Admittedly, charts like this are interesting but can be interpreted in many different ways. I’ve offered a few comments below but I’m curious to hear what you think…

All categories are 2% of total unless otherwise labelled.

Deadpool 2009 Highlights

• Social networks were still “hot” in 2009, coming in at #1 again but taking a smaller chunk of the pie. I also categorized them as vertical social networks because they were all targeted at a specific group of people. All of the failures in this category are remnants of the social networking startup explosion of ‘07.
• In similar fashion to the previous analysis, media streaming took second place (in 2008 I combined video and audio into one category called Media Streaming). In 2009, Music streaming alone had enough failures to warrant a dedicated category…looks like a tough business to be in, filled with lots of competition and licensing nightmares.
• A lot of money was spent on Search over the last few years; another difficult space to be in and what seems to be a lack of consumer interest in vertical search engines.
• MVNOs disappeared from the list…not too surprising.

Notables

• 7 out of the 45 (~16%) deadpool members were divisions of Yahoo! (acquired companies) that were shut down (eg. Maven Networks, GeoCities, FareChase, etc.
• 2 members were founded 10+ years ago (GeoCities and FileFront) both belonging to the File/Web hosting category.

Other Stats

• Average number of years from launch to deadpool (alive period): 2.8 years for 2009 deadpool members (1.8 years for Jan 2007 to Sep 2008 period)
• Longest ‘alive’ period: 10.3 years (6.4 years)
• Shortest ‘alive’ period: 6 months (2 months)
• Standard deviation: 2.1 years (1.25 years)

Kudos to the PostRank team. It isn’t easy to crunch over 2 billion pieces of data, make it look good and provide a great resource for a variety of audiences. Whether you want to learn more about engagement metrics, are looking for the best blogs for a particular topic, want to know who key influencers are or want see how your blog stacks up, PostRank Top Blogs of 2009 has something for you. I encourage you to check it out.

So how did Tech Capital do?

Not too shabby

1. how else will newspapers be able to pay the journalists who collect, synthesize and write the news that more people than ever are reading?
2. it’s time for consumers to realize that the free all-you-can-read buffet is going to disappear because it makes no economic sense.
3. the economic model in which newspapers give away all of their online content doesn’t work…It is becoming obvious newspapers must start charging for their online content

Let’s breakdown the “news ecosystem” to the simple parts.

• Create Content
• Distribute/Publish Content
• Consume Content

Now let’s think about this system in the context of past and present and specifically look at competitive advantage, barriers to entry:

• Create Content
  • People: If you think about journalists and talent then things haven’t changed much from past to present. One of the greatest resources of any company that is inimitable are its people. Sure, things like how Newspapers source stories or carry out their work have changed but at the end of the day you need talented people to write good content.
  • Trust: Newspapers still maintain a competitive advantage in terms of content quality because many people trust traditional media (I use the word trust loosely) over new media. There are signs that this is changing but as it stands today the barrier to overcome this is pretty high.
• Publish/Distribute Content
  • Past: This was a major competitive advantage and barrier to entry. First you needed the facilities and equipment to pump out a physical paper each day and then you needed a way to distribute that paper out to a large number of locations.
  • Present: People still read physical copies of the newspaper but when people talk about the problems with newspapers today they’re talking about the transition to the online world. This is pretty obvious, but the barrier to entry for publishing content online is basically zero, and distribution is equally as easy and therefore no longer a competitive advantage (for static content).
• Consume Content
  • Past: Subscription fee or newsstand purchase. Limited window to access.
  • Present: Still have traditional subscription and newsstand but many people read content online. Easy 24/7 access.

The problems currently experienced by the Newspaper industry are not unique. We’re talking about a business that has enjoyed a profitable “physical” existence for quite some time and is now having a difficult time maintaining revenues/margins as more and more of their customers consume news online. There is no reason why Newspapers can’t charge for content, and there is no special business model that is going to make them the profits they once had. Consumers have choice, and unfortunately most of the stories newspapers distribute are not original. Go to Google News right now and look at the top stories, there will be hundreds, if not thousands of sites that offer you the same news story.

#1 – It’s all about content, trust…and less money
Things have changed and if you believe my “news ecosystem” breakdown then the only real place for newspapers to differentiate themselves is content (I know, this is pretty obvious) and trust. The problem is that only a fraction of most newspaper content is unique (read Clay Shirky’s post on created vs acquired content) which greatly diminishes the value they provide. The competitive advantage does lie with the content creator and as Mark suggests in #1 these journalists need to be paid. But will people pay for content? Sure they will, but only a small portion of them. The rest will go somewhere else for their content needs. Unfortunately the reality for Newspapers is that as daily paper subscriptions decrease their online subs won’t increase at the same rate because there is an abundance of choice. Yes, people will pay for content but just like anything else in the real world they will only pay if they feel it provides enough value at that price. If a paper reports the same news as the free guy, why pay for it? Papers need to use content and trust to their advantage in order to get people to pay. Competition is tough, isn’t it?

#2 – But I like the free all-you-can-read buffet
I don’t think consumers need to “realize” anything. This isn’t how the world works. People don’t make conscious decisions about whether they should give money to corporations to ensure their survival. If free content exists and it is the same as paid content then people will take the free content. Market forces such as competition, supply/demand, etc. will determine what the price is, who survives and who dies. Set a price and let the market decide. This is what scares the hell out of most papers…and of course government regulation can change all of that.

#3 – No Money Mo Problems
As of late there has been a lot of talk about charging for content. I’m waiting for someone to just pull the trigger and get it over with. If you have the best content then you should be able to charge for it. If the revenues are enough to sustain a profitable business then great, if not then welcome to the world of competitive business.

The harsh reality is that in many cases Newspapers will not be able to maintain the revenues/profits that they once knew. How can they? Differentiation through distribution is difficult in the online world and as everybody knows, once the barriers to entry drop and competition increases, competitive differentiation is going to have to come from somewhere else to fight off the downward pressure on prices, margins, etc. You can choose to embrace and adapt to market changes or you can fight them. In many cases fighting them is a losing battle since healthy competition and minimal government intervention always leads to victory for the market.

Assuming the government doesn’t get involved what we’re going to see is a lot of papers close their doors, consolidation, fat-cutting. I don’t foresee people paying for daily news content but there is obviously a business around premium content. New businesses will emerge that are structured to survive in the “new world”, older businesses will need to change to do the same.

So in the end we have the same old story:

• Don’t fight the market…you will lose
• Don’t fight consumer behaviour…you will lose
• Differentiate, provide value and people will pay…whether they pay enough to make you happy is a different story

• The number of people viewing video online is rapidly increasing
• The amount of online video content that can be streamed is increasing (Hulu, Youtube HD, Netflix, Zip, etc.)
• Video quality has increased alongside increases in bandwidth
• There are a host of set top boxes (“internet enabled devices”) able to deliver online video to your television set

The promise of being able to watch any TV show, movie, etc. whenever you want, by streaming video over the internet is here. It sounds great but ISPs have implemented what they call the “Monthly Usage Allowance.” At first glance, these allowances seem perfectly reasonable because they are designed for most people who only check emails, browse the web and download the occasional video. However, if you decide to subscribe to a video streaming service like Netflix (in Canada Zip.ca will be offering a similar service) then the story could change.

How does the “Monthly Usage Allowance” affect someone who wants to start streaming most of their video consumption over the internet? Well, first consider that according to Nielsen media, the average American watches 153 hours of television a month. For the purposes of this post I’ll use Bell and Rogers as example ISPs since they own the majority of the market where I live. Assume that as a Rogers or Bell internet subscriber you’ve decided to watch all of your TV shows, movies, etc. by streaming online video. Based on the video quality (i.e. bitrate) you choose and the internet package you’re subscribed to, the following tables show how many hours of video you could watch until you reach the monthly usage allowance. (By the way, a higher bitrate means higher quality video. i.e. 3800kbps is better than 500kbps):

# of online video hours (per month) – Various Video bitrates

# of online video hours (per day) – Various Video bitrates

What these tables both illustrate is that unless you’re watching low quality video you’re going to reach the monthly data usage allowance before you can reach the 153 hours the average American watches each month. For example, if you have the Bell Performance package and only stream SD Video @ 500kbps, in a month you’ll be able to watch 111.1 hours of video (or 72.6% of the monthly American average) or approximately 3.6 hours of video a day. That’s not too bad but we’re talking about low quality video that is a bit better than Youtube. If you bump up the quality to 1000kbps then you’re getting half the hours/month. Even worse, let’s say you’re paying for premium HD content, with the Bell Performance package you’ll be able to watch about 15 hours a month or roughly half an hour a day (9.6% of the monthly average). Remember, I’m talking about just using your internet connection to watch video only, the number of hours would actually be lower since you have to expect some usage for other online activities.

Some context on video quality:

• Normal Youtube video is around 250kbps; high quality is around 900kbps
• SDTV (normal TV) is anywhere from 1500kbps to 5000kbps (or higher).
• Netflix encodes SD video at 500, 1000, 1600, 2200, and 3400kbps depending on the speed of your connection. HD video is encoded at 2600kbps and 3800kbps.

One other point of view I thought I’d share with you is the length of time in hours you could use your internet connection at full speed potential before reaching the usage allowance.  

The Point
I think there is something important to recognize here. Many companies, startups included, make an assumption that data usage for the traditional consumer (at home) isn’t something to worry about. A video streaming service like Netflix or Zip.ca is based on the premise that bandwidth is available and cheap for the end user but the effect of the amount of data used and how this usage will affect the end customer is often not considered. Sure, your company isn’t restricting the data, but the experience you’re providing could indirectly cost the consumer more money. Is your business relying heavily on an upstream partner? What is your relationship with this partner? Maybe ISPs will increase the caps over time but if they don’t it will be interesting to see how things play out as more data intensive services inevitably become available and the average consumer begins taking advantage of these services.

So what’s the point? This list is an indication of ad dollars spent on magazine advertising –- i.e. less advertising = fewer pages. No surprise that page growth is negative across all categories, but the worst performers are (and again not surprising) either luxury related or more general in nature. Yes this is a simple analysis but considering the large number of startups focused on advertising revenue, maybe they want to think about Parenting/Kids and Health/Lifestyle/Fitness as an opportunity worth targeting.

The story from the word clouds above reads as follows:

• December: Canadian economy, auto industry bailout
• January: Auto (Toyota), banks, jobs, recession
• February: Banks, bankruptcy, jobs, losses, markets, sales/profits, lower/fall

You don’t need a word cloud series to be aware of that story but it highlights the ability to pull out this type of information from 3 months worth of new articles and with only about 5 minutes of work. I thought I’d throw in another example, this time for TechCrunch:

Tweaking
The story for TechCrunch over the last 3 months is rather static, mostly covering the same stuff: facebook, google, iphone, launches, raises, etc. I did the same for ReadWriteWeb and it was almost an identical result, an indication that those topics tend to dominate their articles. Depending on what you’re after it may help to ‘tweak’ the cloud and remove words that are either too dominating or irrelevant. After tweaking I ended up with the revised word clouds:

and the resulting story:

• December: Social (network), search, mobile, video and a lot of Betas
• January: Video, social (network), search, music
• February: Mobile, social (network), kindle, search, video

Sometimes the more interesting finds are the words that are picking up steam over time, in this case: Cloud, Platform, Streaming.

Obviously there are a number of different ways to use word clouds (competitor messaging analysis, news analysis, trending, topic popularity, resume analysis, etc). If you’re interested in creating your own, the two apps I use are Wordle and TagCrowd. Wordle allows you to publicly save word clouds so if you decide to try it out please feel free to share any interesting clouds in the comments section.

First I should declare a few things:

1. Deadpool members are failed companies that have been reported by and entered into the deadpool by TechCrunch. Does this mean the list of companies is biased in some way? It could be but we don’t know for sure.
2. I did a high level analysis to determine the reason these companies failed using publicly available information. Obviously there are many uncertainties in this approach, most notably:
   • many things can contribute to the failure of a company, reducing it to one or two reasons is a difficult task and obviously I wasn’t able to capture the entire picture but this is an approximation
   • what is reported publicly and what really happened aren’t always the same

The Process: I already had the list of deadpool members, from there I spent some more time collecting info to determine:

• the reasons for failure
• the location of the company
• if the company was funded by an outside investor

The result of my analysis:

Reasons
What probably isn’t surprising is that there are a number of common reasons for failure among startups.  What is surprising is that most reported reasons are secondary problems. Sure, a company might have had to close shop because competition was fierce but it could be that they weren’t differentiating themselves effectively, or maybe they blame poor user adoption. The reality is they were trying to solve a problem nobody cared about. When I dug deeper I found that many of the companies suffered from the same usual problems we all hear about on a regular basis.  Some of the key takeaways are:

• Business Model / Funding
  • Lack of funds and cash flow problems came up as a reason for failure in the majority of companies. Again, in many cases this is a secondary problem… If you’ve been at it for a few years and can’t generate enough cash to sustain yourself, at some point the primary problem might not be a lack of funding but the lack of a business model.
  • You need a way to generate money. We’ve all heard this a million times and I’ve written 2 other posts that deal specifically with identifying a suitable business model: Selecting a Business Model, Part 1 and Selecting a Business Model, Part 2
• Competition / Differentiation
  • Granted, it can be a tough market out there but if you can’t carve out some market share for yourself because of competition then maybe there is something else going on. Without a doubt, certain markets are extremely competitive… If that is the case then you should try to answer many of the necessary questions before deciding to enter the market.
  • What differentiates your product from the competition? Why are people going to buy from you or use your service over the competition? Does competing head on with an incumbent that has unlimited resources (compared to you) make sense?
• “False Problem” / User Adoption
  • “User adoption was not what we anticipated.” There are many reasons why this may be but I think in many of these cases you’ll find that the company started off with a “False Problem.”
  • The “False Problem” is when a company tries to solve a problem nobody really cared about in the first place, or creates a new service with a marginal improvement or slight variation over an existing one.
• Shifting Strategy / Implementation
  • Yes a company does have to adapt to market changes and customer needs but when you don’t have a sound strategy you just end up confusing all stakeholders and slowing implementation. Don’t try to reinvent your company every 6 months.
• Partnership Issues / Regulation
  • This is the Single Point of Failure (SPOF) problem. You rely on a single partner to provide data or services and if they disappear or turn off the tap then your business fails. Another extreme example is businesses based on legal loopholes or arbitrage… If these holes get plugged you’re out of business immediately.
  • It’s a good idea to identify if your company has any SPOFs. If you do have a SPOF then you’ll want to figure out a contingency plan in case the unthinkable happens but more importantly, how you can remedy the problem so you will never have to deal with it.

Geography
Maybe this has to do with the TechCrunch bias I mentioned earlier but most of the deadpool members originated from California and Washington state, 72% of the members in fact. Seattle specifically seems to be a hotbed of Deadpool members. Interestingly, no Canadian companies made the list (even though there have been failures) as reported by TC.

The Process: I first identified each TechCrunch deadpool member from 2007 to the present and collected info on:

• The type of Service/Product
• The launch/founded date
• The reason for failure (if available)

I didn’t include certain companies depending on criteria such as acquisitions (Yahoo for example) unless there was a major failure, or companies that had been entered into the deadpool but were revived at a later date. The end result was a list of 43 deadpool companies. (Obviously more than 43 companies failed during this period but remember I’m just looking at the deadpool with a filter).

The second step was to categorize the companies so that I could come up with the following breakdown of the deadpool for the last 21 months or so.

All categories are 2% of total unless otherwise labeled.

There are a numerous ways to interpret this chart. A few examples are:

1. Failures: The chart shows that certain areas such as Social Networks, MVNOs (Mobile Virtual Network Operators), Media Streaming, etc. tend to have higher failure rates than others. In other words, one might think that these are all bad bets and you should stay away.
2. Variety: The chart more or less confirms that nobody is immune from failure. There is an assortment of companies from very different areas that make it into the deadpool.
3. Hot Areas: At a basic level, the chart helps to identify areas that had a high level of startup activity preceding this period. If you reason that the number of failed companies is proportionate to the number of new entrants then the chart shows that a few years ago Social Networks, MVNOs and Media Streaming were hot areas (and may continue to be today).

Unfortunately, the conclusions aren’t that black and white. #1 has some truth to it but not in each case, #2 is pretty obvious, and #3 is probably close to the truth. I think all three of these points are valid on some level but the issue is, without proper context, it is difficult to say with confidence if any of these categories are particularity worse than the other. The MVNO market has certainly been very risky and has seen its fair share of failures with obscene amounts of Venture Capital money being spent but how does that compare to Social Networks or Group SMS? To provide the proper context I would need to take a single category from the pie and then perform a similar analysis on that single category to understand what percentage of companies in that that specific category fail (for example Social Networks), repeat that for each category, and then compare and combine with anecdotal evidence. That is a lot of work, and maybe I’ll get to it some day but for now I’ll skip that and move onto the data I actually have.

Once I had the data collected it was easy to come up with the following:

Average number of months from launch to deadpool: 21 months
Standard Deviation: 15 months

Longest ‘alive’ period: 76 months (6.3 years)
Shortest ‘alive’ period: 2 months

So it looks like if you’re past the 3 year mark the odds of hitting the deadpool drop off dramatically. Nothing surprising here except for the quick 2 month member.

I will continue the analysis and see if commonalities can be found between these companies that may shine some light into what precipitated the journey to the deadpool. I’ll save that for the next post.

I’ve made the list of the 43 companies available for download if you want to check it out. It is in Excel format: Deadpool member (xls)

In the meantime I’d like to hear your thoughts on what the chart means to you and any related stories you may have…

Keep in mind that the models are not necessarily mutually exclusive and you can apply more than one model to a particular business. Also, some of the models may seem very similar but there are usually minor differences that change things considerably from a business process perspective.

I am sure the list below is not an exhaustive one but it includes models that I’ve come across and categorized over the last six months or so. Various sources have contributed to this list, some of which I’ve listed at the end of this post.

Remember, Part 1 is identifying the Basic Business Model Archetype of your business. In Part 2, take a look at the Business Models that are commonly used for your company’s archetype.

I’ve separated the business models into 2 categories:

• Models that intrinsically generate revenue
  • For these models it is clear how you generate revenue. For example, it is obvious that a manufacturer can make money by creating and selling an asset and as such this is the suggested model for a manufacturer.
• Models that do not intrinsically generate revenue
  • For these models, revenue generation is not as straightforward. Instead a variety of “monetization” techniques must be employed in order to generate revenue from the traffic or value they create.

Models that intrinsically generate revenue

• Creator
  • Manufacturer
    • Manufacturer
      • Revenue: Create and sell a physical asset
      • Comment: One of the oldest and most basic models. Design and manufacture something and then sell it at a profit.
  • Inventor
    • Internet Inventor
      • Revenue: Create and sell an intangible asset
      • Example: Creating patents, trademarks, brands, etc. and selling them is an obvious one but also consider the creation of a web properties for the purpose of selling them.
    • Virtual Manufacturer
      • Revenue: Create and sell a virtual asset
      • Example: Facebook Gift Shop allows friends to buy and send virtual gifts at a cost of $1 and up.
      • Comment: The difference between a Virtual Manufacturer and an Internet Inventor is that an Inventor does not sell the same intangible asset more than once, while a Virtual Manufacturer does.
• Distributor
  • Wholesaler/Retailer
    • Online Retailer
      • Revenue: Buy and sell physical assets using an online storefront
      • Example: Well.ca an online health and beauty retailer
    • Catalogue Retailer
      • Revenue: Buy and sell physical assets through the use of a physical and online catalog.
      • Example: SkyMall is a catalog retailer and also uses an online storefront where consumers can purchase items.
    • Clicks & Mortar Retailer
      • Revenue: Buy and sell physical assets through retail locations in combination with an online store.
      • Example: Chapters sells books and other items at its physical retail locations and online.
  • Intellectual Property (IP) Trader
    • IP Trader
      • Revenue: Buy and sell intangible assets
      • Example: Buying and selling a web domain, a web property, or any type of IP. Demand Media, among other things is a company that has a profitable domain portfolio which includes general domain names to typo domains.
      • Comment: This model is very similar to the description of a broker but the difference is that the vendor does at one time or another have “possession” of the intangible asset.
    • Bit Vendor
      • Revenue: Facilitates the transfer of intangible assets
      • Example: Steam provides a service to download and manage video game purchases. Apple iTunes facilitates the purchasing of media such as music, tv shows etc.
      • Comment: The distinction between the Bit Vendor and an IP Trader is that for a Trader, once the asset has been sold it cannot be sold again. A Bit Vendor on the other hand can sell the same intangible asset to satisfy an unlimited amount of demand since it can simply make a copy for each buyer.
• Landlord
  • Financial Landlord
    • Lender
      • Revenue: Lenders provide cash that their customers can use for a limited time in return for a fee
    • Insurer
      • Revenue: Customers pay a premium to insure against losses.
  • Intellectual Landlord (Publisher)
    • Online Publication
      • Revenue: Create unique content, publish it to the Internet and charge users for a copy of it.
      • Example: Some online newspapers continue to charge for access to articles.
      • Comment: This can be a challenging model since there is so much free content available online, making it difficult to consistently provide unique content that users will pay for. Many publications that fall into this category are usually related to independent research findings.
    • Online Archive
      • Revenue: Archive content that is no longer current but still valuable, make it easily searchable and retrievable, and charge for access
      • Example: The NYTimes has archives for the years 1851 – 1980 that are available to be purchased in single or 10-article packages
      • Comment: Online Archive Storage could also be offered as a service to Online Publishers who do not want to invest in the infrastructure needed in order to store and deliver content from over the years
    • Knowledge Merchant
      • Revenue: Leverage existing systems and infrastructure to capture data and sell it in the form of intelligible analytics
      • Example: Before being acquired by Google, FeedBurner charged a premium for providing detailed analytics to publishers.
      • Comment: A knowledge merchant leverages an existing system/technology that serves an unrelated purpose and uses that system to collect valuable information.
    • Software Publisher
      • Revenue: License software that you have published for limited use.
      • Example: Microsoft
      • Comment: The most common business model for software.
    • Audience Measurement
      • Revenue: A market research agency measuring online audiences and consumed content.
      • Example: HitWise provides detailed statistics on Internet user behaviour
      • Comment: Unlike a Knowledge Merchant, an Audience Measurement firm usually relies on 3rd party technology or external data sources to collect valuable information.
    • Piggyback
      • Revenue: Cross-sell premium services to an existing subscriber base
      • Example: Apple exploits the opportunity to cross-sell Apple iPhoto users premium services such as Photo Books & Calendars.
      • Comment: This is a classic approach used by retail outlets that applies well for niche sites and certain subscription based services. The broader the content or service the harder it is to make the connection for a cross-sell opportunity.
    • Subscription
      • Revenue: Sell various levels of service with sliding fees, making it simple to upgrade (or downgrade) a user account as desired
      • Example: The Dogster Plus service is a subscription model that gives users more services and ad-free pages.
    • Trialware
      • Revenue: Trial the application or service for a period of time and then require payment to continue using.
      • Example: Newsgator TopStyle
      • Comment: There is also a similar approach where the software is limited in functionality and a license key must be purchased to gain access to all functions (referred to as Crippleware).
    • Content Services
      • Revenue: Deliver text, audio, or video content and charge users a subscription fee to access to the service
      • Example: Netflix
      • Comment: In many cases this involves streaming services usually with limited use of the asset.
    • Freemium
      • Revenue: Provide some services for free and charge a fee to access premium services
      • Example: Flickr provides a free account that anybody can sign up for that gives you reasonable services. Users have a choice to pay a premium and upgrade to a Pro account which offers additional functionality, storage, and bandwidth usage.
      • Comment: This is a model that is commonly employed and can work quite well if the premium services provide enough value to justify the subscription fee.
    • Trust Services
      • Revenue: Membership associations that have created a public code of conduct that they have committed to follow and uphold. Members pay a fee to be certified and included
      • Example: BBB Online has a reliability program and for a fee will provide member sites with a “Reliability Seal” that identifies sites that have met BBB standards for the web
      • Comment: This model can be very challenging because (a) you have to build up enough goodwill that organizations and consumers place trust in your evaluation and (b) you need to convince organizations that your services are needed.
    • Metered Usage
      • Revenue: Measure a subscribers usage in terms of time, data, etc. and charge them for what they use
      • Example: Amazon Simple Storage Service (S3) can be used to store and retrieve any amount of data and you only pay for what you use.
      • Comment: In this model there is no recurring revenue since users are only charged for what they use. If people are not using your services you will not make any money.
    • Metered Subscriptions
      • Revenue: Users pay a subscription fee and are given access to a limited number of actions, pageviews, etc
      • Example: Slashdot allows users to subscribe to a 1000 page subscription for $5. In return you can view 1000 articles without ads (there are other small benefits).
      • Comment: This model is similar to metered usage but in this case you receive recurring revenue since the user pays a regular subscription fee.
    • White Label
      • Revenue: Take existing systems or software, augment the service and/or provide some sort of added value built on top, apply your brand and sell it.
      • Comment: Benefits of producing a white label app can be: quicker development time since you’re simply building on top of an existing platform, the possibility of leveraging the knowledge and assistance of the original publisher, a uniquely branded product with seamless integration (theoretically).
  • Intellectual Landlord (Attractor)
    • Freemium
      • Revenue: Provide some services for free and charge a fee to access premium services.
      • Comment: See above
  • Contractor
    • Professional Services & Education
      • Revenue: Take an existing product/subject and provide services around that product such as support, training, templates, etc.
      • Example: Red Hat’s linux OS is open source and freely available software so Red Hat makes money by tweaking the kernel, providing software support, training, and other services.
    • Contractor
      • Revenue: Allow people to contract you for service at an agreed price
      • Example: A website or Graphics design firm such as Netfirms.
      • Comment: Used by many as a way to augment their revenue stream but can easily become a distraction.
• Broker
  • Financial Broker
    • Transaction Broker
      • Revenue: Facilitates the transaction between a buyer and seller and charges a fee or takes a percentage commission of the transaction.
      • Example: PayPal provides the means for users to exchange money with each other in a secure manner.
      • Comment: A very simple and traditional method of making money online. Some of the added value for buyers and sellers is the ability to use credit services, to provide some layer of protection from being defrauded, keeping history of sales/purchases, etc.
    • Financial Broker
      • Revenue: Matches buyers and sellers of financial assets such as cash, stocks, bonds, etc and facilitates the transaction.
      • Example: Microcredit services such as Prosper allow groups of people to lend money to people looking to borrow money.
  • Physical Broker
    • Buy/Sell Fulfillment
      • Revenue: Similar to an affiliate/referral model but customer orders are taken by one vendor and are fulfilled by another.
      • Example: CarsDirect allows users to browse car inventory, make a purchase through CarsDirect.com by talking to an agent who will then find the car the buyer is looking for. In the end however, the car will be bought from a dealership being represented by CarsDirect.com who will then take a commission.
      • Comment: Takes it one step further than a Metamediary by participating in the transaction but one step short of an Online Agent since only the order is taken. The real value for a user of a service like this is the time and money saved by not having to shop around, and the benefit of having someone knowledgeable to discuss the purchase with.
    • Online Agent
      • Revenue: Provides service across an entire transaction such as the initial price negotiation, fulfillment, support, etc
      • Example: Expedia negotiates prices with hotels, airlines, etc and then offers these services to customers. They also will fulfill the booking and provide support where needed.
      • Comment: In this case the Online Agent helps match the buyer and seller but takes the transaction one step further by negotiating the price and fulfilling the transaction. Unlike Buy/Sell Fulfillment, the buyer never deals with the seller when using an Online Agent.
    • Online Agent (Priceline)
      • Revenue: The “Name Your Own Price” model matches buyers and sellers, but unlike regular agents who take a flat or percentage fee, in this model the fee is the difference between the sellers asking price and buyers bidding price.
      • Example: Priceline.com created this model.
      • Comment: The acceptable price for a particular asset, during a particular time is given by the sellers but is unknown to the buyers. Buyers are able to make a single bid for the asset or use of and if the bid is within a certain threshold of the sellers asking price the transaction is fulfilled in the same manner as an Online Agent.
    • Auction Broker
      • Revenue: Matches buyers and sellers of physical assets
      • Example: eBay.com
      • Comment: A simple model made famous online by eBay. Allow sellers to post items for sale, manage the auction and charge a flat fee for certain features and a commission once the item is sold.
    • Virtual Storefront
      • Revenue: A service to host online storefronts. The service provider can charge fees such as listing, hosting and/or transaction or commission fees
      • Example: Amazon offers seller accounts where users can list their own items for sale. It gives buyers the option of purchasing goods from another supplier. Amazon charges listing fees and takes a commission once the item is sold. eBay provides a storefront service for merchants the sell products in large volumes.
      • Comment: There aren’t many pure play virtual storefront providers that have the ability to compete with the likes of eBay and Amazon. What is most important with virtual stores (as with real stores) is the amount of traffic you receive.
    • Classifieds
      • Revenue: Provide a place where sellers can list items and buyers can browse listings. Generate revenue by charging a listing or membership fee.
      • Example: Craigslist is the largest classified site but does not charge listing fees except for job postings in specific geographic areas.
  • Intellectual Property Broker
    • Advertising Networks
      • Revenue: Build up a large pool of member sites, attract advertisers and feed ads to member sites.
      • Example: AdBrite has over 32,000 member sites delivering about 760 million impressions a day. Advertisers can choose which types of ads they want to run on which types of sites and then bid on how much they’re willing to pay for clicks.
      • Comment: The larger the member pool along with the quality and # of page views the more that can be charged to advertisers to display their ads.
    • Niche Advertising Network
      • Revenue: The same model as the Advertising Network but dedicated to a niche market. The aim is to provide a quality group of member sites for potential advertisers.
      • Example: DogTime Media is an ad network dedicated to pet-related sites
      • Comment: The benefit of the niche network is the possibility of having a concentrated network of quality member sites that have commonality between them. The network may provide for a better opportunity to target ads.
    • Search Agent
      • Revenue: An automated software agent (usually some proprietary algorithm) that searches on behalf of a ‘interested party’ for goods and/or services to find the price, availability, ratings, etc. If the buyer search is converted into a sale, the agent receives a commission.
      • Example: ePinions.com, Kayak.com all provide product information along with locations on where you can purchase the product.
      • Comment: This model closely resembles a Metamediary but is different in that it is more focused on the job of aggregating information about products and providing a single view for the end user to help them make a decisions. Since its focus is on retail products, it aims to make money by connecting the purchaser with a retailer and taking a commission.
  • Human Resources Broker
    • Human Resources
      • Revenue: Matches buyers and sellers of human services
      • Example: Elance brings together professionals looking to get a job done with those looking for work on a job-to-job basis.
      • Comment: Outsourcing
    • Classifieds
      • Revenue: Match employers with potential employees
      • Example: Monster, Workopolis

Models that do not intrinsically generate revenue

• Creator
  • Inventor
    • Brand Integrated Content
      • Description: This isn’t as much about direct revenue generation as it is about PR and generating buzz
      • Example: BMW Films – A highly successful film series “starring” different BMW models receiving over 100 million film views and numerous awards.
      • Comment: If successful, the buzz that is generated can be used to drive traffic to other web properties like the Buy/Sell Attention model or can be Monetized
• Landlord
  • Intellectual Landlord (Attractor)
    • Portal
      • Description: A web site that usually combines a variety of content and services and in most cases has a search engine.
      • Example: Yahoo!
      • Comment: In the case of Yahoo! they offer many portals such as Travel, Finance, Autos, etc. which all roll-up into their main portal page.
    • Free Registration
      • Description: A site that provides free content but collects usage and demographic data through required registration.
      • Example: NYTimes.com requires users to register in order to get access to their free content.
      • Comment: This information can be used to serve highly targeted ads to the each individual user when they visit the site and log in.
    • Buy/Sell Attention
      • Description: Not a revenue generator but a way to jump start or increase traffic to a related web property by using one property that is performing well to promote the other.
      • Example: Dogster.com launched in January 2004 and was able to use its popularity to help promote a new site it launched in August 2004 called Catster.com.
      • Comment: There are two ways this model tends to manifest itself. Case 1: A site is launched and becomes successful, the authors launch another site to promote the new one and drive traffic to it. Case 2: A site is launched with the sole intention of creating buzz in order to launch another site. The second site is the real target and the first site is only a means to drive traffic.
    • Open/Free Content
      • Description: A group of volunteers work collaboratively to support and create peer-reviewed publications. The content they produce is free and openly available to the world.
      • Example: Wikipedia
      • Comment: In many cases it can be difficult to monetize these sites for any purpose other than site maintenance. There is considerable debate once a site reaches the size of wikipedia on whether it can be monetized without somehow compensating contributors for their time. Wikipedia’s approach has been to collect private donations and hold fund raisers.
    • Social Networks
      • Description: A place where people can come together and share experiences under a common connection such as friendship, business, pets, etc. Some sites are broad in nature while others choose to be niche sites.
      • Example: Facebook is based on Friend connections, LinkedIn is based on Business connections, Dogster is based on Pet connections.
      • Comment: Most social networking sites monetize through contextual advertising or charge for premium services. For a broadly based social site like Facebook or Myspace there is a greater opportunity to grow the user base but it is a challenge to contextually target such a large and diverse group. On the other hand, niche sites like Dogster may have a smaller user base but advertising can be focused on the common interest all users have (in this case Dogs).
    • Loyalty Program
      • Description: Create a loyalty program by providing redeemable points or discounts to users who purchase merchandise from selected retailers.
      • Example: At Airmilesshops.ca you can collect airmiles by shopping at over 90 retailers.
      • Comment: The loyalty program provides users with an incentive to visit the loyalty site first before making purchases to ensure they receive extra benefits such as points and discounts. In return the loyalty program can display highly targeted ads based on the demographic data they collect on the member.
    • Metamediary
      • Description: Provide detailed information/reviews and possibly other services for products but do not actually sell anything.
      • Example: CNET Reviews provides comprehensive information on tech products including performance, value, pricing, etc. and publishes this information on their site. PowerReviews provides the “engine” for retailers to include review capabilities in their online stores.
      • Comment: The examples highlight two different approaches. In one case the products are chosen, reviewed, and the results are published on the respective site. In the second case, technology to facilitate user generated reviews is provided and implemented on a retailers online site.
• Broker
  • Intellectual Property Broker
    • Labour Exchange
      • Description: A service which indirectly provides human intensive labour which would otherwise require a company to pay wages in order to accomplish the same task. This model can be effective only if the cost to run the service is less than the combination of:
        1. The value of the information/data collected and/or the task being carried out and
        2. The savings in wages that are realized by implementing this model.
      • Example: 1-800-GOOG-411 Google provides free directory assistance via the 1-800-GOOG-411 service but in return they are able to train their speech-to-text engine and improve their voice recognition capabilities. Gwap Users play fun games for free but in reality what they are doing is teaching computers and hopefully improve such things as image search, audio search, object recognition, and information for artificial intelligence.

Methods to Monetize

For models that do not intrinsically generate revenue some of the common Monetization Methods employed are:

• Affiliates
  • Pay Per Click (PPC) and Pay Per Action (PPA)
    • Description: Get paid for each click through to an affiliate site or for users that perform an action on an affiliate site.
    • Example:
      • PPC: Shopping.com gets paid for each click through to a merchants site.
      • PPA: PartyPoker will pay on a scale per each user you refer that signs up.
    • Comment: In most cases affiliate programs will pay a set rate for a number of completed actions and/or click throughs and unless you drive a significant amount of traffic to their site this will be non-negotiable.
  • Revenue Share
    • Description: Receive a commission for each user click through and conversion that originates from your site. The commission is usually a percentage of the sale price.
    • Example: The Amazon.com Associate program offers up to 10% in referral fees on all qualifying revenue made through affiliate links links. There are many ways to be creative with this model. For example, Overlay.TV provides an online video platform which allows users to overlay “clickable” information directly onto video, which then ties into a large affiliate network.
• Advertising
  • Search Paid Placement (Keywords/Categories)
    • Description: On a high traffic site offer to sell link space based on keywords and/or categories.
    • Example: Google Adwords will place paid links along with search results based on keywords.
  • Contextual Advertising
    • Description: Advertising related to and placed alongside page content.
    • Example: Google Adsense ads are related to the content on the page. Websiter operators get paid per thousand impressions (CPM) and per click (CPC).
    • Comment: This is the most popular choice to monetize web properties and is often the only source of income. If a site has a large amount of traffic, the simplicity of generating revenue through the placement of contextual ads is what makes this a popular choice.
      • It is worth commenting specifically on the traffic problem. If you are using Contextual Advertising as the only means of generating revenue, then you need to have a very large amount of traffic to drive the number impressions. If this is not the case then hopefully you have a high click through rate (CTR) to compensate. Obviously one would prefer to have a huge amount of traffic and a high CTR but this is not an easy task, and many companies fail to generate any significant revenue this way.
  • Behavioral Targeting
    • Description: Takes contextual advertising one step further by identifying the person viewing the page (usually in a non-personal manner), their preferences, behavior, etc and presenting Ads targeted directly at the unique characteristics of the individual.
    • Example: Blue Lithium, Tacoda
  • Intromercials
    • Description: Users are directed to a landing page where they must watch a Video/Animated Ad before proceeding to the desired content. In most cases users are able to skip the ad and go directly to the content.
    • Example: Forbes employs this method
  • Traditional Ads
    • Description: Traditional Ads such as banner ads that are not contextually targeted, but just served “blindly” from an ad network.
    • Example: Banner ads such as those found on sites like BusinessWeek, eWeek, etc.
  • Watch an Ad (Ultramercials)
    • Description: Require a user to watch a video Ad before providing access to premium content.
    • Example: Salon DayPass gives access to all of Salon.com premium content from 1-8 hours after viewing and clicking through an Ad.
  • Sponsors
    • Description: Set aside “visual real estate” on a site and then sell a portion of that space.
    • Example: TechCrunch charges per month for a 125×125 pixel area
    • Comment: This model works well if (a) you have a site with a high number of unique visitors and a large number of page views and (b) the site is dedicated or limited to a few topics and has a fairly consistent user profile.
• Data Collection
  • Intelligence Vendor
    • Description: Collect demographic information on a user base, create actionable intelligence about users and sell this information.
    • Comment: This type of information is valuable to market research firms, analysts, advertisers, academics and in most cases the information is unique and cannot be copied by competitors.
• Donations
  • Donations, Fundraiser, Grants
    • Description: Provide a valued service (often based on crowdsourcing) that appeals to a large audience and then accept voluntary donations in return for the value created.
    • Example: Wikipedia is funded through donations, fund raisers, grants, etc. and managed to generate over $1.4m in revenue for 2006
    • Comment: This is a model most often employed with the intention of simply maintaining a site. Even a highly trafficked site like Wikipedia (which receives over 40 million unique visitors a month) is only able to generate an ARPU of about $0.003 per month

What Did I Miss? If you know of any other models that aren’t listed above, please share them in the comments section. Can you identify which archetype you think the model belongs to?

Sources
There are a lot of great articles, posts on the web about business models. One site in particular, Business Models on the Web by Michael Rappa was a great resource. Quite a few of the models listed above are from Michael Rappa’s site. Other Notables: The Long Tail Of Business Models and Alternative Online Business Models.

One final note….yes, I do realize that this may be the longest 2 part blog post ever!

Background
Early on in my research I compiled a list of the business models employed by a variety of companies operating online in some way or another. It was apparent that certain models worked well for some companies and others not so well. Shortly after I started to compile this list I discovered a paper written by researchers belonging to the SeeIT Project at the MIT Sloan School of Management. The paper describes a Business Model Classification framework and the differences in financial performance of various business models. The classification framework assisted me in creating — in combination with the list of business models I’d been aggregating — a systematic approach to identify the most appropriate models for a particular business.

Purpose
For some of you, what follows may seem like an academic exercise that doesn’t add any real world value because what you really want to know is “what model is going to make me money?”. If that is the case then maybe you’ll be more interested in Part 2, but do me a favor and keep reading the rest of this paragraph before you decide if you want to continue. When a business is struggling to “find” the best business model, in many cases management tends to review the various models that are out there and then try to apply them to their business to see what fits. It’s akin to people building a product and then trying to figure out the problem it solves after the fact. In many cases this is the wrong approach because you’re starting at the end point. You shouldn’t first identify a business model and then see if it fits to your business. Instead, try to identify exactly what the business does, and then apply that knowledge to the various models that are out there. This may seem like a trivial exercise when your business is, for example, wholesale/retail but what you may not realize is that you’ve probably already done this exercise intuitively. The choice of business model for a wholesaler/retailer is in most cases an obvious one so you don’t consciously think about it. In one quick thought you jump right to the end of the process and determine the model should be a pure play online retailer, or a catalogue retailer, etc. For some businesses however, the path from “My company does X” to “The business models that work for my business are Y” is not that easy or intuitive. The purpose of this post is to assist anyone out there who is having this problem and is struggling to figure out which models are applicable. It will hopefully allow a company to say “This is what we do, this is where we fit, and the models that can work for us are…”. Let’s get started…

Defining Business Model Archetypes

Step 1: What rights are being sold?

The first thing we need to do is identify the type of rights you are selling. The options are:

• Ownership of an asset
• Use of an asset
• Matching of a buyer and seller

Step 2: What is the transformation of the asset?

Secondly we need to determine how much the business transforms the asset. In most cases this only applies if you are selling the ownership of an asset. Options are:

• Significant
  • Purchase an asset, make considerable changes/additions and then sell
• Limited
  • Purchase an asset, sell with no or minor changes/additions

The steps above can be summarized in the following selection table:

Based on the table above you can determine if a business’ basic archetype is Creator, Distributor, Landlord, Broker.

Definition of Basic Archetypes

• Creator: A Creator designs the product, procures the materials/resources needed and transforms these elements into a finished product. Even if the entire manufacturing process is outsourced, one is still a Creator so long as the majority of the design was their doing.
  • Example: Dell Computer designs the computer, buys the required components and assembles the computer according to the customers customizations.
• Distributor: A Distributor buys a product and in most cases resells the identical product to someone else. They can sell through the wholesale or retail channel. In some cases the identical product isn’t sold but it may be packaged differently or value-added services may be offered in conjunction with the product.
  • Example: Future Shop buys merchandise and resells through their retail stores. On certain products they provide extra services such as extended warranty or bundling for additional value.
• Landlord: A Landlord sells the right to use an asset but ownership of the asset remains with the landlord. The right to use the asset can be indefinite or for a specified period of time. The asset does not need to be a physical asset but can also include intangible assets or any other asset that has value and can be “lent out” in exchange for something.
  • Example: Microsoft – People who purchase Windows for their computer are actually purchasing the right to use the software (a license in this case). Ownership remains with Microsoft.
• Broker: A Broker in its simplest form assists in matching potential buyers and sellers. A broker at no point takes ownership of the asset, instead they connect the interested parties and make money through fees charged to either or both sides of the transaction.
  • Example: eBay allows sellers to post assets and buyers to bid on those assets. eBay never takes possession of the asset but receives a commission for the sale and charges a listing fee.

Step 3: What type of asset is it?

For the third step we dig a bit further and take the framework to the next level. We now need to identify the Type of Asset that is being sold. The options are:

• Financial assets
  • Includes cash and other financial instruments like like stocks, bonds, insurance policies, etc.
• Physical assets
  • Includes durable items (computers, cars) as well as non-durable items (food, clothes).
• Intangible assets
  • Includes legally protected IP (patents, copyrights, trademarks, and trade secrets), as well as other intangible assets like the organizations knowledge, goodwill, and brand image.
• Human assets
  • Includes people’s time and effort. People are not “assets” in an accounting sense but their time (and knowledge) can be “rented out” for a fee.

Examples

• Amazon Books (Distributor/Physical Asset):
  1. Amazon is selling the ownership of an asset
  2. Amazon does not transform the asset, therefore asset transformation is Limited
  3. The asset is a Physical Asset
• Google Search (Landlord/Intangible Asset):
  1. Google is “selling” the use of an asset. Their search engine service.
  2. Step 2 is N/A because Google is selling the use of an asset
  3. The asset is an Intangible Asset
• eBay (Broker/Physical Asset):
  1. eBay matches buyers and sellers of physical assets.
  2. Step 2 is N/A
  3. The asset is a Physical Asset
     • not everything on eBay is a physical asset but for the purposes of this example we’ll assume this is the case.

Basic Business Model Archetypes

The following table is the combination of determining the type of asset being sold and identifying which rights are being sold. The result is sixteen detailed business models. I left out models that don’t apply to the Internet (Entrepreneur, Physical Landlord, Financial Trader) and others that are illegal activities (Human Creator, Human Distributor).

Definitions:

• Manufacturer
  • Creates and sells physical assets
  • Example: Dell Computers designs/assembles computers and sells them online.
• Inventor
  • Creates and then sells intangible assets
  • Example: This model is rarely used but an example would be developing patents and selling them outright. The creation of a web properties with the sole intention of selling them could fit into this category.
• Wholesaler/Retailer
  • Buys and sells physical assets
  • Example: Well.ca, an online retailer of health & beauty products.
• Intellectual Property Trader
  • Buys and sells intangible assets
  • Example: The practice of buying/selling web domains. The purchase of business.com by RH Donnely is a good example of this.
• Financial Landlord
  • Lets others use cash (or other financial assets) with certain conditions. There are 2 types of Financial Landlords:
    • Lenders provide cash to clients for a limited time and receive a fee in return usually in the form of interest.
    • Insurers provide clients with the option to access financial reserves if they experience losses and receive a fee in return usually called premium.
• Intellectual Landlord
  • Licenses or otherwise gets paid for limited use of intangible assets. There are 2 types of Intellectual Landlords:
    • Publisher provides limited use of information assets
      • Example: Licensing software (Microsoft).
    • Attractor attracts people’s attention using content and then “sells” that attention to advertisers
      • Example: Many web sites “attract” users with original content and then place ads on their web site to generate revenue.
• Contractor
  • Sells a service provided primarily by people, such as consulting, education, etc.
  • Example: UPS
• Financial Broker
  • Matches buyers and sellers of financial assets
  • Example: Microlending – Prosper matches and consolidates people who are willing to lend money with people who need a small loan and charges a fee to facilitate this.
• Physical Broker
  • Matches buyers and sellers of physical assets
  • Example: eBay
• Intellectual Property Broker
  • Matches buyers and sellers of intangible assets.
  • Example: Ad networks such as AdBrite. Publishers are looking to sell space on their web site and advertisers are looking to buy advertising space. Ad Networks facilitate this transaction and take a share of the money.
• Human Resources Broker
  • Matches buyers and sellers of human services
  • Example: Workopolis and Monster.

Examples

Continuing the example from above with the new information we have on the Basic Business Model Archetypes gives us the following:

• Amazon Books is a Wholesaler/Retailer because
  1. Amazon is Distributor
  2. The asset is a Physical Asset
• Google Search is an Intellectual Landlord because
  1. Google is a Landlord
  2. The asset is an Intangible Asset
• eBay is a Physical Broker because because
  1. eBay is a Broker
  2. The asset is a Physical Asset

Onward

Part 1 only gives us half of the picture. In Part 2 we will explore the various models that can be employed for each Archetype and discuss the models that intrinsically generate revenue and the options available for the monetization of models that do not intrinsically generate revenue.

I realize that without Part 2 I’m leaving you hanging but in the meantime I would love to hear your thoughts on this approach, some of your own examples, differing opinions, disagreements etc.

If you’re interested in reading the complete paper on which this post is based it can be found on the SeeIT Project website.

Update: Part 2 of this piece has been added, to continue reading see: Selecting a Business Model, Part 2