

Conclusion

Amazon Prime saves time, money, and helps me be a reduce my carbon footprint. The $79 might sound like a lot, but it's a drop in the bucket compared to the improvements in quality of life. And if you are like me and dislike shopping in the first place, all of these calculations are meaningless. I would actually be willing to pay more to avoid the mall, but Prime has conveniently given me a better option. You can check it out here (non-affiliate link): http://www.amazon.com/gp/prime

&nbsp;

[1] Personal income in the US: http://en.wikipedia.org/wiki/Personal_income_in_the_United_States

[2] This comes from the CNN money breakdown of American expenses: http://money.cnn.com/interactive/news/economy/us-spending/

[3] For every car in the US, there is 1.3 people, so this seems pretty reasonable. http://en.wikipedia.org/wiki/Motor_vehicle

[4] Presumably if it is newer, you'd be more likely to get good gas mileage, but I'm working with averages here. http://en.wikipedia.org/wiki/Fuel_efficiency

[5] It's actually pretty shocking that we all live so close to Wal-Marts, http://www.statisticbrain.com/wal-mart-company-statistics/

[6] This is an extrapolation from the statistic about living 15 minutes from a Wal-Mart. I do not have a statistical source to back up the conversion from 15 minutes to 8 miles, I'm just guessing you can't go 60 mph on the way to the average Wal-Mart from the average home.

[7] Yes, that means that the average American spends 274.13 hours shopping every year: http://www.bls.gov/news.release/pdf/atus.pdf

[8] 2014 IRS mileage compensation rate: http://www.irs.gov/2014-Standard-Mileage-Rates-for-Business,-Medical-and-Moving-Announced

[9] Expected cost of gasoline: http://www.nytimes.com/2015/01/28/business/energy-environment/after-steadily-falling-price-for-gas-notches-an-increase.html

[10] The study was conducted by Kantar Retail. Their sample size is pitifully small (36 goods) and doesn't take into account the fact that Amazon doesn't cover all goods with Prime shipping: http://www.bloomberg.com/news/2012-06-22/wal-mart-beats-amazon-prices-including-glee-dvd-set.html

[11] The study is actually focused on Buy.com, which presumably has a less efficient distribution network than Amazon: http://campustechnology.com/articles/2009/03/16/shopping-online-more-energy-efficient-say-carnegie-mellon-researchers.aspx

[12] http://compass.ups.com/ups-lightens-up-with-fuel-efficient-plastic-trucks/

Why You Shouldn't Follow Your Dreams

Back when I was in 9th grade, I purchased my first desktop computer. I did extra chores, mowed lawns, and shoveled snow to afford my dream machine. It came tricked out with a 900Mhz AMD Athlon Thunderbird CPU, 128MB of RAM, a GeForce 2 GPU, and a Plexdor CD burner. Housed in a bleak grey metal box, it was my pride and joy. Countless hours of Counter Strike, Total Annihilation, and Starcraft were played on that computer.

Despite my fond memories of that first computer, it was never the perfect performance machine. I had made many compromises to meet my meager budget, and the hardware was out of date less than 60 days after it arrived. So around that time, I vowed that someday I would buy myself the most ridiculous performance-intensive machine that could be had.

A while back I realized all of a sudden that I had more than enough cash to buy that no-holds-barred gaming machine. But desktop computers are no longer terribly practical. And I don't play Counter Strike in the evenings anymore. And my middling laptop runs the Adobe Creative Suite just fine. In fact, I spend most of my time in front of productivity and office applications at work, and in the evenings, I try to spend time away from the computer for the sake of variety. I have the cash, but I longer want that beast computer, and it's sad. When I explained my disappointment to a friend over lunch, he shrugged and said, "of course you wanted that computer, but dreams expire."

He was right. Since making my vow to buy an uber gaming machine, the dream had lost meaning for me, and like a mesofact, I had not updated my list of dreams.

This caused me to not only go through that list, but to re-evaluate how I pursue and value dreams. I have kept a bucket list for several years, but I here propose a short list of reasons why you shouldn't have one. Maybe you shouldn't even follow your dreams.

Destinations are boring.

As a society, we praise big dreams and tell kids to aspire to be astronauts. But as a society, we focus our dreams on destinations, not on the journeys that make them meaningful. I fell prey to this exact problem when we were founding my first startup. I wanted to build a successful startup. For three years, we worked hard, we got some lucky breaks, and these days Skritter is pretty darn successful. I have become an astronaut. And you know what? It's an empty victory, because as it turns out, what I really want is not to have a successful startup. What I want is to work hard on difficult problems with people I like. Notice the difference there: the first dream is a state, "success," the second dream is a process "work hard."

For all the glib discussion of life being a journey, not a destination, we overlook that fact when forming our deepest and most personal goals. Maybe you want to be a famous rapper, a skydiving instructor, or a polyglot. All of these are noble ambitions, but thinking about those dreams in terms of continuing actions rather than destinations makes them more meaningful. You don't want to be a rapper, you want to spend most of your time rapping in a studio. You don't want to be a skydiving instructor, you want to teach newbies how to enjoy their first jump. You don't want to be a polyglot, you want to spend time learning the nuances of dozens of languages.

Destinations are boring, and dreams that rely on them are hollow.

The exotic is kinda meh.

You know what sounds cool? Getting out of a Cleveland winter and working from Costa Rica for 2 months. The problem is that if you value a degree of consistency, recurring familial interaction, predictable diets, stable electricity and internet, reliable transportation, and a hundred other factors you probably take for granted, then it's actually not that cool. That was the story of my 2 months in Costa Rica.

Yet despite the disconnect, whenever I tell people about going to Costa Rica, most people said "I wish I could do that."

Daily life constrains choice sets to the degree that most people can't up and move to Costa Rica for 2 months. And for most people, that's a good thing. Dreams often take the form of overcoming the inertial forces that keep us grounded to the status quo. But it is precisely these forces that often make us happy. Many dreams are predicated on circumstances that by their very nature would make us unhappy.

Dreams made in a vacuum are meaningless.

I drive an extremely economical car and I've always dreamed of having a performance sports car. Recently I got the chance to test drive some exceptionally cool sports cars, and coming back to my little Suzuki Aerio was a relief. Why, you might ask, would I prefer my 1600cc putsy hatchback to a Mercedes Benz E55? I won't go into details, but in forming my dream about owning a sports car, I ignored all of my previous preferences. A Porsche looked cool, but in my daily life I prefer gas economy, reliability, and low insurance bills.

Dreamers are encouraged to think big, and that often implies "outside of our experience." This gets back to something Paul Graham suggests about finding what you love to do: if you aren't tinkering with computers in your free time, do you really want to be a programmer? If you didn't sacrifice to buy a sports car when you were 17, are you really invested in performance automobiles?

We often classify our unqualified aspirations as "dreams" and then foolishly work towards them. Daily experience is often a far better judge of what you will enjoy than a groundless idea of you got from society.

Summary

All of this might lead you to conclude that I think dreams are meaningless, but that's not the case. I think dreams are extremely important, but they often grossly misrepresent what people actually want from their life experience. I am still struggling to find a good way to pursue my own aspirations and would be interested to hear if anyone else has had luck better forming and achieving meaningful dreams.

Magic Tricks

Back in my junior year of college, I switched my major from Cinema Studies to Economics. I was sitting in the office of my favorite professor and adviser, a man who had his Economics degree from Harvard. I was asking about post-graduate options.

"I was thinking about maybe getting an MBA one day. Do Oberlin graduates stand a chance of getting into the Harvard MBA program?"

My adviser smiled in a good-humored way. "Honestly George, I don't know how anyone gets into the Harvard MBA program. I think you have to be magic to pull it off."

When I heard those words, my heart sank, but I took the advice to heart. Seven years and two startups later, I still vividly remember that conversation because it crystallized my understanding of how to pull off a seemingly impossible accomplishment: it's a magic trick.

Getting into the Harvard MBA program might embody a magic trick for most people, but magic tricks encompass anything that appears to be impossible on the surface. Retiring at the age of 30, traveling full time while working remotely, being a C-level executive at a Fortune 500 company, or knowing people of repute. All of these accomplishments work to confound explanation and increase the perceived importance of the speaker. After all, only 1 in a million spend our evenings rubbing shoulders with A-list celebrities and our days on the cover of a business magazine. It's unique, it's interesting, and it defies simple explanation.

But it's a mistake to write off such accomplishments as impossible. It's an even bigger mistake to write them off as unimportant or shallow.

I've learned the hard way that when you are an entrepreneur, it is extremely important to quickly impress people you meet. When all you have is a one-month-old company and a smile, people write you off unless you can quickly portray success. Maybe you didn't go to Harvard or MIT, but the more magic tricks you have under your belt, the more investors, customers, and peers will assume you can accomplish your next audacious goal. This is the first step in actually living your life like you want.

Which bring me to choosing magic tricks. I've used common examples above, but your achievements will be most impressive if they are your own. Maybe you want to start a nonprofit mentoring program and positively impact 1,000,000 kids. Maybe you want to make the next greatest search engine. Maybe you want to replace Netflix and bring the world a decent selection of on-demand movies. You need to define what you want to achieve, and then do it. The coolest thing about magic tricks is that they are cumulative: the more you've already done, the more people will believe you can accomplish the next one.

After all, if you've already started the first commercial space program or played basketball with president Obama, or even been accepted to YCombinator on stage, then you can probably do whatever you want.

It's Sad to See Your Startup Turn into a Business

This post was originally published on 2/1/2013.

When I founded Skritter in 2008 with Nick and Scott, we called it a startup. We raised three rounds of funding, hired developers to help scale our team, and attended startup summits, venture capital panels, and meetups filled with aspiring entrepreneurs working on the next big thing. As with all young startups seeking capital, our business plan growth model had us making 30M in profit in 3-5 years as we took the language learning world by storm.

Four and a half years later, Skritter has become a viable, successful, growing company. We have three employees in addition to the founding team, and have provided employment for twice that many along the way. We've proven that our business model generates profit, that it adds value to customer's lives, and that we can achieve product-market fit.

But somewhere along the line, Skritter stopped being a startup and became a business. And while I am deeply proud of our achievements, the change makes me sad.

When you run a startup, you dream big, you think in terms of conquering entire new markets, challenging entrenched competitors, and changing the world in a big way. You work hard, play harder, and forge lifelong relationships with your co-founders.

Businesses, by comparison, are more modest and mundane. Businesses tend to know whom their customers are, they have a good sense of what makes money and what doesn't, and they don't make a habit of re-investing every penny to try and shoot the moon with a new product. Businesses are like middle aged fathers who just want things to run smoothly without too much fuss. Startups are their star-struck sons spouting poetry to their lovers in moonlit gardens.

Startups are just more exciting, vibrant, and entertaining.

But they also have this frustrating tendency to fizzle out, fail, or explode catastrophically. Founders lose their shirts, relationships are ruined, investors are burned, and once stable, gainfully employed founders end up in their parent's basements applying for jobs to cover their credit card debt.

I'm proud of what Nick, Scott, and I have built at Skritter. I'm proud we achieved the dream of building a profitable company. But if you've ever been there for the startup part, the irrational giddiness you get from building something new, you'll know instinctively what I mean when I say it's sad to see your startup turn into a business.

Why You Should Bootstrap Your First Company

A while back I read Daniel Tenner's excellent article entitled Taking the Leap. Having run my own modestly successful startup for going on 7 years now, I can say with some authority that he makes excellent points. But one thing about the post bothered me: his advice is most applicable to your first startup. That distinction is critical.

Hacker News idolizes people like Steve Jobs, Elon Musk, and other visionaries who take incredible risks in the face of absurd odds. Their stories are dramatic, and it's delicious to read stories of people who buck the system and succeed. But it is a disservice to the less experienced to omit the beginning to every success story: the small successes they had early in life.

The men who mine asteroids and build electric sports cars don't start with those ventures. To illustrate my point, I'd like to tell a quick story.

Back in 2008, my cofounders and I were going door to door trying to raise a minuscule amount of funding. One of our business advisers gave us an introduction to a successful founder turned angel investor who had just sold his company. Everyone was talking about how successful he was, but over the course of developing a mentorship relationship, we heard about how he  got his humble entrepreneurial start. He did it by selling asbestos file folders to legal consultancies at a time when everyone was going digital. The business model was in it's death throes, but he was able to generate enough profit to reinvest it on his next company.

Read that again if you missed it: our visionary angel investor got started selling fireproof file protectors to lawyers who wouldn't need them in a few years.

This sort of story is far from isolated. Success begets success. Elon Musk didn't start with Tesla, he started by selling a $500 computer game called Blastar at the age of 12.

Don't try and shoot the moon on your first startup. Bootstrapping reduces the upside of your ventures, but it also reduces the risk that you'll fail. Daniel Tenner has it right: keep your head down, reduce your burn rate, and if you succeed doing that a few times, Mars, cold fusion, and hover bikes will still be waiting.

Your Startup Need an Intractable Design Problem

Since 2008, I have been tasked with designing and UX testing the entire website for my first startup. From the outside, Skritter is pretty simple, it's a website that teaches students of Chinese and Japanese to better learn and remember their characters. Basically a flashcard program for Chinese and Japanese.

Seems pretty simple, right?

Well, not at all, actually. The problem is that Skritter uses a spaced repetition algorithm that makes it non-obvious to add and manage vocabulary. Unlike a standard flashcard program, you don't just add words to your library, the application does that for you gradually, as you learn and remember more content. So on Skritter, you can't "study" a list, you have to "start adding from" a list. The difference creates all sorts of problems for users. Where are you in the list? What haven't you learned yet? When you stop adding from a list, should we remove all the material you've already added? Spaced repetition makes Skritter powerful and useful, but at the cost of simplicity.

For years I railed against this unintuive aspect of our product. Apple, Dropbox, and a hundred other companies were making it big by making it simple. Although Skritter eventually became a big success, it took a while for people to "get it" and it was an exercise in frustration for me as the product designer.

So I was thrilled to start work on a new startup in 2013. CodeCombat offered me the opportunity to build upon the design lessons learned at Skritter, but in a different arena: game design. We were making a game that taught users to write JavaScript.

Sounds pretty simple, right?

Not at all. Although there are hundreds of educational games teaching everything from typing to math, surprisingly few teach users to code. And those that do serve more as antipatterns than examples of successful design. For CodeCombat, we couldn't just rely on typical game mechanics, because we were supposed to be teaching users how to code the very behaviors most games take for granted like move up, for instance. As a result, simple things like unit selection, setting waypoints, choosing actions, and resource allocation turn into non-obvious design conundrums.

When I was working on Skritter, I used to think, "Someday I'll be able to work on a product that's simple and obvious, boy that'll be sweet." But CodeCombat taught me is that if you aren't running into seemingly intractable design problems, that's a strong indicator the product isn't solving real problems.

There are products out there that won by simplifying a complex problem; perhaps it's file uploads, or listening to music, or sharing photos with friends. But it's a mistake to assume that because the end result is simple that it was simple to design. As Apple has proven time and time again, making things simple is extremely difficult.

So, if your startup isn't solving an intractable design problem, find one.

Minecraft Isn't Educational

I have spent the majority of my professional career building edtech products. First I taught tens of thousands of students Chinese and Japanese, then I taught millions of kids to code. I know a lot about building educational products; games in particular. As a lover of Minecraft and an edtech game designer, I'm here to tell you that Minecraft isn't an educational game.

For those not familiar, Minecraft has several game modes none of which are games in the sense of having levels, bosses, missions, and achievements. Minecraft more closely resembles a digital sandbox with varying levels of abstraction. Survival mode is a bounded sandbox with randomly generating baddies. Creative mode is digital Legos. None of Minecraft's game modes explicitly teach the player anything. That's right: there is no educational content in Minecraft whatsoever. There are no lessons, tutorials, grades, or tests, there is no backstory, no plot, no puzzles, no brainteasers, riddles, math, or history. Nothing in the game tries to teach anyone anything. 

I hear you crying out "but Minecraft holds kid's interest long enough that they learn to mod the game, or build simple circuits, or build historic structures. Surely that's educational!"

But take a look at that line of reasoning again: the advocates don't claim that Minecraft teaches anything. They claim that kids like it enough that they may end up teaching themselves something unrelated while playing. The girl who likes computers learns enough Java to mod the game. The boy that likes building things constructs interesting structures. But to say that learning in the pursuit of addictive entertainment is educational is sloppy and unfair reasoning. By that same logic, Grand Theft Auto 5 is educational because some kids get so into it that they memorize the geography of LA to minimize transit between missions [1]. 

The reason that parents, schools, and kids call Minecraft educational is that it combines the addictive behavior of video games with the least offensive content imaginable. What learning occurs in the course of that addiction is labeled educational, but is no more useful to kids than anything else they voluntarily spend equal amounts of time on.

Why does the distinction matter? Because it's misleading educators and game designers. Spoiler alert for people making edtech games: there's very little to learn from Minecraft because as I mentioned above, it doesn't teach anything. Spoiler alert for teachers: Minecraft won't teach your students anything useful [2]. 

I love Minecraft and have played for much longer than I'd like to admit. So has my wife. So has my brother. As a game, it's great; but as education, it's no better than World of Warcraft. If you want your kid to learn, you'd be better off letting them follow their interests and educating themselves.

[1] Yeah, yeah, "Los Santos." Everyone knows it's LA.

[2] Even though it will keep them entertained for a class period with little to no chance that parents will complain.

Call Me When 3D Printing Becomes Practical

This was originally posted on 1/4/2013, but I'm still a skeptical grump about 3D printing.

Over the holidays I finally got around to reading Wired's effusive article about Makerbot and the coming 3D printing revolution. I get it: 3D printing is going to take over the world. It's going to eventually let me download a car, and that's very cool. But in the interim, 3D printing appears to be nothing more than a distraction.

I want to own useful, practical, and cost-effective things. Making a plastic belt buckle, or RC plane wing, or clothes hanger isn't terribly compelling. And sit-around items like action figures don't meet the practicality rubric. Even if I were interested in making these things, I wouldn't want them made of plastic. For most US consumers, plastic is a poor substitute for the metals and alloys that we have come to expect in quality consumer devices. The real clincher though, is the ready availability of superior substitutes. I'm busy enough that learning to use a CAD program to create a plastic sub-component of an equivalent metal device I can purchase in a fully functional form for $10 on Amazon just doesn't make much sense. And I'm guessing that I'm not alone here.

The revolution in 3D printing is going to come when disinterested folks like me can download, customize, and effortlessly create complex products from the comfort of my own home without having to become proficient in CAD software and the vagueries of 3D printing hardware.  Right now 3D printing is like the personal computer market in the late 80s; it has explosive growth potential and the possibility to disrupt our system of commerce right down the foundation, but it's all but inaccessible to anyone but engineers sporting the modern equivalents of pocket protectors.

I bothered to right this not to slam companies like Makerbot or tear down gushing writeups like the one I read in Wired. Makerbot is doing great work and Wired always gushes about new tech as though it will single-handedly bring about the singularity tomorrow. But until I can download that car I was talking about, articles about 3D printing are just distractions.

Effort and Reward: Correlation, Not Causation

A few weeks back I was watching The Queen of Versailles, which is an independent documentary about billionaire timeshare mogul David Siegel and his quest to build the largest house in America. At the beginning of the movie, the director interviews Siegel about how Westgate Resorts got it's start. He talked about how he founded the company when he was young and naive, worked like hell, and managed to grow the company to billions in sales. What struck me about his description was how similar it is to the way I describe working on my first startup, with one critical difference: Skritter is a tiny bit less less profitable.

This got me thinking about the nature of effort vs reward. It's a common misconception among entrepreneurs that the harder you work, the more successful you become. This workaholic mentality causes people to sideline important aspects of their lives to maximize a perceived chance to make it big. I believe that effort and reward are correlated. If your goal is to become a millionaire, you are far more likely to reach your goal working very hard on a bunch of ventures than if you stay at a job that allows you to relax and coast. The age old motto "God helps those that help themselves," seems true.

But the the amount of reward you enjoy for your effort is randomly distributed. If it weren't, David Siegel of Westgate fame would have had to work a thousand times harder/longer than someone whose startup makes $1M/yr. Since that clearly isn't possible in a normal human lifespan, I'm forced to conclude that there is a big component of luck involved in the rewards anyone reaps from their efforts.

Having a deep understanding of that fact is important because it helps workaholics like me from over-investing in work. The truth is that I probably stand about the same chance of retiring early from a new venture whether I pace myself or work 100 hour weeks. Effort is important for success, but marginal effort is just that, and it seems a horrible waste to labor under the delusion that another few hours of work will be the difference between a decent living and early retirement.

Let's Partner Into Prosperity

Most business partnerships are a waste of time. Guy Kawasaki says so, Paul Graham says so (see the section at the bottom), and I have learned from personal advice that both men speak truth. The thing is, partnering is most appealing and dangerous to a startup early on. In those critical months and years where credibility is scarce, partnerships seem to offer a quick path to legitimacy and (your partner will lead you to believe) wealth. So it's imperative to develop resistance and skepticism to partnership offers. But how?

Well, one method stumbled right into my lap recently. This is a spam message that I received last week:

"LET'S PARTNER INTO PROSPERITY:  Kudos!!! You've got a very good work going here. I've been contracted to develop a website and a phone application that can help people in a particular Country to learn their three different dialects. It's a multimillion $ Project to be funded by the Government. I understand that a lot of scamming bullshit is going on online but you won't need to spend a dime of yours, all we need is the service of a person that has the knowledge required which would be magnanimously remunerated. I don't know much about language software design, if you do or if you know anyone that can partner with me on this please mail me now without any delay: address@yahoo.com Do you have a website? If yes, what's your website? I'm waiting ... Success!!!"

It's got all the elements of a bad partnership: vague intentions, an appeal to the legitimacy of some large organization (the Government!), a nod to skeptics, and call to action. My advice to you: the next time someone proposes a partnership, simply tack "... Success!!!" to what they say to remind yourself that most partnerships are a waste of time. What's scary is that many seemingly legitimate partnership offers are more dangerous than this example because they lure you into wasting time on them. At least in this case I can just click delete and get on with my day.

Why You Can't Admit Personal Mistakes on the Internet

My uncle is an entrepreneur-turned-corporate executive and has achieved considerable career success in his life, which is just to say that he's imposing and has a track record of getting what he wants.

One summer between my junior and senior years of college he hired me to help landscape his lawn. Several weeks after I had started, I made a small mistake.

"Oh, sorry about that, I'll just --"

"George, never say you're sorry" he said.

"Sure, I just meant that --"

"I know what you meant, but never say sorry, it makes you appear weak. Instead of saying you're sorry, fix the problem." He paused for a moment, looking at me, and then went back inside.

His advice ran so counter to what I had been taught that it stuck in my memory. I had always been taught to admit mistakes and correct them, but here he was suggesting I skip the first step. I'm still not sure I agree with it completely, but like every good over-generalization it has a nugget of truth: admitting a personal blunder to someone who has no empathy for you is ill-advised. That summer was all about me becoming employable, so I'm pretty sure my uncle meant that I would appear weak to a boss or employer, but the rule holds doubly-true for internet readers.

In the past, I have been tempted to write blogs about my goof-ups. I have almost always decided not to write such posts, not for want of subject material, but because they don't have any upside for me or my company. In the best case, people will think I'm humble, in the worst case they will think I'm incompetent. Simply put, I'm not willing to risk being labeled the latter for the former. And neither should you.

Self-aware people admit failure to get constructive criticism and seek catharsis. You aren't likely to get either if you bare your soul to the anonymous masses on the internet by first saying "I'm sorry."

Hypothetical Number Inflation

I read a lot of Hacker News and it always strikes how big the numbers are in other people's blog posts. When people write about a big exit on HN, they talk about $100M not $500,000.

When I was in college, and especially when I was running my first startup, reading those articles made me feel like a failure. From what I could tell, the world was populated exclusively by companies "struggling" to break $10M in yearly sales, and founders learning how to manage similarly "modest" liquidity events.

Now that I've been blogging for a few years, I can say from experience that most of those numbers are bogus. Survivorship bias aside, bloggers suffer from what I like to call Hypothetical Number Inflation.

When I write about a topic, I want to make a point, and the point is very rarely to be realistic about ordinary business metrics. I might want to prove that the relationship between effort and reward is correlative not causative, or that bootstrapping your first business makes sense. These are opinions that I've thought through and believe, but it helps to rally some ballpark numbers to make the case.

And therein lies the problem: in the course of arguing a point, it behooves me as a writer to push my numbers to the logical extreme to avoid losing my readers to unrelated quantitative niggling.

If I want to provide an example of a successful business in a blog post, I want the example to be unequivocal. If I choose a number that's too low, the effect is like Dr. Evil asking for 1 MILLION dollars, readers stop and think "wait a minute, $1,000,000 isn't successful to me." When that happens, I lose that reader before they hear the entirety of my argument.

Since the HN crowd is so affluent (or at least says that it's affluent), writers that want to be taken seriously need to choose hypothetical numbers that boggle the mind and leave no doubt as to as to the writer's intent.

So, if you get frustrated when you read authors talking about "small" exists in the mid-8 figures or "low" executive compensation in the high 7 figures, remember that these numbers don't represent the median, but the extreme. Better yet, decide for yourself how many zeros constitute success or failure and ignore writers like me.

Now You Are a Geek: Saying Goodbye to Wave

This was originally published on 2/2/2012.

Google Wave closed write access several days ago. This had been on my radar for some time, but as a heavy user and advocate of the service, it's difficult to see it go.

Seeing a pet project die is a right of passage for a career technologist. It's not just that it's a coming of age event, it's an important part of becoming better at building and improving ideas.

When I was young, I didn't think about technology or specific projects as transitory. I had such a short period of reference that everything was effectively permanent. I used Altavista for search because that was the only search engine that had ever existed and I played Descent 2 because it was the best game. I spent a lot of time playing games, trying operating systems, and installing utilities without thinking much about the people behind those monoliths of time and effort.

When you don't have an ownership stake in what you consume, it's easy to remain agnostic about games, operating systems, programming languages, server platforms, SDKs, editors, project management styles, deployment techniques,  and a hundred other topics about which technologists care deeply.  As I transitioned from being a consumer of web technology to a producer, I began forming opinions about the tech that surrounded me.

I love Google App Engine, dislike AWS, love Python, hate Java, dislike Windows, but think Mac OSX is for snobs, snub Firefox for Chrome, love Winamp, hate iTunes, loath antivirus software, prefer Google hangouts to Skype, and a hundred other subjective preferences. For having so many pet technologies, I have seen remarkably few get abandoned. Among those that I have lost, Wave has been the most important.

I was enamored with Wave because of the ideals and dreams of the people who invented it. I don't know much about the creators of Winamp [1] but I could put a face to the Wave development team and I wanted them to dethrone email, destroy chat programs, and better organize all web communication.

The fact that it failed makes me more of a real geek. I invested heavily in Wave and will now pay the price [2]. But in going away, Wave has also taught me some valuable lessons. I've learned that even well-funded projects die, technology doesn't often win out against established behaviors, and evangelism can only go so far.

The pace of innovation in web technology is accelerating and there is a tendency to avoid investing in any platform, but I think that it is the mark of a mature and invested hacker to have a small cemetery of pet technologies to grieve for. Don't hang on and be that backwards guy who wants to implement everything in Pascal, but remember what you've lost and use it to build better products in the future.

[1] This is because I wasn't old enough when Winamp was disruptive and being talked about, I have heard anecdotes that suggest it is quite an interesting story.

[2] Endless exporting and unsorted data migrations await. There is a strong temptation to accept the data loss and start over on another platform, but there is data in Wave that I simply cannot afford to lose.

Are Titles Important at a Startup?

My two co-founders and I signed our first company into existence on a hot summer day in 2008. Prior to signing the document, we had to decide what our titles would be. Our lawyer assured us that we could give ourselves any titles we wanted and it would not have an impact on the legal process. We thought it would be fun to sport executive titles, so we each chose a grandiose executive title and then returned to our shared apartment and kept hacking.

After several years at both companies, only the CEO title stuck. Although we remember what job titles we each picked, they ended up being largely unimportant. 

Size Does Matter

Organizations give titles to members to help people outside the organization understand who they are speaking with. When you get interviewed for a job, you want to make sure you aren't talking to a sales associate, you want so-and-so from HR. When a member of the press deals with a company, speaking with the CEO lends more weight than if the representative is a summer marketing intern.

But the orienting power of a title breaks down when the organization being represented is small. If you are being interviewed for a position at a 5 person startup, it's less important who you talk to because you can be more certain you are talking to someone of importance. Similarly, if a member of the press talks to one of a three person founding team, it matters less what title they chose and more that they are one of the co-founders.

For small startups, job titles are less important, if not entirely meaningless--except for the title of CEO.

Head Honcho

So why does the CEO badge remain important even at a tiny company? At both of my companies, we made it clear to our customers and business partners who we were, and we didn't made any attempts to look larger than we are. So at least in theory, an email from any one of us should have carried equal weight. Internally this was true, but externally, it remained useful to call one person the head honcho. Even in a super-small startup, it helps clarify to outsiders who the organization recognizes as the leader or point of contact.

None of us made business decisions without unanimous consent, we all checked each other's commits, and we all read each other's support emails. But startups deal with thousands of people in their lifespan and it's easier on everybody to avoid explaining egalitarian cooperation and simply point to one guy and say "he's boss."

This doesn't mean, however, that tiny startups can get away without a clear internal division of labor. Titles need not be assigned, but somebody has to know it's their job to file this year's tax return on time. At both of my startups, I was referred to as "head businesser." Nick was a "programmer," and Scott was a "programmer and accountant." The fact that Nick and Scott were labeled CTO and CFO on paper was only superficially important. Their understanding of their roles at the company, however, was of paramount importance.

Conclusion

If you are starting a startup and are wondering whether you and your cofounders should have titles, I would give this simple advice: decide who the CEO is and stop worrying about other three letter acronyms. Incidentally, this is exactly the advice that YCombinator tells it's founders.