Nothing Ventured, Nothing Gained

A collection of essays by Matt Barrie

Crowdfunding

The Internet has reached a tipping point in the last year or so now that two billion people are online. Two billion potential customers, coupled with the ability to distribute a product or service instantly, which has resulted in revenue growth rates of companies that is unprecedented in history.

This has also led to the tipping point for a number of new breed one-to-many business models; outsourcing begat crowdsourcing, and now crowdfunding is emerging into prime time.

Kickstarter, a site where “projects” can post a snazzy video and seek funding from the public to make it reality, has had such success that it apparently funds more arts projects in the US than the National Endowment for the Arts. I personally think it’s an incredible site for many sorts of companies that would never get a look in from venture capital. In particular, this is the new frontier for many hardware or gadget companies that would never get through the door at a traditional venture firm due to the deep burn, risk and high capital cost of this genre’s profile. Although, these businesses can still be lucrative if you get it right; even a one trick wonder like GoPro recently obtained a $2.25 billion valuation when FoxConn bought in.

The really, really great thing about Kickstarter however is that you’re not selling equity or raising debt- it’s all pre-sales. This makes it an incredible avenue for companies to pursue (just wait until the music industry catches on); although the “investors” are just starting to discover now that given the model is a few few years old, that Kickstarter isn’t a store, and that venture capital is a hard, long term portfolio play sort of gig. Most startups fail, and consumers aren’t really used to paying for something on Amazon, only to be told 18 months later after two delays the product isn’t going to ship.

Which brings me onto the next frontier of crowdfunding; crowdfunding equity in startups. The US is heading this way one year into the JOBS act, and other countries are following suit. The end game has the potential to disrupt the entire venture capital industry and turn GPs into stock pickers.

Ultimately where I think the venture industry is heading long term is to some sort of a mainstream model based on a stock exchange, where Joe Public will be able to invest in early stage, speculative technology stocks. This is no different to what already exists in the mining industry, for example, with exchanges like the TSX or ASX, where you can get a listing away with a spread of 400 investors holding $2,000 of shares each, at a market capitalisation of only $10 million. If you’ve been following what’s been going on with SecondMarket and Sharespost, it certainly seems to be heading in that direction, with Sharespost and Nasdaq recently announcing Nasdaq Private Market. Technically, there are exchanges today like London’s AIM or the TSX Venture Exchange where you can list an early stage technology company, but to date they haven’t been a spectacularly successful place to do business due to it being relatively costly, the liquidity not being there yet and the pressures of being a public company such as the focus on short term financial performance and continuous disclosure rules. Perhaps Nasdaq Private Market will find a way to synthesise a low cost, liquid private company technology exchange that will work because it can find a balance here.

You only need to glance at Kickstarter for a few minutes to see that startups are going to love raising crowdfunded equity. It will suddenly become really easy to raise seed stage funding if you can put together a good pitch, especially because . Not just that, but I would expect the terms to be remarkably better than many early stage venture capitalists would offer since the company itself would likely do the drafting of the documents (which would probably head towards an industry standard set of docs like the YCombinator Series AA Equity Financing Termsheet).

Although a few VCs might initially feel threatened about equity crowdfunding, I think they are going to end up loving it too. For a start, more startups are going to be successful at raising funds which means dealflow for VCs, which is great. But more importantly, VCs are going to absolutely love equity crowdfunding, because they will discover pretty rapidly that when they come into a later round that they won’t be negotiating with any investor counterparty.

Therein lies the rub- who exactly is going to represent the interests of 500 investors holding $2,000 worth of shares each? The model works on a regulated exchange because companies usually only list one class of stock- ordinary or common stock (although there has been an increasing trend with ). If you are successful enough to get your company to the point of listing, there’s a set of clauses in your typical financing documentation under the heading “Automatic Conversion” that kick in:

Automatic Conversion: All of the Series A Preferred shall be automatically converted into Common Stock, at the then applicable conversion price, upon the closing of a firmly underwritten public offering of shares of Common Stock of the Company at a per share price not less than [three] times the Original Purchase Price (as adjusted for stock splits, dividends and the like) per share and for a total offering of not less than [$20] million (before deduction of underwriters commissions and expenses) (a “Qualified IPO”). All, or a portion of, each share of the Series A Preferred shall be automatically converted into Common Stock, at the then applicable conversion price in the event that the holders of at least a majorityof the outstanding Series A Preferred consent to such conversion.

This means that every shareholder is in the same boat, rowing in the same direction after you list. This is opposed to the traditional venture model which is a nuclear arms race of stacked liquidation preferences, vetoes and control rights. With a piecemeal approach to introducing equity crowdfunding, I fear that the ultimate loser here will be the general public, who will discover a few years down the track all the tricks of the VC trade which has been traditionally structured as last in-first out. This is why the golden rule of being a VC is to never run out of dry powder, and why a venture capitalist will always keep funds in reserve for follow on financings. You see, when you have a loss making venture, the golden rule can also be translated roughly as he who has the gold rules.

If you haven’t been through multiple rounds in a venture financed company before, you won’t know what it’s like to go through successive rounds of dilution with the weight of layers of stacked liquidation preference on top. While senior liquidation preferences are coming down, in a crowdfunded world of dumb money I wouldn’t bet on them staying down. You see, if in the Series A someone comes in with $5m on a pre-money of $10m with a senior participating liquidation preference of 2x, that means that when the company sells, that 2x the money invested (or $10m) is returned first, then all shareholders (including the investor) share the proceeds. When the Series B comes in, they will want a liq pref at least on par with the As, and since round size usually increases outside of a cram down you might be looking at say $20m invested on a 2x. So a total of $50m now is taken off the top, before everyone shares. So if the business sells for less than $50m, the investors take it all. Liquidation preference was originally designed to return cash to financial investors first in a wipeout scenario- with non participating preferred, at an exit the investor doesn’t participate after being paid back their multiple, and so much choose between that capped return, or convert into common and share in the upside return with the rest of the shareholders. It has since been warped in some circumstances provide double dipping for venture returns in upside scenarios. I’m not sure the crowdfunded money is going to know how this works, and with investments capped at 2% of earnings or $10,000, I am not particularly sure if the investment is going to be meaningful enough for the dumb money to care, either.

This is just one trick of the trade. There’s plenty of others where the company itself might be complicit in moral hazard. For example, some companies often top up a management team during a financing round with stock options, which leads to further dilution. How are follow on financings going to be managed from the perspective of the crowd in a private company world to ensure that the public isn’t going to be axle grease steamrolling venture fund (and founder) profits? In the absence of continuous disclosure or mandatory financial reporting how are voting thresholds going to work? With public companies, a minority bother to vote on critical decisions such as board remuneration (sometimes they don’t even bother to vote in a takeover!), but the protective mechanisms are set up in such a way to deal with that. How is the public going to even going to be able to form a view of what their stake is worth?

There’s tons of other tricks that investors and management can play with the capitalisation table- at the end of the day, in the absence of running on a regulated exchange, who is going to stand up for the rights of the crowdfunded equity holders in these discussions, or will this dumb money become the gearing for venture funds?

On top of all this, venture capital is a hard, long term portfolio play. . For the economics to work, basically one of your portfolio needs to be a real knockout event that’s capable of returning all the funds raised from your limited partners in one go. Half your investments will return zero, a couple might exit for positive returns if you’re lucky and the rest will be liquidations returning cents int the dollar. Founders will love crowdfunding, VCs will probably learn to love it, but the crowd is about to play this game of poker blindfolded at a table full of pros. The worst thing about this, is that they will probably only discover this three or four years down the track, when the first crowdfunded companies start to exit.

Online Apocalypto

I observed with schadenfreude the complete debacle we call the traditional Australian retail industry as they screwed up a pathetic attempt at an online sale a couple of weeks ago. Over 150 retailers participated in “Click Frenzy”, paying up to $30,000 to a third party in an attempt to mimic the CyberMonday sale that racks up over $1 billion in sales each year in the US.

At 8pm, I checked the ability of the traditional bricks & mortar retailers to run a one day online sale in the year 2012. The tally kept on Facebook:

  • Big W: Down.  JSP Processing Error.
  • Harvey Norman: Down. “Sorry, we are down for maintenance. We will be back online shortly.”
  • Myer: Down. Timed out.
  • David Jones: Down. Site unable to serve content.
  • ClickFrenzy.com.au: Down. Serving 504 errors.

One of my friends wryly commented: “Amazon: UP!”.

I wasn’t sure whether our traditional retailers were completely and utterly incompetent or whether they thought they were being evil marketing geniuses by taking a leaf out of Apple’s playbook with a stage managed “ZOMG@! WE SOLD SO MUCH OUR SERVERS CRASHED!!”.

Heaven forbid, a little bit of predictable traffic comes in, and their websites that normally rank 70,000th on the Internet blip into 20,000th position for a day. It’s not like they could have seen that coming by looking at say, last year’s Christmas traffic where they also temporarily surged into the top 20,000. Either way, in today’s Internet it’s still a piddling amount of traffic.

It might be funny if it were 1997. Unfortunately it’s 2012, and the only thing I hear is the deep rumble and choo choo of a freight train coming, and it’s called the Amazon Express. Frankly, I can’t wait until it lubricates the tracks with the bones of our pathetic, overpriced joke of a retail industry.

We’re getting absolutely ripped off here in Australia. I can order a book from Book Depository in the United Kingdom for cheaper than I can find here in Sydney, even though it also gets printed here! Not only that, but it will arrive in two days with free international shipping. I can buy a shirt that would sell locally for $300 from Hawes & Curtis, also based the UK, and pay all up, including freight, $33 a piece. And this is coming from the UK for heaven’s sake, not China!

Take this for a beauty: when the video game Crysis came out, it sold in Hong Kong stores for $50 AUD, but in Australia sold for $100. The icing on the cake? Both versions were manufactured in Australia, with identical versions of the game box and disc, right down to the Australian censor ratings on the box!

Everyday in the news, I see either Gerry Harvey or Solomon Lew bleating on about how the economy has “run out of time” and that Australian retail is uncompetitive because the $1000 GST-free threshold on overseas imports is too high.

Sorry guys, that’s not the reason why you are uncompetitive. And, as the NAB report on Monday pointed out, 74% of Australia’s $11.3 billion in total online sales actually comes from domestic Australian businesses. Some new entrants, like Gerry Harvey’s arch nemesis Ruslan Kogan, are doing quite well. It’s just that for years you big bricks & mortar retailers were arrogant, and had non-existent or poorly functioning website presences because you didn’t want to change consumer behaviour to buy online as it would threaten all your legacy infrastructure.

Compare and contrast this to the new face of US retailing. While David Jones and Harvey Norman can’t even run an online bake sale, Amazon has been busily building the biggest cloud computing infrastructure.. in the world.

Frankly, I’m not sure why Gerry cares about the GST, after all he’s been pretty smart. Gerry doesn’t really sell anything anymore as he’s parlayed all the retail risk onto his franchisees and really just runs a mini-Westfield. That’s why the stock price is trading at the book price of the property. The only problem for Gerry is that the rent is propped up by assistance packages and I can’t possibly think of a new entrant retailer who would move into the property once the Harvey Norman stores vacate.

As for the phony GST argument- be careful of what you wish for. In the United States, bricks & mortar retailers complained for years with similar poppycock that Amazon had an unfair advantage due to a 1992 Supreme Court ruling which stated that only firms with a physical presence in a state are required to collect taxes from residents. So Amazon could sell a laptop for $1,000 that physical retailers sold for $1,100- and rely upon consumers to file a “use tax” with their local state for the extra $100, which of course they never did. Sound familiar?

Fed up, Amazon dropped the fight and suddenly turned around promising to invest hundreds of millions of dollars and hire thousands of locals in high tax states like California and New York. If they had to go to the trouble to collect the tax, they may as well go all out. And they did, building gigantic distribution centres right next door to these populous centres, and in doing so are destroying the sole remaining advantage that local retail has: instant gratification. Using robotics from Kiva Systems, a company specialised in automated logistics that Amazon bought for $775 million back in March, we’re now talking same day delivery. Witness the last asphyxiating breath of traditional US retail.

Now Sol and Gerry, in case you didn’t notice, earlier this year there were rumours of Amazon scouting the property around Sydney and Melbourne for 40,000 to 50,000m2 warehouses. Your problem certainly isn’t the GST on overseas goods.

And if you connect the dots, on the 12th of November, Amazon announced the launch of its new Asia Pacific Data Centre for the AWS network- based in Sydney. Beware of Greeks bearing gifts; they didn’t do this to help you fix your websites.

Australian bricks & mortar retail feels like the ending of the 2006 film Apocalypto, Mel Gibson’s opus depicting the decline of the Mayan civilisation. In the final climatic scene, warriors from two squabbling tribes spill out onto a beach after an epic chase through the jungle- to witness first contact. While Gerry and Sol argue about the GST, the Spanish have arrived- rowing ashore in giant conquistador ships.

We all know how this story ends.

I, for one, welcome our new online retail overlords.

How to Hack a Billion Dollars in Revenue in 3 Years

Distribution has become unprecedented. The rate at which consumers are adopting technology is unparalleled in history- and accelerating. And by adopting, I mean purchasing.

Consider this; it took over 45 years for 50% of US households to have a telephone, 15 years to have a cell phone, 10 years to get the Internet, 2 years to get the iPhone/iPod Touch, and 1.5 for the iPad. Today that’s 60 million units sold!

Moving to software, the adoption rates are ridiculous: Twitter went from zero to 500 million users in six years, Facebook from zero to almost a billion in eight. But Cityville, a silly game within Facebook, went from zero to 100 million users in an astounding 41 days.

Internet businesses- and face it, all businesses today are Internet businesses- are in the remarkable position of being able to achieve parabolic growth through harnessing marketing channels which I call the distribution firehoses of the Internet. Tom Friedman proclaims, “The world today is hyperconnected”; today there’s two billion people on the Internet, that’s two billion people online, networked and accessible through technology. Distribution firehoses are channels that allow you to reach these two billion potential customers more or less instantly.

The bleeding edge today is Growth Hacking and Data Science, and it’s where analytics, statistics, computer science and marketing meet. Growth hacking quintessentially looks for platforms that provide one to many relationships (read: one to millions), and develops smart ways to harness them quickly. 

Tech companies are literally hiring rocket scientists to exploit these channels and acquire customers quickly before they get crowded and the advantage fades as they get costly or restricted. My Growth team at Freelancer.com includes a PhD in Quantum Physics, a valedictorian in Mechatronics, computer scientists, statisticians and machine learning experts. The Harvard Business Review calls Data Science “The Sexiest Job of the 21st Century”.

If you want hypergrowth, conventional marketing won’t get you there, because traditional marketing methods involves absolute, linear growth and not relative, exponential growth. If I buy billboards on a freeway somewhere, the number of users that might sign up to my service or buy something from my website will just be a bump, depending how many billboards I buy, where they are, how long they show and what the creative looks like. But if I figure out a trick to get you you to refer all your friends, and your friends to tell their friends, and so on- you can see the growth is not just a rapid exponential, but it’s also relative to the size of my userbase. These techniques get even more powerful the more users you have, and grow exponentially without a corresponding exponential increase in cost base. This is why Peter Thiel said recently, ”You should have zero advertising spend. If your growth isn’t viral, it’s fake.”

Marketeers always knew that “word of mouth” was incredibly powerful, but being mathematical neanderthals living in the Paper Age they were never able to quantify it, or figure out how to amplify it.

This is why I have a VP Growth, but no VP Marketing.

Virtually all of the massive consumer Internet companies you hear about today tapped into a distribution firehose to get ridiculous growth and big so fast; there’s simply no other way it can be done so quickly.

Google is the original distribution firehose of the Internet. What do the two billion people on the Internet do all day? Type random things into Google and click on the links that get spewed back. Figure out how to rank highly for a large basket of keywords and you can get a ridiculous amount of traffic for almost negligible cost. One company hired 13,000 freelancers to churn out quality content ranging from “How to tie a bow tie” to “How to lay carpet” for a network of websites which contained Google ads next to each article. Fast forward five years and in 2011 Demand Media went public in with $325m in revenue. So egregious was this business model (In 2011, Google generated $11 billion in “Ad Network” revenue, about 30% of which came from the likes of Demand Media), that it was forced to change it’s algorithm in a series of ensuing “Panda” updates. 

Facebook is another distribution firehose; every few months Zuckerberg opens up a new feature and turns Facebook into a viral swamp for a short time allowing those that get in quick to grow at an astounding pace. Years ago it started with “You’ve been bitten by a Zombie!” apps spamming the hell out of your friends. The latest incarnation is the “Open Graph” API which allowed apps from Viddy, SocialCam and The Guardian to grow at stupendous rates. Viddy gained 17 million new users in 7 days by spamming your feed with videos that your friends are watching. Of course, to watch you also had to install the app, which in turn also spammed your friend’s feeds. You can see how this can grow exponentially if everyone has hundreds of friends. The bonanza lasted for about 7 weeks before Facebook disconnected posting to the feed. 

However, the best exploiter of the Facebook firehose has been Zynga, the publisher of sucky flash games like Cityville. The company uses the psychology of “gamification” (an engagement amplifier) to drive the one to many amplification through using Facebook as a distribution firehose for their “social games”. How does it work? In order to get anywhere in Zynga game, you need to constantly pester your friends (ironically making them them about the least “social” games I’ve ever played). Facebook, of course, is the platform of choice for distributing to friends, making it an ideal fit (Zynga probably also did a fair bit for Facebook’s growth). Zynga’s business model? Around 98% of people don’t pay a cent, but 2% for some reason decide that buying a virtual cow or barnyard in one of these games is good value. The three and a half year old company (at the time) went public last year with revenues of $1.16 billion. Yes, billion with a B. It’s certainly got me scratching my head. Go figure.

The Apple AppStore is a distribution firehose for mobile with over 30 billion apps downloaded. Three students from the University of Helsinki got in quick and managed to cement their game at the top of the Games leaderboard early, resulting in over a billion downloads. In 2011 their company, Rovio, generated 75 million euros in revenue from Angry Birds, a game which involves flicking little cartoon birds at buildings. Today the AppStore is crowded; with over 700,000 apps it’s tough for new entrants to even get a look in.

Groupon got to $1.6 billion in revenue in 2.5 years on the back of growth hacking email to get users to spam their address books. It also went public last year, on a valuation of $12 billion. Pinterest is now trying to do a similar sort of thing on the web, and gets more traffic than ESPN, CNN and AOL. Paypal was a primitive early example of using this technique, where you got $10 in your account if you referred a friend. Although it was pretty successful for Paypal, it wasn’t as refined as today’s techniques, because it got quite expensive pretty fast. However, it still allowed them to sell to eBay for $1.5 billion in four years.

Airbnb built a billion dollar company in four years on the back of smart integration (and rumoured spamming) of Craigslist. App.net looks like they’re trying to do it with Hacker News.

There’s plenty more distribution firehoses that people have yet to figure out fully; Twitter (followers & search), Reddit (readers), Amazon (buyers), Kickstarter (financiers), YouTube (watchers) and my company, Freelancer (workers). 

These firehoses are also only going to get better and better for at least a decade to come. Remember, 66% of the world’s population still isn’t on the Internet. By 2020, the number of people online will grow from 2 to 5 billion. That’s 5 billion customers. Imagine if someone actually harnessed these firehoses to do something useful.

Make money without working: automate your business

Imagine a business that makes money on autopilot, requiring no oversight, no management, no intervention - no effort on your part. A dream business.

Using a combination of software, freelancers and online platforms, we are on the verge of being able to create a new type of business to realise that dream, one that is fully automated. A business that ticks along as a black box for making money.

The holy grail is to build automated businesses that run completely in software, with no humans ‘in the loop’. Finding these opportunities isn’t easy, but more and more circumstances are arising as software becomes ubiquitous and online marketplaces grow.

If you can’t completely remove humans from the loop, you can at least substitute full-time employees with task-based work performed by online freelancers. Through a few mouse clicks we can access a global workforce of millions of these freelancers, who are skilled in just about any task you can possibly imagine. They’re inexpensive, employed on-demand and available instantly, any time of day. All you need to do is go to a marketplace like Freelancer.com, click “Post Project”, and fill in a simple title, description and budget for your project. The median time the first bid will come in for a basic website project on Freelancer is 167 seconds; meaning that you can now assemble a team blindingly fast. Even better, using an API, you can do this from software.

Probably the biggest money making bots in existence today are the algorithms that power trading activity at quantitative hedge funds. Today, many hundreds of billions of dollars are put at play by these black boxes for making money. Bots dominate the financial markets to such an extent that a recent estimate holds high frequency trading algorithms responsible for 70% of all completed transactions, and for 99.9% of all quotations on financial exchanges.

Money making bots are creeping up in other weird and wonderful places. One highly active area for bots is in online gaming. One of the earliest examples of this was a bot that someone wrote for an early MMO (Massively Multiplayer Online Game); I believe it was Everquest. All this bot did was pick up an acorn, which appeared in a particular place periodically. The bot would then walk to another location in the game to sell this acorn for a small amount of in-game currency. To do this manually was incredibly tedious, and the earnings were tiny; automated in software, this could easily be done ad infinitum. All big MMOs today have thriving real money exchanges, whether sanctioned officially or not by the games companies, where virtual currency and items can be bought and sold in exchange for cash. So in-game currency translates to real world cash; as bizarre as it seems, this bot was making money!

Today, this practice is called gold farming, and it uses a combination of bots and inexpensive labour to power an industry that generates an estimated $3 billion in revenue per annum. The sophistication of bots in online gaming doesn’t stop there, either. There are a plethora of other bots active today; one recent example in the game Diablo III exploits arbitrage across the regional in-game auction houses, much like financial bots do in the real world arbitraging dual-listed stocks.

However, it’s possible now to also automate more conventional businesses, the sorts of ones that you or I might set up.

I set up a small automated business as an experiment a few years ago. It is basically a website that sells paints, glues and other arts & craft supplies online to retail customers. It’s basically a clone of the website a wholesaler has, with an identical set of products, a different skin, and a 2x markup on prices. When an order comes in, the software automatically places the same order with the wholesaler, who dropships the goods to the customer with my company’s branding on the box for a small fee per order. The payments are taken automatically using Paypal, which integrates into the shopping cart software. I hired a freelancer out of the Philippines to manage customer service. Their job can be described by an algorithm; (1) answer all the incoming customer queries by email and live chat, (2) check the drop shipper’s website for any new products or prices and update my site (this could actually be automated in software), (3) twice a month put a few items on sale and send an email newsletter out, (4) deal with any returns or refunds (you can set up privileges for this easily in Paypal) and (5) when the balance hits a certain level, sweep it to the bank account. 

Up front, I personally put in a little bit of work to set this all up. This involved selecting a name for the site, buying a domain name, getting a logo designed (used a freelancer), buying a theme for the shopping cart (TemplateMonster.com has a theme for popular shopping cart software for just about any industry you can think of), customizing the theme (freelancer), and loading all the products in from the wholesaler (freelancer). I then hired an SEO expert to get me on the front page of Google for some high traffic keywords related to art & craft and optimise some of the product pages.

The shopping cart software is open source and free (ZenCart), the domain name costs me $7 a year, and the hosting about $10 a month. The drop shipping costs a few dollars extra per order than would normally be the case if I shipped the items myself. The freelancer is hired full time, and costs $300 a month. The site brings in an average of $300 a day in sales pretty reliably- which is about $90,000 a year. After all the costs and shipping it nets about $30,000 a year in earnings before tax. The great thing about this? I don’t do a thing. Zero. Zip. Zilch. In fact, I logged into the Paypal account the other day for the first time in a year, and that’s only because I’m writing this article and wanted to check the finances. It’s been so long since I logged into the admin panel for the shopping cart, I’ve forgotten the password. Once a month I get an invoice from the freelancer, but it’s only a couple of clicks to pay it (through Freelancer).

If I wanted to, I’m sure I could find a number of industries where this could be done. Funnily enough the economies of scale would also bring the costs down, as a freelancer could probably run a couple of sites of this magnitude comfortably, and I could share the hosting account.

Additionally, there’s a huge opportunity in selling the inventory through online marketplaces like eBay, instead of just relying upon good SEO and search traffic to find customers. There’s some great inventory management software for online auctions that automates much of the process, including inventory management, selling, labelling and packaging; as well as integrating with most accounting packages.

Not so long ago I met with one of the founders of another automated business called PlugInSEO. These guys wrote a software plugin for the Shopify shopping cart that installs for free with a click, and will scan your online shop against a list of checks to see how friendly it is for SEO. Based upon how badly you score, an option is presented to fix the problem- for a varying price. Payments are collected by Paypal, and upon checkout, the software automatically posts a project to have the problem fixed by a freelancer with whom the company has a pre-arranged pricing agreement. The app is now one of the top ten most popular apps for Shopify, and runs on autopilot. By being in the top 10 list with its “freemium” business model (try before you buy), nothing really needs to be done to market the business. Customers simply install for a free scan, see they have problems and are asked to make a purchasing decision precisely at the right point in time. This is another example of a smart use of an online marketplace to find customers, and automated use of freelancers to fulfil the work. This app doesn’t really need any maintenance, and just sits there being a cash cow for the developers that wrote it.

I can see in the not too distant future other slightly more complicated examples of businesses on autopilot. Imagine, for example. an online store that sells virtual content such as logos, website templates, sound clips or document templates. Such a store might be able to stock itself by hiring freelancers to generate content. Depending on which items are selling, the software might solicit more of this content to stock. Provided the store continues to make a profit (or has money in the bank to pay freelancers), this might continue perpetually. This operation could be further refined; as content is being produced, the software might hire other freelancers to review and check the content.

The software business could market itself in a number of ways; a search engine marketing (SEM) program could be automatically created via Google’s Adwords API for each item as it’s added, drawing funds from the bank account. The ad copy for this could be automatically generated or freelancer-generated. Alternately, freelance link builders could also be hired through the Freelancer API. As each link is built, the software checks to see that it is not spammy, of a certain page rank and thematic (basically what Google deems ‘high quality’). As each link is submitted, it’s trivial for software to run this check, and if the link costs $1, the software could release from escrow 1 cent per day for each of the next 100 days while the link is active. If the link goes down, for whatever reason, then the payments stop. Alternately the content might be sold automatically through a marketplace like eBay, Envato or Freemarket.com. 

So long as the content sells and the bot is making a profit, the software could potentially run in perpetuity. Although this example might seem like sci-fi, all the technology exists today and is quite straightforward to implement. Like all businesses, the key risk is really market risk, in determining the right content to sell. 

We could go even further with this example- perhaps the content is actually generated software itself; contrived examples might be Crossword Puzzle or Sudoku books. And that’s just getting started- if you want to see just how crazy the content that software can produce today by itself, I encourage you to look at some of the work of Philip M. Parker. This professor from INSEAD has developed software that takes a topic, and then scans the Internet for data and automatically spits out books. So far he has over 107,000 titles listed on Amazon, making him one the most prolific authors in the world. The topics of these books range from medical texts on diseases as obscure as spinal stenosis or autoimmune hepatitis, with some so obscure that maybe only a handful of people in the world have them. He also publishes a series of quotation collections and an econometric series on the future demand for certain products in various regions in the world, including the 2008 Diagram Prize winner for the Oddest Title of the Year “The 2009-2014 World Outlook for 60-milligram Containers of Fromage Frais”. I’m not making this up- watch this video if you want your mind blown.

The trick with automation is to reduce items of work to a series of steps, or basically an algorithm, and then to figure out whether this algorithm can be implemented in software, or whether you can hire a freelancer to do it for you. Not only can any job that you can describe as an algorithm be effectively outsourced (to software and/or freelancers), but once you’ve accomplished that, you can also parallelise and scale it up. Imagine suddenly an operation ten times, one hundred times, or one thousand times as big.

Increasingly this is easier and easier to do. In the future, perhaps the entire e-commerce landscape might be dominated by bots operating on cutthroat margins, that make it impossible for us puny humans to compete.

Bootstrapping

So you’ve got your killer idea, that’s got the potential to disrupt a market the size of Texas. You’ve assembled an A grade team, and have attracted grey hair with operating experience in the industry you’re going to disrupt as an advisor, or to your board. You’ve done the financials and have a strong feeling for the fundamentals of the business. You’ve worked together a plan and have an idea of where the capability gaps are and who you need to hire. You’ve also talked to a lot of customers and they like your idea, and have given feedback on how to get it to the sweet spot where they’d make a purchase decision.


You need to raise some money to get going- how do you go about it? Capital raising is as much an art form as it is a science. Firstly- and I can’t stress this enough- the best place to raise money from is from your customers, by selling something to them! If you can bootstrap your company, it’s by far the best option. You don’t dilute your shareholding, and you retain absolute control of the business. Bootstrapping introduces strong discipline around management of costs and cash flow. Cash is King! I always consider this my plan A before thinking about trying to raise funds.

An innovative way to raise money quickly from your customers is by crowdfunding using Kickstarter.com. This website, and others like it, are going to completely disrupt early stage venture capital by allowing companies to take pre-sales quickly. Recently, a startup called Pebble listed their project- customisable, internet-connected watches- after not being able to raise funds from Silicon Valley VCs. If you committed $115, they’d ship you a watch when it was ready, for $235 two watches and early access to their software development kit and so on. They were aiming to raise $100,000.. and ended up closing $10,266,846 from customers! What’s so great about this is that it’s revenue, not a dilutive equity raising. Kickstarter today funds more projects than the National Endowment for the Arts in the US (the Australian government should absolutely pass legislation to allow innovation on the crowdfunding model).

Ship early, ship often. Get customers using the product as soon as possible and providing feedback. Solve pain points and don’t be afraid to ask customers to pay. Keep your operating costs low and get your business to what Paul Graham calls “Ramen Profitability” quickly (enough to cover you eating noodles while developing the product). If you can do that,  you’ve massively de-risked the business.

Be careful here that you’re selling your core offering. I see a lot of startups trying to get going by consulting on the side. This ends up more often than not being an unhelpful distraction to the business.

If you can’t figure out how to bootstrap, the first thing to do is figure out how much money to raise. As a startup CEO you have one job and one job only- to keep the company funded. Run out of cash, and you’re dead in the water. The oldest trick in the book for a potential investor to play is to simply wait until you run out of cash. The less money in the bank, the more desperate you’ll be and the worse terms you’ll be given. Run out of cash completely and all bets are off- at best everyone’s equity position is renegotiable, including yours. At worst, if you can’t pay your debts as and when they fall due, that’s insolvent trading, and you can get into a huge amount of trouble.

For an illiquid, cashflow negative business, you need to think about raising money in terms of how much operating runway it’s going to give you. At minimum you need to raise enough money to get you a milestone that will deliver you the next increment in value for the company. Ideally, you’d like to raise enough to get you cashflow positive, but that’s not always possible. Here you want to ensure that you have enough time to hit enough goals so that the next investors that come along are prepared to pay a higher price than the old investors (an “up round”). If you fail to do this, or deliver flat results, you’ll most likely have a “down round”, and that’s if you can get anyone interested to invest at all. Down rounds are toxic to companies, they destroy morale and pretty much signal the end for most businesses. So make sure you raise enough money to get you to the next level.

Finally, bear in mind that it usually takes about 3-6 months to raise a round of financing for a private, early stage company. Sometimes it can take 12-18 months in a bad market (ironically, the best time to raise money.. because if you can pull it off, it’s unlikely you’re going to have many competitors!), so plan for that.

As a ballpark, most early stage companies aim to raise enough to operate about 18 months or so. Always raise more than less if you can, and always look to raise money when things are looking good and you don’t need it!


Dealing with the Board

Every company is required to have a board of directors. Putting together the right board for the company is critical as are all people issues- having the right founding team, and hiring right. Having an A grade idea but a B grade team behind you is a disaster waiting to happen (and unfundable). Having the A grade team, but a B grade idea however works; your star performers will surely figure out an A grade execution plan once you start building the product and talking to customers (or change the idea!).

A bad board, or even a couple of bad directors can be absolutely toxic to the company. I know, I’ve been there. A good board acts as mentor and backup to the CEO, but at the same time is prepared to ask the hard questions when needed and stand out of the way at other times.

The size of a board is inversely proportional to its effectiveness, just like all committees. Keep your board small. Three is best- typically a founder (e.g. CTO), the CEO and a Non-executive Chairman. Five is next best (so in votes, the outcome is clear). Then four, but you’ll have to give the Chairman a casting vote. Any other number is bad in my experience.

Of course, the composition of the board is actually determined by the shareholders- each director is voted on separately for their appointment and removal- but when you’re just starting out, the founders will own the majority of the shares so you can set up it up right to start.

Ideally you will want the Chairman (and non-execs) to be experienced and successful in the industry you’re entering. Importantly, all board members need to have operating experience. Startups are like a roller coaster; one day you’re on top of the world, the next you think you’re staring into the abyss chewing broken glass. If someone has run a company before, they’ll know this. Junior VCs with no operating experience will freak out the minute any bad news comes in and will start meddling in running the company. This is toxic. At best it can be a huge distraction for the team, at worst it can kill the company. A board’s role is to direct the CEO (hence “director”), the CEO’s role is to run the company with their management team. Some of the best advice I ever had when I was at my wits’ end dealing with one very unpleasant venture capitalist and other junior alternate venture capital director came from a grizzly old technology CEO, “Matt, as CEO of a startup you have one job, and one job only- to keep the company funded. The board has two jobs; to hire and fire the CEO. If they really don’t like what you’re doing, they’ll fire you. So don’t worry about it.”

When you raise venture capital, it’s usual that a venture capitalist will want a board seat. Typically angel investors won’t, because their investment will be small and directors do have legal liability over their decisions (or lack of).

Typically VCs will push in the early stage of a company’s life to have a board of five- two appointed by them, two by the founders, and an independent non-executive. This is where young entrepreneurs often lose control of the business, because the “independent” ends up being not so independent after all. He who has board control can fire the CEO and effectively change the business plan of the company. I’m a strong believer that founders should retain board control until the company is well established.

How do you remunerate the board? Founders shouldn’t get anything for sitting on a board while they are still working at the company, and neither should venture capitalists. That toxic venture capitalist insisted we pay their partner $25,000 per annum, even though he had carry in the venture fund (which owned shares). They insisted on this to invest, and our lead insisted on a co-investor to share risk (it was 2001 in the wake of the tech crash). Of course if one VC got the cash, the other wanted it too. It was fair to then pay the three non-executives. Next thing you know we had an expense of $125,000 per annum that was completely unnecessary for a company at that stage had only four people and whose technology consisted of a powerpoint presentation.

Only remunerate the non-executive directors, and this depends on time commitment and how early they join (how risky it is). An outside CEO costs about 10%, maybe 15%, but less as the company matures (millions in revenue). The Chairman is usually pro-rata this on time (say 2%). Someone who is really experienced and there at company formation who is critical to getting things going might get as much as 5%. NEDs later on might get 0.25%. Vest all over 2 years with no cliff.

The people decision are the most important, make sure you get them right!

Assembling a Team

Got an idea to disrupt a billion dollar industry? Time to assemble a crack team. First stop- the founders. Founders are the heart and soul of the company, they come up with the idea, stick together through thick and thin, have a burning desire to change the world and will crawl over broken glass to make it happen. Don’t think of considering anyone who whinges about leaving their current job, needing to be home at 5pm to look after the family, the risk involved, or their initial salary.

You are forming the A team, the Harlem Globetrotters. You and your co-founder need to be Larry Bird and Michael Jordan. You are not here to make friends, you’re here to make a difference.

Two founders are best, no more than four else the equity pie (shares) gets split too many ways. The founding team need to be handsomely rewarded in the end for giving up their social life, friends, family and health. After dilution, the total founder’s pie reduces to 40% or less at the time of exit.

Your cofounder must be extremely easy to get along with. You are proposing marriage, and if you divorce it’s potentially destruction of the company. At best, one of you will lose big time and the friendship over. I’m not keen on real husband/wife or boyfriend/girlfriend teams. It always goes wrong and it’s impossible for employees to work for them (you can’t complain about them, you can’t fire them, etc).

Start a business with friends, you’re bound to end up enemies. Your social life will tank at minimum. It’s far better off to start with business colleagues with which you already have a tried and tested working relationship, socialise occasionally but are not “best friends”.

One founder must be deeply technical with very strong product knowledge. The other must be a confident leader, speaker, marketeer (they are the chief salesman and CEO designate), but don’t pick a founder because they solely have “business skills”. I am a big fan of the founder CEO. No-one else will ever have the passion you have. I prefer all founders to be deeply technical. You might end up being a good CEO but you will never be GREAT unless you are deep on product and technology. Your customers, users and investors won’t respect you.

Picking up the business bit is easy. Just talk to lots of customers! Choose the worst ones first (just like fundraising), that you don’t care if you blow. By the time you get to the ones you really want you’ll have all the answers, a refined pitch and be super confident.

Splitting equity amongst founders- equal shares will end up getting less arguments and bad feelings down the track. It’s best to use it as a starting point then modify slightly for contribution to the company and more importantly timing of the others coming in. More risk = more reward.

Always put a vesting agreement in place for any equity or option issuance. ALWAYS. A vesting agreement allows you to allocate shares to someone, but they don’t “earn” them until you stay a certain amount of time with the company. Search google for a template. A typical “silicon valley” style vesting agreement is four years with a one year cliff. This means that you get your allocation but nothing happens for a year. On that anniversary, you get 25% of your shares. After that, each month you get 1/36th, until your shares are fully vested after four years. If a founder is going to leave, it’s most likely within 12 months. If that is the case, they get nothing, and they don’t screw up the cap table for everyone else who stays on. You don’t commit for more than 12 months? Tough.

Do this before you speak to investors to start the vesting clock now, not at the closing. Always put employment agreements / contracts in place upfront for everyone that works a day or more. Otherwise your intellectual property will be messy and you’ll get held to ransom later cleaning it up. You can put $0 now as your salary, but also another reasonable figure on hitting a sales or financing target. That way you’re not in a three way negotiation with investors over a good valuation being traded off over job title, position and pay. It’s off the table.

Always issue equity for founders at company formation, when it doesn’t cost anything. Never issue without a vesting agreement- to anyone. If you issue later when there is value in the company and you might have to pay a lot of money to buy your shares.

Employees. Always hire people smarter than you, the best you can afford. Look for PhDs (poor, hungry, driven). Reward them well. Hire slow, fire fast. I’m more comfortable making a 24 year old who is smart, deeply technical, passionate, hungry who is easy to get on as a Vice President before an experienced, veteran ‘outsider’. But start the job titles low and increase them as they prove their worth. There’s less risk of catastrophe.

Normally you would issue options to employees with a strike (exercise) price at the current fair market value of the company. This way they don’t pay tax on the grant and only benefit from upside. Only in 2009 the Australian government royally screwed option grants, creating a tax liability at each vesting point. Great stuff, guys. You really stuffed the technology industry.

I hope this advice helps!

Industries to Disrupt

Opportunity is everywhere.

Look around you. Although you might be unlikely to notice it, there’s billions of dollars of opportunity around you right now. Everywhere you see a customer in pain, there’s opportunity. Everywhere you see inefficiency, there’s potential opportunity. Everywhere you see someone with a job description that can be described by an algorithm, there’s potential opportunity.

Humans are inefficient, expensive, error prone and unreliable. Replace someone with a few lines of code, and a business can potentially become scalable, efficient- and global.

There’s a remarkable number of jobs out there that can be replaced with software or a website. All around us, industries are being disruptively transformed as code replaces humans, bandwidth replaces trucks and bits replace dead trees.

Simultaneously, these new Internet businesses are consolidating astronomical wealth into the entrepreneurs that see the immense opportunity before them and start companies to capture it.  There’s not many industries in the world where a 28 year old can make $15 billion dollars in 8 years. That’s billion with a ‘B’.

Life is generally inefficient. Facebook made it easier to stay in contact with your friends (and for marketers to target advertising). Amazon made it easier to buy things. eBay made it easier to sell things. Solve life’s inefficiencies and there’s opportunity. Make the solution disruptive- 10x better- and you can potentially make a great business out of it.

To get you going, here’s some industries I would like to see disrupted. Businesses that annoy me, that are painful, and are inefficient. Industries I would like to see remapped- disrupted by new, tech savvy upstarts.

What on earth do real estate agents do? Talk about an archaic, inefficient industry. We have all these little shops occupying the most expensive real estate in every single suburb, housing people whose sole purpose it seems is to fetch some keys and open the front door of a place to let you take a look. And there won’t just be one of them, there will be a whole strip of 5 agents in a row smack bang in the best location in town. Do they find buyers and tenants for your property? No, 80% of leads come from Domain and RealEstate.com.au nowadays. Do they really think that putting their mug on advertisements really brings in a better price? Surely, it’s possible to sell a house or rent a property over the Internet. Surely, you can do a rental inspection and organise a plumber the same way. This is an industry that’s from the dark ages, that’s highly fragmented and just waiting to be decimated. Just think of all that high end rental property that will be unlocked when this industry disrupts. (Disclaimer: I’m Chairman of a company, Leasate, that’s going to do just this).

The recruiting industry has been hacked at a bit over the years, but there’s still a huge way to go before it’s efficient. How on earth can a recruiter justify charging 15% of the first year’s salary to rummage through an old database of CVs, make a few phone calls and line up a meeting? Really? LinkedIn and SEEK are a start, but there’s a multi-billion dollar opportunity or two still to come disrupting this industry. SEEK needs a bit of a competition. They manage to charge $250-500 to post a couple of paragraphs on their website! The marginal cost for SEEK to provide this service is zero. That’s a big markup.

Don’t get me started about the legal industry. I’ve paid out more for corporate lawyers than the GDP of some small kingdoms in Africa. In the tech industry, lawyers manage to charge $40,000 or $50,000 just to draft the documents for a venture financing. Why? We know they just go to the main file server and pluck out a template. The space is crying out for a TemplateMonster.com to come around and replace $10,000 of “$525 per hour legal services” with a $99 smart online template.

Accountants. How many of you love doing your tax every year? For me, it involves rummaging through my old mail (printed on dead trees) and dumping a shoe box of statements and receipts (printed on dead trees) at the accountants. For three figures an hour, they perform data entry, sorting and filtering algorithm on the dead trees. There’s got to be a better way, let alone save all those trees that get chopped down to be turned into Woolworth’s receipts.

I love FindMyiPhone. Now can someone please make FindMyKeys, FindMyWallet, FindMyCat? Why do we need to carry a wallet? Why do I have to carry rectangular bits of plastic around with me to function in society? Keys? I have to carry hunks of metal with squiggly lines carved into them to get into places? Really?

Want a couple of hundred more ideas? Go to Craiglists or eBay- pick off a vertical and do it better. My company, Freelancer, took freelance jobs. OKCupid took dating. Airbnb took vacation rentals. Etsy took craft. There’s hundreds more to go around.

Opportunity is everywhere- go and grab it.

Disruption

As I write this, Industries across the world are undergoing spectacular transformation as the software and the Internet disrupt incumbents at a pace never ever seen before. Yesterday’s titans like Walmart, Best Buy, Kodak, Blockbuster, Borders, Sony, David Jones and Harvey Norman are waking up overnight to find their business gutted. The weapon of mass destruction? Pencil necked geeks armed with websites and mobile phone apps, who use software to disintermediate and automate, leaving yesterday’s goliaths to asphyxiate from their multi-billion dollar supply chains and infrastructure. We are well and truly in a new age- the Internet age.

Disruption is occurring everywhere: retail, the labour markets, telecommunications, marketing. The bigger they are, the harder they fall.

Forward looking, the transition from one paradigm to the next may not be so obvious, especially when burdened with inertia from legacy businesses, however in hindsight it’s usually quite obvious.

Imagine I told you I had a great idea for a business and wanted you to invest. My idea: we’re going to find really great websites, ones with niche high-quality information about a specific topic. Then what we’re going to do is print all these websites onto dead trees. That’s right, we’re going to go out to the forest, chop down lots of trees, and then print websites on them. We’ll then find hundreds of big trucks, and load them up with hundreds of thousands of these dead trees, and ship them to every single suburb on Earth, where we are going to set up little shops that sell the dead trees. Of course, these websites keep updating, so we’re going to have to ship out a new set again on a monthly, weekly, sometimes daily basis.

If I told you that was my idea, I’m pretty sure you would say I was nuts. But that’s exactly what we do with this magazine (BRW) you’re reading.

In the example of online, disruption is on many fronts - it’s over 10x cheaper to ship versus print; it’s 10x faster to break news. It’s 10x more convenient- you can access the entire library with only a web browser. It’s much easier to search.

It’s often thought that incumbents don’t see disruptive technologies coming. That’s absolutely not the case- most of the time they are well aware of the new paradigm. Kodak invented the digital camera, yet went Chapter 11 after failing to capitalise on the idea before the eventual shift to digital photography. Incumbents are usually well aware and actively involved in developing the new paradigm.

Why do they then fail?

Disruption cuts both ways. Continuing our example, it’s now 10x easier for advertisers to measure ROI from buying an ad- you can simply track a purchase through someone clicking through a link- and it’s 10x cheaper for new competitors to enter the market, who can do so cheaper, unburdened with legacy costs and infrastructure.

This is something that the music industry knows too well. When the music distribution changed from CD to the Internet, the marginal cost of production dropped to zero. Copying bits is free.
The music industry was well aware of this; however, the Sword of Damocles was hanging over their head. Do they embrace the Internet and jump in 100% to the new model- rendering their existing manufacturing and distribution network worthless- spurring a change from CDs to mp3s, potentially encouraging piracy- potentially fatally as consumer behaviour changes? Or do they pan the new model, arguing that mp3s are somehow inferior, sue new digital entrants and any customer who downloads a song?

Big companies are weighed down by inertia. Their desire to change is stifled by the fact that the new model generates 10x less revenue. So they undertake a token effort with a new business unit as they try to figure out a business model. What practically happens next is that in every resource allocation meeting, the old business units are still bringing home the bacon, so marketing budgets and headcount continue to flow the old models, not the new. Quarter on quarter, the old businesses slowly bring in less and less- and while the new businesses bring in more and more revenue, they usually don’t come anywhere near replacing the lost revenue from the old model. They fail to realign.

And then Apple comes in and eats your lunch.

A complete outsider — someone with nothing to lose, who doesn’t care about the smaller opportunity in the new paradigm, one who can consider entering the industry even as a loss leader to sell something else (iPods, iPhones, iPads). They figure out that consumers really enjoy being able to instantly play any song that’s ever been recorded, on demand, for 99 cents.. without having to catch a bus 5km to a store to buy their music etched onto squashed circles of plastic. They’re OK making a billion dollars, turning the incumbent ten-billion dollar businesses into shipwrecks.

Software and the Internet is in the process of disrupting every industry I can think of. Apple’s market cap is now bigger than Greece, Portugal and Spain combined. Amazon.com will soon be larger than Walmart. Kogan is kicking Gerry Harvey around the block from a nondescript office whose sign is sticky taped to the front door. 

Sometimes you’ve just got to make the leap before you get eaten.

Business Models

For every product or service there’s also a zillion ways to sell it into the market. This is what’s called a business model.

Let’s say I develop a new disruptive type of antivirus technology that detects viruses 10x better than competitors.

I could sell it shrink wrapped in a box through stores for a one-off price of $99 (traditional retail). I could sell it online as a subscription for $9 a month where the customer has to keep paying for the latest virus updates (subscription based Internet services). I could give the product away for free to home users to build my brand recognition up, and instead focus on selling to enterprise customers (enterprise software). I could give it away to everyone and put it on the big free software download sites for mass distribution, allowing it to detect viruses but charge customers $49 if they want the viruses removed (this is called a “freemium” model- give a limited version away for free, and charge for a premium upgrade). I could pre-package it on a slim form factor computer and sell it as an antivirus firewall to businesses (hardware). Alternately I could just supply the technology to the big antivirus companies for millions of dollars a year for them to incorporate in their own projects (OEM licensing). I could go on for hours!

Each of these business models has its pros and cons, and today antivirus technology is sold in all these ways and more.  

Figuring out the most lucrative way to commercialise such technology was exactly the problem I faced at my last company, Sensory Networks.  We had made breakthroughs in high performance (gigabit) pattern matching technology, which was implemented in an integrated circuit. We chose to sell OEM to big vendors who designed the chip into network equipment such as antivirus firewalls.

We decided to sell OEM because the sales process was easy for us (it was a technical sale to engineers), we expected to make millions per sale (we were going into products that ship in the tens of thousands) and our sales force could be small (the market was concentrated into a few customers).

Of all the business models, I think we ended up picking the worst!

We discovered that to sell technology OEM there’s an extremely long sales cycle. The products we were selling into (firewalls) had a new version every year. Since we were a physical component that had to be built in at the factory, the sales cycle could be up to two years long (you might not get into this year’s product and have to wait a year).

Next we found out that even though our customers sold hardware, they didn’t have any hardware engineers to design us in! The firewalls looked flash, but they were just standard PCs in a slim line case with some pre-installed software and a nice paint job. To sell to them we needed to put the chip on a card (like a graphics card), and provide a software library to drive it. To then sell that card we needed to integrate it into their software or bundle apps! When we finally landed our first sale, McAfee told us they wanted another company to manufacture it. Next thing you know we were in the test equipment business, because McAfee’s factory needed to test the cards.

Even after we were designed in and working, we faced another dilemma: the rest of McAfee’s other software wasn’t ready. Do I send my engineers to help them so I can start selling chips, or wait? Now I’m in the consulting business!

Somehow a business that I thought would be highly scalable (design a chip, sell millions) ends up being completely unscalable. Selling OEM is not a cookie-cutter sales process. It’s a long, complex, custom job each sale with many factors outside your control.

On top of this, even though the firewalls sold for $20,000 retail, I could only sell my cards for $500, making $250 a piece. That’s because after going though the value chain of wholesalers, distributors and so forth, every dollar of cost into hardware adds about $8 to the retail price. Vendors didn’t want to take the hit to their margins, and couldn’t increase prices (an inelastic market). It was a total squeeze play. I was a component inside a component inside a component.

In hindsight, I should have just packaged the technology in a firewall appliance and sold direct to enterprise myself. That’s a simple, repeatable sales process.

In the meantime I saw others in the same market absolutely kill it! Simon Clausen, one of Australia’s great tech entrepreneurs, built up PCTools as a freemium antivirus software business. Spyware Doctor was free to download, would detect viruses on your computer, and ask if you’d like them removed. If so, you had to put in your credit card and pay $49.95.. per year. He put it on downloads.com, where it became the number one download. Since the first thing millions do when they get a virus on their computer is type “free antivirus” into Google, this was brilliant. Talk about a pain killer and awesome business model. You have the customer held hostage with a gun to his head right when they are feeling the most pain! No surprise Simon managed to bootstrap that company with a $1000 initial investment to over $40 million in revenue.

So before you sell think about how you’re going to do it. You might sink or swim depending on the business model.