Moving home

I decided to move this bog to a new home hosted on Typepad. Please click on the following link to go my home:

http://1vc.typepad.com/

If you were subscribed to the RSS feed of my blog via the Feedburner link on the right of this page, then "no worries", I updated Feedburner to the new feed. If you were subscribed to the direct feed from Blogger, then please update your subscription so you pick up my new blog location.

The next application platform - the Browser????

We tend to think of the browser as a way of displaying web pages; a piece of software that can render HTML and graphics in a consistent and universal way. The browser has become the ubiquitous display device much as the IBM 3270 display terminal was back in the era of the mainframe.

But the browser is also programmable - you can extend it with plugins that run as native code on the PC, you can download Java applets that run within the Java virtual machine and then there is JavaScript which can be embedded in web pages and is interpreted as the page is rendered by the browser.

If we think about JavaScript at all, most of us probably think about animated menus, mouse over effects etc. Then along came AJAX - combining JavaScript with lightweight XML interactions with the remote web server behind the scenes; all designed to make the web page interaction more interactive. AJAX is still relatively new and only recently have frameworks and other tools been made available that reduces the learning curve for the average programmer to write complex code. As a measure of just how far things have come, Zimbra has implemented an Outlook like interface using AJAX.

This was impressive enough but this last week I came across two applications that were implemented entirely in JavaScript and run on the client with no host interaction. The first was TiddlyWiki - a lightweight Wiki implementation that doesn't require a host - it runs entirely in the browser. TiddlyWiki has a growing community of developers implementing plugins (macros) to extend TiddlyWiki itself. The second is a complete Chess game that runs in JavaScript; this is a demonstration of the capabilities of the tools produced by an Australian company called Morfik.

Last but not least, there are now a number of contenders for WebOS - an operating system environment implemented within the browser to further facilitate application development.

So where is all this going? My hunch is that the combination of these tools will become the platform of choice for implementing SOA (Service Oriented Architecture) applications and displace the heavyweight frameworks of .NET and other contenders.

Big can be ugly!

What happens when you start offering a service such as on-line data storage to hundreds of millions of users? What system design and scaling problems will Google, Yahoo and MSN find as they start offering 2 GB of Email and say 5 GB of file storage to their users?

Back in the good old days users didn't keep every mail message they received and the size of their disk made everyone a good housekeeper, tidying up unwanted files and being very picky about what they kept. Those days are gone - user behavior has changed as the underlying economics have become more favorable.

2 GB of Email doesn't go very far if you never delete the messages you receive - even if you do, the information you keep is growing on a daily basis. 5 GB of file storage isn't what it used to be either, given digital everything - pictures, audio and shortly lots of video. Unlike the phone system, its going to be hard to make reasonable assumptions about the utilization rates of these new services - because of changing user behavior, the utilization rates are going to be higher - much higher!

The phone system is built around the assumption that not everyone wants to use it at the same time - it's designed around an assumption of the maximum number of concurrent users that will be active in the peak hour of the day. If that number is exceeded (as it did after the Loma Prieta earthquake near the Bay Area in 1989) you don't get to make a call - you pick up the phone but there's no dial tone and you are placed in a queue until capacity becomes available.

With services such as on-line storage or email that you never throw away, the assumptions you make about the rate at which data is added by a user or the typical utilization rates get to be critical - you won't be popular if you start throwing the user's data away when you start getting overloaded!

Look at the math for a moment...

• 100 Million users
• 2 GB email, 5 GB disk space
• 7 GB per user
• 700 Million Gigabytes

700,000,000 gigabytes! Ouch! Even with 500 GB disks, that's still 1.4 Million disk drives. But wait!, it has to be backed up... I don't even want to think about using tape or optical tape/disk/etc. for this amount of storage - with that many disks spinning, even disks with one million hours between failures leave me looking at a couple of disk failures an hour! Ugh!

RAID 5 would be a solution - but needs another drive to hold the parity for every three drives of capacity - so now we have 1.75 million disk drives. That's just the DISK DRIVES - think of how much power, rack space, servers and cooling you need to drive them... Last but not least, think of the bandwidth you would need to provide even a reasonable subset of the users access to their data.

As you can see from the math, bringing all this to a single location or even a small number of locations leaves you with a scaling problem beyond belief. There are some problems you simply have to solve by moving data to the edge of the network an distribute the scaling problem.

One rumored solution to this was reported here - a very interesting idea and a nifty way of scaling the solution.

This week the press (NYT, WSJ etc) were busy reporting the capital equipment spends of Google ($1.5B projected in 2006) and MSN - with the prospect of bringing these kind of services online, no wonder!

In the meantime, I think I'm going to buy a big USB FLASH drive!

Anything you can think of... 2.0

The list is growing... Web 2.0, Radio 2.0, Voice 2.0. Enterprise 2.0...

There are lots of lists of Web 2.0 companies, I'll start a list of the areas/topics to which the 2.0 suffix has been added!

Here's a start with some sample companies...

• Web 2.0
  
• Plenty of lists without starting a new one! Check out eHub for a running update.
  
• Radio 2.0
• Last.FM
• Mercora
  
• Pandora
• Voice 2.0
• Iotum
• Enterprise 2.0
• Socialtext
  
• Storage 2.0
• Amazon S3
  
• Box
• Google G-Drive (rumored, not yet released)
• Microsoft Live Drive (not yet released)

You have to look deeper than the 2.0 suffix to see what's in common across these different areas; they all leverage similar trends -

• Web enabled platform
• Simple, well defined interfaces
• Leverage other services, applications etc. as extensions
• Apply relevance, tagging, social networks, your life (interests, tastes, calendar, email...)
  

Put another way, the world is becoming one big mash up! This is very cool especially when you add mobility into the mix. What's happening is that folks are taking all the infrastructure (wide spread, inexpensive communications (the Internet), Open Source, the Web, cellular, VoIP etc.) that has been built out and applying it to new areas of opportunity.

If you have thoughts about new areas to which you can apply the 2.0 suffix, let me know - I'll add them to the list!

Crunch point

Definition: Crunch point (krŭnch point), a transition milestone in the evolution of a company where:

1. What used to work, doesn't anymore.
2. A marked transition needs to occur in the focus of the CEO and/or team.
3. New skills are required to take the company to the next phase but are missing in the team.

Examples:

• Moving from development to revenue generation
• Ramping sales through key milestones (Zero revenue, $1M, $10M...)
• Fund raising
• Expansion to address new customers, markets, geographies etc.
  

From the examples, it's easy to see that every company will go through multiple crunch points and that how you deal with them will determine whether the company evolves to the top of the food chain or fails the Darwin test.

Blindingly obvious? Perhaps, but if so, why do so many companies stumble painfully through their crunch points and others go through them like a hot knife through butter?

In one word, PREPARATION.

Preparation is about identifying the milestones that remove risk, establish proof points and demonstrate necessary skills are in place. Too often the focus is on the next crunch point without thinking about the big picture. Like the saying about draining the swamp, it's too easy to get consumed with the current crisis and lose sight of the bigger picture.

Successful CEOs think about the big picture all the time while making sure their team is on top of what must be done to get through the next crunch point. This requires the CEO to be self-aware of their own strengths and knowing when to hire people who bring in necessary skills, effect change or step aside.

Companies big and small have their crunch points; it's the CEO's job to anticipate them and make sure that the company is prepared.

How far can "ad-supported" go?

Thinking about the growth in on-line advertising and how it's being used to support "free" services got me wondering how far "ad-supported" business models could be taken. I started playing with the idea of a "free", ad-supported cell phone some time ago. Ads could be delivered to the phone in many different ways, for example:

• 15 second audio ads during the call set-up time
  
• SMS ads based on the location of the phone
• Streamed video ads based on the demographics and interests of the phone user
• Location specific coupons or Spiffs accessed by the user

If you start thinking about advertising in different ways (a topic that Robert wrote about here), you could imagine getting an ad delivered to your phone as you pass an electronics store - "$200 off this high end LCD TV - only $1300 - click here to purchase then go to the pickup counter". Having access to a mobile device with a user of known demographics could generate new forms of advertising.

The idea would be to create a MVNO (Mobile Virtual Network Operator) like Virgin Mobile for example - the company buys wholesale minutes of airtime from established cellular providers and so doesn't have to build out their own infrastructure. Phones are given away to users who agree to get the ads in exchange for free airtime.

Of course there are a lot of issues that would have to be addressed such as:

• Capital required to build a big enough user base to take the model cash flow positive
• User acceptance of getting ads on a cell phone
• Frequency of ad delivery that users will tolerate
• Rates paid by advertisers for targeted (location, demographics etc) ads
• Acquisition of advertisers
  

The biggest question is whether the economics of this model would work out so I started doing some research into the underlying factors - price per minute of delivering the service, typical usage patterns (how many minutes per month) and the cost of giving away the handset.

Price/minute: I looked at the price per minute that underlies the different plans offered by the cellular network operators here in the Bay Area. The effective price ranges from about 9 cents/min for the low end plans and drops to 3 cents/min for the high end plan. At these rates the cellular operators make money and don't forget these are end-user prices - the underlying operating costs are lower. The high end plan at $199.99 for 6000 minutes (not including night/weekend time!) probably gets closest to the underlying cost - that's the 3 cent rate.

Usage: How many minutes of use per month would you have to cover for a typical user? Virgin Mobile is a public MVNO and their first half financials for FY2006 show that the average user consumes 752 minutes of use per month. The U.S. cellular networks promote a 900 minute plan as covering most user's needs so that resonates with the 750 number from Virgin's financial report.

Handset cost: This is a harder one to track down but you can find phones sold without a contract (and hence likely with little or no equipment subsidy) for less than $30. You'd want to be giving away a reasonably attractive phone so say $50 is the cost of the give away.

This coarse brush run at the economics (venture math!) works out as:

• 750 minutes of use at 3 cents/minute = $22.50 per month
• $50 for the phone with a useful life of 12 months = $4 per month

So to break even, you have to be able to cover $26.50 of expense per month. Depending on the forms of advertising and the conversion rates, its not hard to believe that you could generate a profit once you had a large enough user base!

When I started thinking about this model I had thought that the companies most likely to implement this model would be Google or Yahoo. These companies have the brand, millions of users and the capital strength needed to get the model profitable.

In digging into the likely economics underlying this model, I came across a new company called Xero Mobile which was described in some detail in this article. I'm surprised that this didn't get more press (I missed this when it was announced 2 weeks ago) - Xero is setting out to build an ad-supported MVNO targeted at college students.

Admittedly this is a high level view but it looks feasible to have an ad-supported cell phone service. Is this the limit of how far we can take ad-supported business models - let me know what you think!

Blocking online ads

One of the best reasons to use Firefox as a browser is the AdBlock extension. AdBlock together with the AdBlock filter set works seamlessly with Firefox to block advertisements from being displayed when you browse. AdBlock's default configuration leaves the space that would have been filled by the ad so that the page renders correctly.

A side effect of AdBlock is that web pages load much faster! It turns out that much of the delay in loading a web page is waiting for the ad servers to deliver ads or their tracking beacons (little one pixel transparent images that help the ad network track what you are doing).

Imagine the impact on the Internet advertising market if every browser came equipped with AdBlock like capabilities? Google is complaining about the upcoming IE7 release from Microsoft because the toolbar search box defaults to sending search queries to MSN. Think how much louder Google would yell if Microsoft shipped IE7 with ad blocking technology and the default filter set permitted ads only from Microsoft approved ad sources!

The impact of AdBlock is limited; currently it only works with Firefox and its not shipped in the standard distribution. Even if AdBlock functionality gets implemented for IE7, the majority of users stick with the defaults that are configured at installation time.

Too bad! It would be interesting to see the counter measures that the ad networks would come up with to defeat the AdBlockers (Newton's 3rd law applied to the Internet - to every security measure, an equal and opposite counter measure). In the meantime, if you use Firefox, you can download AdBlock and its associated helper (AdBlock Filterset.G) by clicking on the links.

Tags: Google Advertising

Net neutrality - the howls continue!

When I initially wrote about Net Neutrality a month ago, I thought the whole thing would blow over. The concept that a service provider would block access to particular content or internet destinations seemed unlikely. I was wrong on both counts! Technorati shows the Net Neutrality issue is gaining strength...



and this post I found shows that at least one rural service provider blocked access to Vonage.

At the NVCA Annual meeting this week, Net Neutrality was described as one of the largest regulatory threats to the growth of early stage companies along with a number of other serious issues including SEC regulations requiring expensing of stock options and Sarbannes-Oxley.

The issue arose when several of the larger carriers suggested that they were going to offer premium services that were targeted at rich media such as streaming video. The idea of premium grade internet service isn't new. When I ran the software development of the IOS back in my Cisco days, we conceived of tiered services on the 'net - a base "best efforts" service and protected Quality of Service (QoS) that protected certain kinds of traffic so that they always had preferred bandwidth.

As content providers start to offer streaming HD movies over the Internet something will have to change! - you don't want to be hauling high bandwidth traffic to millions of different end-points over the same core network (much too expensive!). More content will have to be moved to the edge of the network and the last mile access network will have to accommodate serious bandwidth improvements. Looking at the financial performance of Akamai (NASDAQ: AKAM) over the last year, content IS moving to the edge and the build outs by Verizon et al are all edge focused.

I've decided to add my own howl to the issue; at a minimum, service providers should be required to support a basic level of Internet access that provides equal access to all destinations and content. Much as you can dial any number you want on your telephone, you should be free to select any service or content via the Internet.

If the service providers want to offer a premium grade service that gives high bandwidth, guaranteed QoS for a specific service, more power to them. If there is sufficient demand and the benefits warrant the price paid, the market will give them a payback.

In the meantime, I'm still predicting that the new providers of content (like Google, Yahoo etc.) will get more and more into the access network business - either on their own or with partners, so that they can protect the user experience.

Tags: NVCA Net Neutrality

Sales strategy - shotgun or rifle

The transition from the development phase of a startup to selling a product is perhaps the most nerve racking and dangerous in the life of a company. Last month I wrote about the risks of parallel mistakes when beginning the process of ramping sales so this month, I thought I'd add a few thoughts about choosing the initial customer prospects.

In the lead up to getting the product into beta, its appropriate to cast a wide net for prospective customers. The shotgun is an analogy here - its useful for hitting a moving target where you get close enough to score a hit without having to have precise aim. Match up the capabilities of your product with prospects who should be a fit but cast a wide net - go after different size prospects (revenue, number of products, employees etc.) in a wide range of different industries.

Look at the prospects where you get early traction; prospects who agree to be beta customers are an indicator - every commercial prospect has more than enough things to do without taking on the work of beta testing your product. Make sure you do the work to qualify a beta prospect - a good way to do this is to work with the prospect to agree a detailed beta test plan; this should set the prospects expectations, define what the beta will test and assign roles and responsibilities for both the prospect and the company. Put together a plan that includes who allocates the resources, what will be tested and defines the completion criteria.

Don't be bashful in this process; work with the prospect to identify the required configuration and provide a quote of how much the configuration will cost. Early on in the process, ask the prospect if upon successful completion of the beta, they will buy the product. This gives you an early indication of how serious the prospect considers your product and whether you are solving a problem that the prospect considers important enough to pay for the solution. You can extend this idea post beta and provide pilot installations as part of the first 5-10 sales prospects.

The goal is two fold;

• Get reference customers while minimizing the sales cycle - do everything you can to facilitate moving the prospect to a close (the "beta" plan).
• Get enough prospects to be able to recognize the type of prospects that you have a good chance to close in a reasonable time versus the lookie-loos.

Once you've got the first 5-10 prospects converted to initial customers, its time to pull out the rifle. The rifle is good at hitting visible, stationary targets. You want to score hits to the bullseye so get the marketing folks working double time to generate sales leads that match the criteria of your initial sales. Look for the defining attributes that help you identify why these early customers bought such as;

• Customer application environment - what surrounds your product that helps or hinders the sale?
• Customer vertical (legal, retail, manufacturing, financial etc) - will other companies in the same vertical have similar problems and hence be receptive to your solution
• ...

Part of building the repeatable sales process I wrote about last time is getting a good match of your product capabilities to the business needs and problems of your customer. Be thoughtful about casting the wide net but then go laser focused once you find the initial sweetspots...

One more thing; analyze the prospects you DIDNT get as betas or early sales. What set them apart from where you had success? Use this to help identify what capabilities are missing from your product or help avoid prospects that are a poor fit for the current product.

Making money out of old blogs

As you can see over on the right side of this page, I use Feedburner to syndicate my RSS feed. Feedburner collects some limited stats about your feed, for example how many times the feed has been pulled and by what kind of reader. Privacy is maintained as the IP address or domain of the reader isn't disclosed.

As I looked at the Feedburner stats today I saw a new reader program being logged - it was described as "Biz360 Spider" and included an email address and the URL for the Biz360 web site.

Curious minds need to know!... so I dutifully pulled up their web site and browsed around. Biz360 is a market intelligence company (venture financed by Foundation, Granite Ventures and others it would appear) that provides service to corporations by mining the news, databases etc. to establish a viewpoint for the corporation.

They also mine blogs using Feedster to determine real blogs from splogs and provide an end user perspective back to the corporation. Add another way of making money from the blogosphere to advertising and syndication. I wonder what other ways will be found to monetize user generated content?

P.S. I bet Philips didn't need Biz360 to understand the PR snafu they created by their patent filing to stop the viewer switching channels during commercials!

In-situ telecom upgrade - a first?

I got a postcard in the snail mail yesterday from Comcast who provides my Internet access via data over cable. The postcard is newsworthy for two reasons and both of them are firsts to me;

• It THANKED me for being a customer
• It told me that Comcast had UPGRADED my service from 7Mbps to 8Mbps - for free!

In all the years I've been a customer of different telecom service providers, I've never had the service upgraded without having to swap to a new facility (sure I upgraded! dial-up to ISDN to T1 to DSL and finally to Data over Cable - I have a need for speed!) and certainly never for free.

It highlights the fact that the cable guys "get it" - they recognize you have a choice and that they live in a competitive world. The Telcos have a long way to go in this regard.

What is more impressive is that this upgrade was done seamlessly; no equipment change on the customer premise and no on-site visit.

Hats off to Comcast on this one!

Will the future will be asymmetrically challenged?

A couple of weeks ago, as I was thinking about HTTP as a convenient way of punching holes through firewalls, I realized a future solution for those of us that use multiple computers; where do you keep your data so that its always accessible?

The problem is access to large amounts of data that get regularly updated. In my case, its my email that gets filed into folders. I use a hosted Exchange solution from DeskOptional (a great service provider BTW) that I access from my Treo using Goodlink, in the office from my laptop and from home with my desk top computer. I tend to file a lot of information into a set of Outlook PST files that live on the file server at home, hence the problem. The file server can't be accessed from my Treo and is challenging to access away from home.

Of course there are multiple solutions; from keeping all the data on the Exchange server, use a VPN to get secure access to my home network, all the way to keeping the PST file on a USB Flash Drive.

What would be really cool would be to keep all the data in the net! I started thinking about a file system that could be accessed over HTTP and a quick search showed that I wasn't the first one down this particular path.

I didn't think about it much more (too busy getting Ridgelift off the ground to go back to writing code :-) until I read about a new service that Microsoft is soon to launch - Live Drive. All though the details are still sketchy, both Microsoft and Google are talking about on-line data repositories that allows you to access your data from anywhere. Given how the costs of storage and bandwidth have fallen, the economics behind a subscription based on-line storage system are pretty attractive.

However, if this is the answer to my particular problem, it raises another challenge and the reason for the title of this post. A big percentage (maybe the majority) of broadband access is made over links that are asymmetric in terms of downstream and upstream bandwidth. For example, my home access to the Internet comes via data over cable - 6 Mbps down and 1 Mbps up.

Normally this is ok, in most systems, read operations outnumber write operations by 10:1 or 20:1 - we consume more information than we generate but a growing number of applications (digital pictures, digital video etc.) are eroding this disparity between reads and writes - at least for significant periods such as when you save all the pictures from your digital camera.

I've already been a victim of this change when uploading digital pictures of my kids sporting events to share with other families, friends etc. It takes a long time to upload a few hundred hi-res images when you are pushing data up a 1 Mbps pipe. Moving more data into the net will highlight this problem even more.

Most network build outs provide asymmetric service to the end user; it saves a lot of infrastructure and hence cost. Will new applications continue the erosion of the read to write ratio plus new services like Live Drive, cause a re-think of asymmetric links???

Web 2.0 - mostly features, not companies

My friend and former colleague at USVP, Baris Karadogan posted a list he'd compiled of Web 2.0 companies - its long! Baris segmented the list by area of focus, for example collaboration, messaging etc.

There are two things that strike me about the list;

1. Each category is really crowded!
2. Most of the categories are features that should be part of something larger, not large enough in their own right to be built out as a company.

Of course there are exceptions of categories where real companies have been built but they are few and far between in the list. Unless you get to be the first and biggest in a category, the odds of being acquired to be part of a bigger company aren't very attractive!

It's worth spending time thinking about what categories haven't been established and where enough value can be delivered to build a company, not a feature.

The majority of categories and companies are consumer facing; the enterprise space for Web 2.0 looks like a good hunting ground. Let me know if you'd like to brainstorm ideas or chat about opportunities.

How to limit progress - block what you fear!

I read an article in the WSJ last week about the growing number of companies that are blocking access to some services on the web because of a real or perceived security risk. Some of the services being blocked were predictable - Bit Torrent for example, but some of the others included IM and web hosted email like Hotmail etc.

This got me thinking about how every security measure tends to invoke a counter-security response; for example, Meebo provides web based IM access that needs nothing more than a browser - no client, no client side protocol, just goes through a firewall over HTTP (port 80).

There are a lot of different protocols that get carried using HTTP as a transport protocol for exactly this reason; web access is pervasive and has become a key business tool so there aren't many companies that actively block HTTP at their firewall. Some companies are now blocking access to particular web sites - both by name and address one would suspect.

From a business opportunity, I wonder what other mainstream applications could be enhanced and carried over HTTP or for that matter, what kinds of services that get blocked by the network security police might be interesting opportunities for counter-measures (but that make money!).

If you have ideas in this space, I'd like to hear about them.

Basic economics and Web 2.0

Robert's latest article for the DealMaker highlights a problem with a lot of Web 2.0 companies - it seems that the lessons from the 2000 Internet bubble have been forgotten. Bubble 1.0 was marked with a land grab mentality - build it and they will come. Without a plan for making money, even millions of loyal users won't help you survive.

Web 2.0 companies have to comply with the same business fundamentals as every other company. To win, you have to make a profit!

There's a simple equation I'd like to see considered in business plans;

Life time value of user > Acquistion cost of user + cost to service user + (overhead expense / number of users)

The life time value of a user could be the revenue generated through advertizing, product or service sales, lead generation, etc. It's the monetization plan that says what value is going to be created for the company by the user's actions.

The right hand side of the equation is a simple way of looking at what it costs you to get the user and deliver your business proposition to them. If you are the first mover in a particular segment your user acquistion costs are likely to be lower than for the companies that follow you - the first mover advantage.

Over time, you have to get the revenue greater than the costs - it's a pretty simple equation but one that seems to get lost in the rush of a land grab.

Soaring on updrafts!

On the quasi-humourous side, one of my colleagues forwarded me a pointer to this video.

It's both a great demonstration of the power of an updraft such as we're working to generate with Ridgelift Ventures as well as a great illustration of the power of the individual on the future of media!

I guess you have to go with the flow!

On-base percentage

Billy Beane, the general manager of the Oakland A's took a different approach to hiring baseball players by going back to the fundamentals of the game. The object of baseball is to get a player on base and then move him around the bases to score a run. It's not just about hitting home runs - these may be crowd pleasers but getting on base turns out to be the more "economic" way of playing the game. Billy went back to the fundamentals of the game and applied the scientific method to hiring baseball players. He analyzed a broader range of player stats that were indicative of the players ability to help win a game, not just the typical stats such as home runs, RBIs, etc.

By applying fundamentals and measurement, Billy made the A's one of the most successful teams in baseball and did it on a budget that was about one third of the other top teams. In economic terms, the A's win more games on fewer dollars and their cost per run earned is a factor of 5 or more less than their competitors.

You can apply the same process to business. Go back to the fundamentals - the object of a business is to make a profit. A profitable company has control of it's own destiny and no longer depends on infusions of cash to keep it going.

You can apply the scientific method to any business and at any stage of its life. A thoughtful business plan based on experience and observation provides the foundation for building the company. No business plan is perfect, it's only as good as the assumptions that are used in it's construction. The feedback loop is established through carefully chosen metrics that help to prove (or disprove) the assumptions behind the plan.

Both the type of business and the phase of the company's life factor into the choice of metrics. The key is to choose the key metrics that give early validation of assumptions and help refine the business plan. Get buy-in and feedback from the board of directors on the metrics that you choose to measure the business so that the metrics form part of the company dashboard. Be careful not to get carried away with too many metrics - there is always cost and effort associated with generating meaningful data.

Depending on the type of business, keeping track of progress in the very early stages of company life may lend itself more to milestones than metrics. Choose milestones that are of a fine enough granularity that you can track progress from week to week. When slips occur (and its inevitably when, not if), re-run the plan to determine the impact and consider how to recover.

The right metrics (or milestones) will give you insight into the reality of your plan and give you a tool to help determine how different actions drive you towards profitability. Like the A's, this will drive you towards the most economic way to build your business.

Another nail in Broadcast's coffin

Pick a point in the future, say 20 years out... it's not hard to believe that Broadcast media as we know it today will be dead.

The writing is already on the wall; entertainment consumption of broadcast TV is dropping, TiVo time shifts what YOU want to watch (and lets you skip the ads) while the number of hours people spend consuming Internet content is on the rise.

In the limit, the future isn't about Broadcast, its about Unicast; content that you choose, caters to your tastes and at times and places that are convenient to you. The majority of quality content is likely to still be produced for broad audience appeal because of financial constraints but you will have a staggering array of choices of content to listen to, watch or participate.

How will you find and select the content you want? Never mind "600 channels and nothing to watch", you will have access to so much content that finding the "good stuff" will be a daunting task.

One answer might lie in the extensions of word of mouth; for example, my kids got me watching Lost on TV (well really TiVo) - left to my own devices it's unlikely that I would have "discovered" Lost - I simply don't watch that much TV. As I wrote in my earlier posting about finding good stuff in the Blogosphere, my kids form part of my editorial review board.

Other answers might lie in new innovations like Pandora - Pandora is a really cool streaming service that provides music tailored to your choices much as Ringo (developed at the MIT Media Lab, an earlier approach to using social networking to identify music you would probably like given a sampling of your tastes) did almost 10 years ago. The confluence of on-line music distribution, broadband and advertising offers a way of building a new service.

Being a professional connector-of-the-dots (watching for early data points that suggest a trend), I can't help but wonder what other ways we will find to help navigate content.

Wireless when you can, Wired if you must!

The next time you are out and about, notice how many people are on their cell phones. Walking, driving, sitting in a restaurant, all around people are using wireless connectivity. The Blackberry and other PDA like phones are now being taken up by non-business people - email on the go is now mainstream.

Good wireless connectivity (good = wide coverage, decent bandwidth and affordability) is a catalyst that enables new and better services. Communication in many areas is now pervassive and available even without the tether of a wired connection.

This isn't a new phenomena, I co-led the original investments in @Road (NASDAQ: ARDI) alongside Pete Thomas (a great investor BTW) now of ATA Ventures in 1998. @Road leveraged GPS, the Internet and Wireless Data to become a leader in Mobile Resource Management (MRM). MRM provides business intelligence on the use of mobile assets - trucks, road sweepers, garbage trucks...) that enables better productivity, work scheduling etc and saves a ton of money.

In the same way that easily accessible communication provided by the Internet changed the way we think about applications, content and services, wireless gives us the opportunity to think about different solutions to a new set of problems.

Here's a recent example, Mapquest is introducing a new range of services that enable a cell phone to be used in place of an in-vehcile navigation system. This is another example of what I call an Infrastructure-LESS Services Company - a company that leverages all the infrastructure that has been built out over the last 10 years including the Internet, GPS, cellular, Open Source (yes, I view Open Source as infrastructure - a set of capabilities that gives leverage in building new things), etc.

An interesting thought to ponder is what other kinds of opportunities does wireless open up in similar ways? If you can, go wireless, if you can't then wired lets you leverage much of the same infrastructure but without the benefit of portability.

Director's Cut

I just received this week's update of Investment Strategy Outlook, a weekly newsletter written by Jim Anderson of Silicon Valley Bank's Asset Management group. Jim's a good writer and the pieces are often thought provoking.

This week's lead is titled "Director's Cut" and its about the impact that a really active and insightful board member can have on your board. Jim cites the examples of Steve Jobs joining the Disney board and Jerry York joining the GM board. I read an article last week (it was either in the NYT or the WSJ) on Jerry and his impact both on IBM and now on GM.

Jim compares the roles of these very different people and describes them as change agents. They are that and much more; they have pattern recognition. Both of these board members have had extensive experience and can connect the dots - they see the pattern based on their experiences and can relate them to a new business and its own problems.

Now, here's the rub; instead of bringing this kind of change agent in to a challenging situation, what if you had the same expertise along side you from the beginning in a private company? Someone as a working board member who could apply pattern recognition skills to the challenges of a business while the "challenges" are still small and help the CEO and management team from the beginning?

That's much more (to coin a Brit expression) my cup of tea!

Next wave of network build out?

There’s been a whole lot written over the last few weeks about the Telco’s saying that they are going to charge extra for delivery of premium content (such as video) from Google, Yahoo et al to their users. Aside from the obvious point that users are already paying to use communications facilities from the Telco’s, it makes me wonder whether we’re close to another wave of network build out.

The New York Times yesterday had an article about the Telco mega mergers and again posed the question of the Telco’s charging more for content delivery. The same article also talked about the increasing rate of network infrastructure spent by the Telco’s – one could say “Finally!” especially after the capacity glut that resulted from the Bubble build-out. Interestingly, for a couple of years after the bubble, the research guys at Goldman-Sachs did some in-depth reviews of capacity and capital equipment spending and projected that the glut would last through to at least 2007 – so maybe we’re getting pretty close.

My guess is that a move to charge for “premium” content delivery by the Telco’s could spur another round of network build out by Yahoo, Google and MSN. They already have extensive networks of their own that are used to pull content over private backbones. The next wave will be at the edge of the network. You can see some of the precursors of this move in Google teaming up to provide Metro WiFi services as well as their purchases of dark fiber reported by Om Malek and others over the last few months.

Another wave of edge build out could spur the creation of new services and new communication products. There’s been a virtual drought of new ideas in this space over the last few years and it would be nice to see people starting to think outside the box again.

Build before you burn

One of my co-founders of Ridgelift, Robert Goldberg wrote an interesting article on proving out the business model of a start-up before spending all the hard raised capital. The idea is that you put together a set of milestones that step wise answer questions and reduce risk before significantly increasing the burn rate of your company.

I guess another way to look at it is that you don't make all your mistakes in parallel! A common issue with start-ups as they get into sales is building out the sales force too rapidly. Common wisdom is to hire a VP of Sales, build out all the territory sales people and sales engineers in one go. No better way of increasing expenses rapidly! Then you find that there a problems selling, perhaps the product isn't a perfect fit for the customer or you find the sales cycle in longer than expected.

The end result is parallel learning and its expensive!

Better to hire one or two sales people and figure out what it takes to get a repeatable sale. If you are the CEO of the company, you should be out making the sales calls with the sales people so that you can see the issues first hand. A really good VP of Sales can be invaluable in this phase as long as they are prepared to get their hands dirty as one of the two sales people and are more of a hunter than a farmer.

Once you've got a repeatable sales process, you can expand the sales force with more confidence that the revenue growth will be more in sync with the expense growth.

Finding stuff in the Blogosphere

Last weekend I got to hang out with some of the brightest minds I’ve been fortunate to meet in a long time and we got talking about blogs. Several of the folks are active bloggers and a straw poll showed that pretty much everyone but me are active readers of multiple blogs. Naturally, all eyes turned to me – “Gee Stu, as a tech investing VC you don’t read blogs?” was pretty much the unspoken question.

As it turns out, I do read blogs – more accurately blog articles that are sent to me by friends. The blogosphere is huge and there is a great deal of very good content but it’s too difficult to find. A search engine is a blunt tool for finding relevant content unless you are interested in the same stuff as everyone else. What’s missing is the editorial review that provides the “Hey people, read this” effect. My friends are a virtual editorial board for my access to blogs.

Ted Shelton’s new company Personal Bee is a step towards engaging the community as an editorial board for those of us with more focused interests.