Hello from VON in Boston. I spent some time today at the “unconference”, and really enjoyed myself. It was great to see in person all the bloggers I’ve been reading over the last year. It’s strange: I thought they were all taller.
As the new guy in town, I mostly just observed but — as promised — it was a genuine interactive environment. That’s a nice escape from the standard session format.
Some of the issues that came up:
* Who’s life is it anyway?
Aswath Rao challenged the statement that they give you phone service “for life” in exchange for a one time cost of $400. “It’s really the lifetime of the box, isn’t it?” he asked. Dennis confirmed that yes, when your Ooma box dies, you have to buy a new one.
* Biz Model
Carl Ford took a few swings at their business model. I loved this line: “I smell head count. Which means I smell burn rate. How are you going to monetize these customers down the road?”
* Ease of Use
Carl also warned about run-away customer support costs. Dennis replied with some interesting stats that I hadn’t heard before: “60% of their users were able to do the install in 15 minutes. 90% within 30.” That’s impressive. For comparison, here is a video showing a PhoneGnome installation.
* Watch the fine print
Andy Abramson was very critical of Ooma for downplaying some major T’s and C’s till the end of the sign-up process. I didn’t follow all of the details there but, apparently, the small print allows them to take over as your long distance provider and change settings on your phone service.
I asked about a security matter: “Can I listen in on conversations from other Ooma users flowing through my Ooma box?” (See here for some background on the issue.) That set the room twittering (the old verbal kind, not the online kind). Dennis repeated the corporate line several times: “We take the security of our customers very seriously.” but that didn’t seem to placate the crowd.
* Bigger picture
What I found most interesting was Dennis downplaying their distributed termination technology. And although my inner geek thinks it’s really really cool, I have to admit it’s starting to look like more trouble than it’s worth. He called it a “short term solution” along the way to “an all-peered approach.” He wants us to focus instead on the built-in features of the system (instant second line, smart answering-machine) and the potential of this platform for the future.
For the record, let me say that I’m a big fan of the Ooma. After Jon and Alec, I think I’m the 3rd guy in Canada to have one. And, more importantly, I want them to succeed as a platform because I want to use the Ooma network to get FonCloud features out to consumers. (It will be a heck of a lot easier than doing carrier deals.)
Sadly, the lively discussion got cut short by the next presentation. The intention (well-placed) was to show how cool Adhearsion is, but it killed the buzz for me. Yes, I get it: high level languages let you put together simple apps quickly. But I don’t need to see it … being…. typed… in… real… time.
[This is the second post inspired by the NMS Connect conference. See also: There's More to Pipes than Smart vs Dumb.]
Voice 2.0 companies can be divided fairly cleanly into two groups. Those who use carriers to deliver features and connect with their users and those who “go direct”. In some cases, two companies have a similar product but on the opposite sides of this divide.
For example, GotVoice will retrieve your voicemail, transcribe it and send it to you. SpinVox does the same thing, but through carrier cooperation. (They just announced a deal with Rogers, Canada’s largest mobile carrier.) Similarly, Kirusa provides a “Voice SMS” service through carriers. Pinger provides a similar product but direct to consumer.
Kirusa’s carrier strategy turned out fruitful, but it certainly wasn’t easy. I met CEO Inderpal Singh Mumick a few months ago through a mutual connection and, at the Connect conference, I spoke at length to CTO Ewald Anderl. The stories I heard from both of them followed the same theme I’ve been hearing from entrepreneurs and investors for several years: If you want these deals, you need experienced biz dev people and you need patience. Both of those things translate into “you need cash”. Many start-ups have run out of gas waiting for that carrier deal to get signed; or if it’s signed, for the service to be launched; or if it’s launched, for the marketing push to materialize.
With all this trouble, why do we continue to line up outside the carrier’s door? Because there are things that only carriers can bring to the table…
Carriers communicate regularly with their customers through the billing process and marketing campaigns. They also have large PR budgets and a team that can tune your product pitch to match their user base. Here’s a TV ad, run by GrameenPhone in Bangladesh, showcasing the emotional power of Kirusa’a Voice SMS product. How many start-ups can afford a TV ad?
(Direct link to video here.)
I’m looking forward to seeing how Rogers deploys the SpinVox product, in particular how they choose to promote it. If they’re smart, they will automatically give everyone a 15-day free trial. (I’ve used CallWave and I have to say that getting your voicemail as text is addictive. It’s painful to go back.) That kind of broad exposure to new customers is something only a carrier can do.
Carriers can integrate a new product into the phone experience in ways that aren’t possible otherwise. Back to Kirusa vs Pinger: Both provide a way to record an audio message and send it to someone. With Kirusa, the process is streamlined by using “short-codes” with the pound or star key. But with Pinger, you have to call Pinger first and that adds significant friction to the process.
411 is another example. There are several options for free directory service such as 800-Goog-411 or 800-free-411. But the majority of people continue to dial 411 the old fashioned way and pay a fee of a dollar or more (and rising). There is a powerful mental lock-in with those 3-number short-codes and only a carrier can assign them.
Services that require an application on the mobile device face additional barriers: Consumers are leery of downloading anything to their phones and the process is not very smooth. Carriers can solve both problems by pre-installing software.Susan Norris, a 25 year veteran of the carrier world, summarize this during one of the sessions: “If someone has to remember to use an application, it won’t work. It needs to come at them.”
This one is self-explanatory — carriers have the infrastructure to create, send out and collect on monthly bills — but don’t underestimate it. Keep in mind the importance of: telephone customer support; setting up free trials, bundles and other promotions; integration with pre-paid plans; extending the right amount of credit to each user.
SoonR recently refocused their efforts to carrier partnerships. Their new CEO, Patrick McVeigh, specifically cited billing as one of the drivers: “[Our] strategy is to go with carriers or large partners … who already have existing billing relationships.” (More on that from GigaOm.)
The bottom line
The Jaiku story shows us that, with an excellent product, you can grow without carrier involvement at all and reach a nice exit. (Google bought them earlier this week.) But counting on an early buy-out isn’t a solid strategy. Stan Reiss, from Matrix Partners, put it this way during one of the Connect sessions: “As a start-up you have to find a way to stand up to the carriers in the long run.”
In my view, the best approach is to build a product that can go directly to consumers right out of the gate, and then repackage it for carriers down the road. This way, by the time you are approaching a carrier, your user base has helped you refine the product and (hopefully) become a source of revenue. That’s our plan with Fōncloud.
I was at the O’Reilly’s ETel conference when Jaiku launched. (The running joke there was “This conference brought to you by the letter J” … Jaxster, Jangl, Jaduka, Jaiku.)
The immediate comparison was made with Twitter. (From Mashable: Jaiku Launches, Looks Like Twitter To Me.) And that rivalry continued to define them, just search on “Jaiku vs Twitter” for some fun reading.
Yes, those two are similar on a functional level, but I always felt that the key value to Jaiku is their mobile application and the way it’s integrated tightly with their web app. Designing a good mobile web app — one that people will use beyond the initial facination — is tough. There is still a significant barrier to downloading and installing a mobile app so getting users to jump that barrier is very impressive. (Doubly so, given that this is not about cheaper calling.)
Om says: “…time for me to sit down and start making sense of all the Google acquisitions in the mobile and VoIP space…”
I think their recent acquisitions illuminate their GooglePhone strategy: Google is looking for ways to improve the phone experience without cooperation from the carriers.
They’ve learned a lot from the Apple-AT&T arrangement (especially given the Google-Apple connection at board level) and the lesson is that carrier integration is messy and takes a long time. (And it doesn’t get less messy after you launch – e.g. unlocking, bricking, and lawsuits.)
But at the same time, Google knows carrier deals are the only way to have broad availability of the GooglePhone (even if they win the spectrum auction).
Can’t live with ‘em, can’t live without ‘em. So what do you do?
You make sure your product works well out of the gate without the carriers, then use your early success to negotiate good deals. The ideal situation for the GooglePhone is that it’s great on its own, but AWESOME with carrier integration.
So… from GrandCentral they get inbound call management (routing and filtering). DIDs can be handed out to users without any carrier assistance. From Jaiku and Zingku they get mobile messaging and presence apps that work well “off-deck” — again no carrier assistance. Also from those two, they get teams that are good at the nuances of designing mobile UIs that people like. (As mentioned above — that isn’t easy.)
[This is the first of several posts digesting what I learned at the very informative NMS Connect conference. Their CTO, Brough Turner, was live-blogging and so you can read the blow-by-blow action of each session here.]
One of the themes that ran across multiple sessions and hallway discussions was: What should be the role of the wireless carrier in new data-driven telephony applications? Or, as it is commonly simplified: Should the carriers be providing dumb or smart pipes?
In the “dumb pipe” approach, the carrier provides direct internet access to handsets and then stays out of the way. They make money by charging per bit and treat all bits the same. The hope is that having an open platform will attract developers to create millions of mobile applications, some of which will become runaway hits that can survive on advertising, while others find ways to monetize the long tail of niche users. (We’re nearing 3 billion people on mobile phones so it’s hard to imagine that you could foresee their needs through design-by-committee.)
On the other hand, the “smart pipe” approach has the carrier act as both content and access provider, controlling the whole the experience. This allows the carrier to charge based on function, not on bits, and to prioritize traffic on their networks more intelligently. That way real-time applications like video calls won’t be sacrificed by someone on the same cell tower downloading a song. And that way we can keep the mobile world more “tidy” — free from spam, phishing and other nasty things. I’m certainly willing to pay a bit extra for that.
This issue was at the center of the conference’s most lively debate. Rooting for dumb pipes was Seamus McAteer from M:Metrics “The endgame is the mobile internet, not something from the 3GPP. They will never keep up with innovation. That’s what users want.”
In the other corner, Andrew Bud from mBlox: “I fundamentally disagree. The internet model is not a great way for service providers to get any value….”
I’m not sure where this dumb vs smart paradigm started although it certainly got a big boost with the 1997 essay Rise of the Stupid Network. As a short-hand for broad discussion, “smart vs dumb” is useful but as we drill-down into the details we need better terminology. The word “smart” is just as ambiguous when describing a pipe as when describing a person. A pipe can be smart in different ways.
As it turns out Andrew was arguing in favor of a particular kind of smart pipe, where carriers partner with 3rd parties for content but keep close control (and a cut of revenue) over all transactions happening on their network. (It’s no coincidence that his company enables exactly this kind of relationship.)
The nature of a carrier-type organization is not conducive to the creativity you need to create new applications. Dean Bubley nailed this point with a question from the audience — and I’m paraphrasing here — “The coolness lifespan of an internet service is about 18-24 months, which is about how long it takes for a carrier to roll out a service. So how can you possibly keep up?” Russ McGuire from Sprint-Nextel admitted: “We know we are not going to be the innovators. But we want to be sure we attract innovators in a fashion that doesn’t leave us in the position of being a dumb pipe.”
On the other hand, there is a strong and valuable relationship between carrier and subscriber that doesn’t exist in the wireline world. Michael Scully from Virgin Mobile put it this way: “If I buy a ringtone on my cell phone, even from a third party, and if there’s a problem, I call my cellular provider. After all, they’re the ones that billed me for it, and I may not even know how to contact a third party if I even know that I bought from one. Conversely, if I have an issue with an order on Amazon, I don’t call Comcast.”
Carriers have the infrastructure to do itemized billing, access to credit and payment history and hardware-based identity tracking (through the SIM card). It would be a shame to throw this all away and put each new mobile application in its own “sand-box” that knows nothing about me and needs its own user name, password, etc.
So this leaves us with a stand-off that has led to some pretty unfortunate situations as chronicled by Corey Doctorow in a recent essay…
Every time we butt up against some rotten ‘feature’ in the mobile phone world — screwy data-pricing… carrier-locked handsets, handsets that can’t run user-installed software, APIs that can’t talk to the phone’s radio hardware … phones that won’t play your own music or movies, phones whose numbers can’t be ported to another carrier, phones where number portability takes weeks or months, “unlimited” data-plans that cut you off if you use too much data — everyone’s got someone else to blame. It’s the greedy, stupid, dinosauric carriers. It’s the wimpy, gutless phone manufacturers. It’s the coked-up Hollyweird fat cats from the record industry. It starts to feel like a Mexican standoff, three tough guys, each pointing a gun at the others’ heads, deadlocked and unwilling to risk anything to break the standoff.
I’ll be a bit less dramatic and say simply that both sides need to give some ground:
- Carriers need to realize that creativity lives at the edges, and that they will never be able to design the next Twitter or Mig33 or Facebook.
- Developers need to realize that there is unique value in the relationship between carrier and subscriber and it would be a shame to throw it away.
Last year, one of the best events I attended was the O’Reilly Emerging Telephony conference. And one of the most interesting people I met there was George Kontopidis from NMS. He helped put the Voice 2.0 hype in perspective with some straight talk about the realities of getting new features out in wide scale mobile deployments.
Later that year, I went to TMC’s Internet Telephony conference, and one of the most interesting people I met was Brough Turner, who is the CTO at NMS. This time, it was IMS hype that got cut down to size. Brough’s blog has been an excellent source of information ever since.
Getting our carrier strategy right is critical to the success of FōnCloud and so learning how that industry ticks is a key goal for me.
- November 2011
- May 2011
- April 2011
- March 2011
- February 2011
- December 2010
- November 2010
- October 2010
- April 2010
- February 2010
- January 2010
- September 2009
- August 2009
- July 2009
- June 2009
- May 2009
- April 2009
- March 2009
- February 2009
- January 2009
- December 2008
- September 2008
- August 2008
- July 2008
- June 2008
- May 2008
- April 2008
- March 2008
- February 2008
- January 2008
- December 2007
- November 2007
- October 2007
- September 2007
- August 2007
- July 2007
- June 2007
- May 2007
- March 2007
- February 2007
- January 2007