I was recently asked what I thought about the future of voicemail.  I had to give this some thought.  Voicemail isn't one of those things I think about often.  In all honesty, I try to avoid voicemail.  I regard voicemail as a nuisance.  In fact, I haven't even set up my voicemail box at Cisco yet, and I've been working there almost a year.

But why does it have to be this way?  Surely there's value in hearing someone's voice rather than forcing them to type out a message (especially if they're a hunt-and-peck'er).  Not only can I catch the nuance in their voice and potentially derive more information from the message, but it's much easier on the person who's sending the message, especially if they don't spend much time in front of a keyboard.  Voice, by its nature, has a much lower participation premium associated with using it than text does.  There must be a solution to this problem, because there is value to be added there.

For one thing, there's the unified communications issue.  I hate multiple phone numbers.  My online identity is already fragmented enough, I don't want to be forced to deal with multiple phone numbers as well.  I have a phone, it's my cell phone.  Anything beyond that is superfluous and a hassle.  I never, EVER use my office phone, except for outgoing calls when I'm sitting at my desk.  I don't use my home phone, either.  I don't even know the number.  The ONLY number I ever give out anymore is my GrandCentral number.  A device is just a conduit to my online identity, which is my connection to the network.  That's my definition of unified communication.  I don't care if it's my desktop, my laptop, my desk phone, my mobile phone or my refrigerator, I am still me.  I only need one point of contact, the device is just a consequence of my location.  Don't force me to remember more than one identifier.  In the short term I'm ok with separate identifiers for data, email and voice, but that too will converge (see:  XMPP!)

So why do I use GrandCentral as my sole phone number?  Because I can point it at other phone numbers that I don't then have to remember, so if I have to switch mobile numbers at some point it's no big deal.  No need to send out updates to my entire contact list, they already have it.  And because I tell GrandCentral how to handle the call.  If the caller is one of my VIP contacts they can get through to me at any time.  If they're not, I can set up rules around what times they're allowed to contact me before they're sent to voicemail.  If I don't know the person calling (they're not in my contact list), GrandCentral will ask them to identify themselves before asking me if I want to take their call.  This saves me time, my most precious asset.

Beyond that, GrandCentral extends what I can do with a voicemail.  The voicemail I get with my mobile provider is dumb.  All I can do is check it, save it, delete it, and forward it (if I'm lucky, this has never actually worked for me).  Look at all the things GrandCentral will let me do with a voicemail:

There's a whole slew of things I can do with this that adds value to the voicemail beyond the original message:

• Flag it for followup
• Add the caller to my contacts
• Send the voicemail via email
• Map where the call came from
• Embed the voicemail on a Web page

The only thing missing is to send the voicemail via instant message or Twitter!  Well, that and speech-to-text.  I really love personal voice recognition, and Jott has really gotten me spoiled on it.  There are times, like when I'm driving, when I can't type something easily and the only data input method I have available to me is voice.  At these times there is no alternative to voice if I want to send someone a message.  HOWEVER, they might be like me and would rather read the message in text form instead of voicemail.  Speech-to-text allows me to take a voice message and insert it into the global data stream, whether that be the internal email system, Twitter, or an application.

Voicemail is just another type of message.  The method I use to put the message into the network should be irrelevant, whether it's a keyboard, my voice, or a video camera.  All that matters is that the recipient receives the message and that it is EASY for him to use it (not just hear/see it, but USE it).  If there is even a momentary hesitation because he has to think about the steps to retrieve it or remember a password or a number, the message delivery has FAILED for all intents and purposes.  And once he has the message, it should also be easy for him to re-use the message in another system as he sees fit, delivery should not be the end of the message's useful life.

So that is my thinking on voicemail in the traditional sense.  I call you, don't reach you, and leave a message.  I think this type of voicemail has somewhat limited value.

In another sense, however, voicemail can be extremely powerful.  If you think of it as voice MESSAGES instead of voice MAIL, a whole new world of opportunity opens up for it.  Instead of a verbal sticky note, the voice message is freed to do all kinds of things that the voicemail paradigm doesn't really allow.  GrandCentral hints at this, but there is more.

For example, consider the traditional voicemail publishing paradigm.  It is almost always one to one.  One sender, one recipient.  But this is not always how a voice message SHOULD be distributed.  If you crack this nut open and allow voice messages to be broadcast to a wider audience, it becomes much more powerful.  Instead of simply saying "call me back", you can use the power of voice and video to convey the nuances in a conversation and text just can't capture.

Podcasting hints at this, but the participation premium is too high.  You must have a podcasting system set up, and very few people will take the time to rehearse a podcast and polish it until it's ready for distribution.  If it's quick and easy to broadcast a rich media message to the people who want to hear/see it, however, much more content will flow and communication will improve.  This is the Twitter philosophy and it applies to voice and video as well as text.

I'm thinking of real-time, rich conversations here.  Virtual meetings save time and money, yes?  Why must it always be a scheduled meeting?  If I'm working with one person in San Jose and another in Timbuktu, why should we have to schedule a time to meet when somebody's going to have to get up in the middle of the night?  Instead, let's inject our thoughts into a real-time rich-media conversation.  If I'm working on it and have a thought about something, I should be able to use a Jott-like application to send that thought in verbal form to everyone who wants to hear it, instead of trying to figure out a way to remember it for the next meeting.

The traditional meeting is broken in a bad way, and the whole concept needs to be rethought.  It just doesn't work well for widely distributed teams (let alone small, local teams--the potential for abuse here is just mind-boggling).  It's inefficient and it's not ENOUGH interaction.  Instant messaging helps, but it doesn't scale well, and it relies of people to take the initiative to tell everyone who needs to know what they're saying.  What's needed is a multi-media conversation cloud, and what better way to approach that than with the tool that's been used for conversations since the 1800's, voice?  Yay, we'll be back in 1890 again someday soon! :)