DoCoMo 4G: 300 Mbps to your Mobile

NTT DoCoMo is about to embark on an ambitious project that provides cellphone users the ability to achieve speeds of up to 300 Mbps on their handsets by the time 2009 rolls around. That, plus at&t's new positioning as a wireless company with landline assets, plus the fact that global "voice account" installed base and growth are killing landlines, has to be disquieting for lots of us who grew up on the wireline side of the business. Wireless is going to keep changing things more than some might like.

SunRocket post mortem

Chris Koehncke, a former SunRocket excutive now back at BroadSoft, has this to say about what went wrong at SunRocket (I'm not so convinced the customer acquisition strategy was necessarily so doomed to failure, but more on that later): "It appears that SunRocket was indeed acquired and with it the vast majority of the remaining employees unceremoniously laid off sans a small group of 15-20 of the folks necessary to keep the network alive. SunRocket network suppliers were notified late Friday so they wouldn't turn them off.

With 200k subscribers and say a $40m revenue base, if they simply stop marketing, give crappy customer support and run it on a shoestring -- it's a nice little pocket business. Playing in the fringes.

In hindsight, it was kinda of crazy to spend $200+ to acquire a customer whose annual spend was only $200+. Hard to outrun the economics. The hopes, of course, was that if you reached a decent size, people would come to you and you could reduce your marketing costs. Vonage certainly hasn't seen this, so SunRocket wasn't going to be much better.

Customer service cost also can eat you alive. Anytime you call for customer service, figure it's costing that company between $0.75 and $1.00 PER MINUTE to talk to you. For a typical 10 minute call that's $8-10. Call twice in a month and for a SunRocket $17 a month revenue, they lost money on you. It's also amazing the number of people who buy cheap are also the same group who complain constantly. They want it all and they're willing to bitch about it.

Sprint's decision to whack 1000 customers who were constantly calling customer support (I suspect their initial list had 10k names on it) is wise. Perhaps, SunRocket simply should not have had telephony support at all and just offered email support.

Some folks have just crappy internet service and SunRocket was never going to fix that. Perhaps after 3 customer support calls, SunRocket should have pro-actively canceled the customer's account. Why keep a customer that will never be happy?

Perhaps rather than being a nationwide telephone service, SunRocket should have focused on a specific region or vertical consumer group. Better targeting their marketing spend and creating a reputation in a specific niche.

VC's are an inpatient lot. They give a start-up money and urge them to spend it as fast as possible and grow as quickly as they can. In doing that, the start-up makes mistakes, does stupid things, hires the wrong people, it's a wild ride for sure. VC's don't mind failures, that's the business they're in, but if it's going to fail, they want it fail fast so they can move on to the next hot idea. SunRocket had to build a complete telephone company literally overnight with hundreds of moving parts so it's not surprising mistakes were made.

The genesis of the idea of SunRocket is still valid, create something different that people will be loyal to and ultimately becomes virally marketed. But by buying off the shelf VoIP and traditional telecom products the ability to differentiate the service, other than with a pricing model, was nil. You can't buy innovation in a catalog."

Discovering Business Models

The problem with discovering business models is that what works for some does not work for all. Back in 1998 and 1999 the stock answer provided by just about any competitive local exchange carrier executive essentially was that the firm in question would get "one percent of a $250 billion market."

These days people ask how Facebook, messaging, collaboration, video or other portals will make any money. The most popular answer is some variant of the old CLEC standby. U.S. advertising currently is about a $153 billion a year business. Portal X will get one percent of that.

Look, it clearly works for four companies: Google, AOL, Yahoo! and MSN. The "four horsemen" get about 60 percent of all Internet advertising. It isn't going to work for most application, communications or portal providers, just as it never worked for most CLECs.

The biggest two "CLECs"--the former AT&T and WorldCom/MCI--threw in the towel in defeat. And those two had more than 40 percent of all "CLEC" revenues between them.

So people assume that fast-growing and useful sites such as Facebook will find some way to make money besides traditional advertising. And there is precedent for such discovery.

Google was equally clueless about its business model, but managed to discover one.
So just because a company has no idea how it will make money, doesn't mean it will not discover a means.

On the other hand, that doesn't mean it ALWAYS will find the answer. And though I'd have to say I am fairly confident Facebook will discover a model, as Google did, that doesn't mean thousands of other sites will be so lucky. Thousands of sites obviously cannot use the Internet advertising model, even if it is fast growing, because most fo the rewards will go a relative handful of companies.

More Enterprise Moves for Google

Google now is any tier one communications service provider's biggest fear, stated or unstated, and it mostly is stated. So Google's acquisition of Postini won't help clam any such fears. Google gains instant access to Postini’s client base of more than 35,000 businesses including many large businesses.

Postini will help Google further refine its "software as a service" architecture for critical business applications, moving Google further into the business services and software space.

Postini has offered managed email security services to corporate customers since 1999.

Jangl to Voice Enable Facebook

Jangl is getting ready to announce Phonebook for Facebook, which puts calling and voicemail right in a user profile and inbox on Facebook. The new feature will allow Facebook members who both have the application to call each other, visually manage voice mail messages in the Facebook inbox and show a current online presence. Voice is becoming an application available within a user's current context.

Lights Out for SunRocket?

SunRocket appears to be refusing to accept new customers. Try calling the call center to sign up. Vonage apparently has offered to buy SunRocket for no cash, simply to provide continuity of service. No word on any response from SunRocket. A couple more bidders, said to be undercapitalized themselves, also are poking around. Not a happy day, at all.

25 Startup Commandments: Great Stuff!

1. Your idea isn't new. Pick an idea; at least 50 other people have thought of it. Get over your stunning brilliance and realize that execution matters more.

2. Stealth startups suck. You're not working on the Manhattan Project, Einstein. Get something out as quickly as possible and promote the hell out of it.

3. If you don't have scaling problems, you're not growing fast enough.

4. If you're successful, people will try to take advantage of you. Hope that you're in that position, and hope that you're smart enough to not fall for it.

5. People will tell you they know more than you do. If that's really the case, you shouldn't be doing your startup.

6. Your competition will inflate their numbers. Take any startup traffic number and slash it in half. At least.

7. Perfection is the enemy of good enough. Leonardo could paint the Mona Lisa only once. You, Bob Ross, can push a bug release every 5 minutes because you were at least smart enough to do a web app.

8. The size of your startup is not a reflection of your manhood. More employees does not make you more of a man (or woman as the case may be).

9. You don't need business development people. If you're successful, companies will come to you. The deals will still be distractions and not worth doing, but at least you're not spending any effort trying to get them.

10. You have to be wrong in the head to start a company. But we have all the fun.

11. Starting a company will teach you what it's like to be a manic depressive. They, at least, can take medication.

12. Your startup isn't succeeding? You have two options: go home with your tail between your legs or do something about it. What's it going to be?

13. If you don't pay attention to your competition, they will turn out to be geniuses and will crush you. If you do pay attention to them, they will turn out to be idiots and you will have wasted your time. Which would you prefer?

14. Startups are not a democracy. Want a democracy? Go run for class president, Bueller.

15. You're doing a web app, right? This isn't the 1980s. Your crummy, half-assed web app will still be more successful than your competitor's most polished software application.

+10 More Startup Commandments

1. You will have at least one catastrophe every three months.

2. Outsource effectively, or be effectively outsourced.

3. Do you thrive on stress and ambiguity? You'd better.

4. The best way to get outside funding is to be successful already. Stupid but true. But you, cheapskate, don't need money, right?

5. People will think your idea sucks. They're even probably right. The only way to prove them wrong is to succeed.

6. A startup will require your complete attention and devotion. Thought your first love in High School was clingy? You can't take out a restraining order on your startup.

7. Being an entrepreneur requires a healthy amount of ignorance. Note I did not say stupidity.

8. Your software sucks. So what. Everyone else's does also, and re-architecting is the kiss of death for a startup. Startups are no place for architecture astronauts.

9. You do have a public API, right?

10. Abject Terror. Overwhelming Joy. Monstrous Greed. Embrace and harness these emotions you must.

Provided by Mark Fletcher, http://www.startupping.com/