Infectious

Google Wave vs. Cc:Betty: Group communication & collaboration

If you’ve been keeping track of innovative products on the web than you probably saw the pre-launch of Google Wave last week at the Google I/O conference. If you didn’t see the demo take a look because it’s slick.

If you don’t have time for the demo (it's over 1+hours) let me wrap up Google Wave in a nutshell (explained by Google):

• A wave is equal parts conversation and document. People can communicate and work together with richly formatted text, photos, videos, maps, and more.
• A wave is shared. Any participant can reply anywhere in the message, edit the content and add participants at any point in the process. Then playback lets anyone rewind the wave to see who said what and when.
• A wave is live. With live transmission as you type, participants on a wave can have faster conversations, see edits and interact with extensions in real-time. 

Here is a Wave in action:

Cc:Betty's 'Mailspace' concept is like a Wave
One drawback to this innovation is that Google Wave won’t be launched for the next year. But, you don’t have to wait if you’re looking for a better way to communicate and collaborate with groups. Instead, you’ll want to look at Cc:Betty which is available today. Here is the demo.
Watch for the similarities between Google Wave and Cc:Betty.

In a nutshell, Cc:Betty’s ‘Mailspace’ concept focuses on many of the same use cases as Google Wave. They both are focused on the fact that email hasn’t innovated to accommodate group interactions. Here is a Mailspace in action:

What’s surprising to me is that these solutions haven’t come along earlier. Regardless, both of these projects are ambitious. But if they weren’t, what would be the point;)

Patterns of Startup Success: Vision Matters

Vision is a big word when your company is just an idea on the white board. In fact, I rarely use the word but instead ask the question 'What do you hope for your company and customers 10 years out?' What's so remarkable about this question is that it asks an entrepreneur to take a perspective on what the world will look like in the future. It also drives every decision thereafter.

Vision is driven by an entrepreneurs perspective

Every entrepreneur comes to the game with a different set of experiences. Those set of experiences determine how they see an opportunity. In fact, two entrepreneurs coming from different experiences will more times than not see an opportunity differently.

Right out of business school (1999) I worked for a startup called Sylantro Systems which was started by a group of entrepreneurs from the Telecom space. Pete Bonee, the CEO, wanted to deliver hosted IP telecommunication services to businesses. Pete's vision came from his experiences delivering hosted voicemail solutions (Octel) and hosted conferencing solutions (Latitude) to service providers. He saw the world through a lens that was heavily carrier-centric. This ultimately led Sylantro down a path that required it to play in the carrier world. This also led Sylantro down the path of tying its success to service providers. Regardless of the outcome every entrepreneur comes to the table with a set of experiences that drive their truisms.

Vision does not come by sitting in a room

Because vision is about having a perspective it shouldn't be built sitting in a room. It should be built by triangulating as many inputs as possible. Looking at the current mega-trends (i.e. aging population), meeting with potential customers, scoping competitors and industry experts should all go into a vision's crafting.

Vision matters

In the end, vision has an enormous impact on the bottom-line. There is no more personal example for me than looking at the current outcomes of Yahoo! and Google. When I worked for Yahoo! in early 2000 Yahoo's vision had something to do with providing users with delightful web experiences. It's pathetic I can't  remember the details and I bet you'd find very few who could in those days. In fact, I did a web search and couldn't find anything specific on the subject! However, almost every Yahoo! employee at the time knew Google's vision, "To organize the world's information and make it universally accessible and useful." The current numbers tell the rest of the story.

Patterns of Startup Success: Customer Development using Test, Rinse, Repeat

Chris Lunt is currently the CEO of  Nombray, a startup that helps you to own your name on the web. Simply...Get a domain, collect your content, control your destiny. In the following guest post, he elaborates on some of his experiences at Friendster during it's meteoric growth. Read on for his facinating thoughts around Customer Development and the value of "Test, Rinse and Repeat".

Friendster was the first modern social network to gain a large audience, very quickly growing to 3 million registrations in the summer of 2003.  Jonathan Abrams hired me to help with the scaling issues that appeared with such rapid growth.  Beyond the scaling issues, there was a broader question of "what is this site for?"  Jonathan had conceived of the site in a dating context, but his VCs encouraged him to think more broadly.  What followed was both a tragedy and a comedy.  And I share the blame and the acclaim for the strange journey Friendster began at that time.

Jonathan had a grand vision of the site as a portal that would frame all of your web interactions, encompassing music, instant messaging, blogging, etc.  He laid these ideas out immediately after the 2004 New Year's Day.  At the time, the traffic to the site was completely performance bound, and Jeff Winner (the VP of Engineering) called for a complete rewrite of the site.  As we began that exercise, the question was raised, how do we prioritize these new features?  Which begs the question, what on earth are people doing on the site now?

Friendster had the advantage that most small startups don't have: traffic.  Blogs everywhere extoll the virtues of A/B testing.  Which ad is better?  Just run them both and measure!  That's great if you're at Google, or you're running a $10,000 ad campaign, but here's the uncomfortable truth for most of you reading this:

You do not have enough traffic to A/B test

Statistical significance happens when you have scores of people signing up for your product.  In the very beginning, you're going to have to rely on something else.  Direct interaction with people.  You can run web surveys, you can run focus groups.  You can do user testing in a lab.  At my next startup, WisdomArk, we did exactly that.  The only trouble there is:

The plural of anecdote is not data

"One of the users in the focus group said that they'd pay a lot of money for our service!"  You'll find that people say they'll do things that they won't follow through on (run a consumer product long enough, and you get a very dim view of the human attention span).  And you'll find that the very process of finding people who are willing to give you feedback colors the results you get.

Regardless of the scale of use you have, by all means use the data you have.  You have 100 active users?  Talk to them directly!  Narrow in on a couple of key questions.  I'll hazard that this is true:

The seed to success is within your first 100 active users

Rarely do companies get it right on the first try with their product, but the really rewarding problem to solve is probably known by one of your early users, who came to your product in the hope of solving it. 

When I tried recruiting people in 2003, engineers would say, "People tried this before, what changed that's making it work now?"  It took me a long time to figure that out.  I created a visualization of our traffic that made it apparent.  I took all the data for a specific day of traffic, and for each page on the service, I drew a circle, sized proportionally to the amount of hits it had.  Then I drew arrows between the circles, showing where people went from one page to another.  The thickness of the arrow matched the amount of traffic.  When I laid it out, it was apparent what people were doing.  They came to the site, went to their friends' pages, and then looked at all their pictures.  People were coming to the site, to look at pictures of their friends.

At the time, we allowed people to have 3 pictures on the site.  I went to the CEO, and said, "make it 6".  Immediately, our traffic went up.  So we raised it to 12, and then finally made it unlimited.   (For those of you who don't understand why we didn't just go directly to unlimited, remember at the time that even eBay didn't let people store photos--the cost of disk storage was just starting to plummet, and it was becoming economically viable for the first time.  Myspace didn't even do their own photo storage, they dealt with PhotoBucket.)  There's another key lesson here.  Figure out a cheap and fast way to figure out if your hypothesis has merit.  If it does, keep doubling down:

Find a small test, and if it works, iterate until it doesn't

I then did some digging around, and realized why Friendster (and more broadly, social networking) took off in 2003.  2003 was the first year that digital cameras outsold analog cameras.  Social Networking is a phenomenon triggered by a new technology: cheap digital cameras.  How much of Facebook's traffic today still centers around photos?  (As an aside, this also was an omen for the then coming irrelevance of Friendster: photo pages do not monetize well.)

This was all important because we always felt that the people search was the most important feature we had, and it's the one we spent most of our resources on--it's the feature our brightest engineers worked on.  You could search Friendster for "Women, between the ages of 25 and 30, within 2 degrees of me, and within 25 miles of my house".  That's great when you're a dating site, but what if people are really just at your site to ogle all the pretty people, and see the silly pictures their friends had put up?  Did we blow our opportunity by working on the wrong problem?  (I have my own argument about that: Friendster succeeded, but with the wrong audience, but that's a story for another time.)

The whole exercise was a positive example of Friendster working well as a company.
1. Ask the right question.  "What feature is really drawing people to the site?"
2. Get the data.  "Time to parse the logs."
3. Synthesize the data.  "How do I understand this?  Can I turn it into a single diagram?"
4. Draw the conclusion.  "People really come here to look at photos."
5. Make the change.  "Let's offer 6 photos."
6. Measure and repeat.  "Hey, that boosted traffic, let's offer 12!"

Get your credit score for free? Mint money? Get a personal loan?

 

What do all of these questions have in common? Well, I spent yesterday attending the Finovate conference finding that out. The conference has a similar format to DEMO but for the ‘financial technology’ space. Overall, 30+ companies presented for 7 minutes each. While there were more ‘features’ than companies there were some intriguing services. Interestingly, there was a void of companies in the B2B payment space and vertical payments space.

In case you're interested, here were the highlights:
-There were 3 types of companies that presented at the conference (personal finance, p2p lending and lead gen variants).
-There were several personal finance companies that have sprung up over the last year after Mint received so much publicity. Other companies in the space include Wesabe, Rudder and Green Sherpa. Basically, this space seems overly crowded at this point. Mint says they are signing up 25k users per week purely on word of mouth. Will it last without some major distribution partnership?
-The P2P lending space (Lending Club, Pertuity Direct, Prosper) is also crowded. The big news at the conference was that Prosper emerged after a six-month quiet period to register their platform with the SEC. Prosper has re-opened to borrowers in all 50 states and lenders in California. Currently, Lending Club and Pertuity Direct are the only two p2p lending platforms available to most investors throughout the United States. Prosper's model is different because, among other things, loan rates are determined by an auction among lenders.
-Lead gen models were prolific amongst this crowd. Interesting companies using this model include Mint, Credit Karma, and BillShrink. The interesting thing  is that these companies have created products that offer value to the user. They have turned the slimy lead gen model on it’s head by thinking about the user first.

The Nugget:
One of the most difficult problems amongst all of these startups continues to be distribution. In a space that isn’t viral and SEO/SEM is very difficult, startups are looking at ways to build traffic. Hence, there were several announcements of white label solutions and integrations with larger players. I doubt most of these companies will be able to build up enough traffic...fast enough...without these traffic partnerships.

Patterns of Startup Success

Over the last decade I've worked with dozens of startups and even participated at the inception of 7 of them from initial concept to funding. So, I've been meaning to capture some of my thoughts on 'patterns of startup success' as I've seen similar tactics that work across companies. Here are a list of items that I plan to write or tap entrepreneurs I've worked with to write about:

1. Have a perspective...vision makes a difference
2. Get to market quickly...that’s your market research...the product might initially be crappy...fail fast
3. The business model is critical...don’t overlook this exercise...but keep it simple
4. Know where you will get distribution, distribution, distribution
5. Marketing on the cheap...market to passionate users...build loyalty through community management
6. Hire carefully...keep it lean
7. Iterate quickly...customer development and data driven development
8. Focus initial product on instant gratification and frictionless first-time usage
9. There are always trolls...listen to your heavy users
10. Forget the 60 page business plan...develop a pitch...tell a story...10 slides...20 min long
11. Achieve key milestones that reduce risk before seeking funding

Calling the series 'Patterns of Startup Success' is a bit of a misnomer because a lot of the posts will be about missteps that led to an understanding of what works. Also, even knowing these patterns doesn't mean you won't make the same mistakes in the future. Building a startup is inherently difficult and you're bound to make mistakes but having experience hopefully reduces the chance of missteps.

I'd be interested in hearing from you if you'd like me to touch on any particular topic. I'll try my best to accommodate requests if I feel like I can do the subject justice.   

Big Company Innovation needs to Look Externally

There has been a tremendous amount written about 'How to Innovate at Big Companies' but internal development of successful web services are hard to come by. Take Yahoo! for example. Over the last decade, what was the last successful consumer service developed internally? You might say, Flickr or Yahoo! Groups but you'd be wrong. Both of these services came from acquisitions. In fact, the only internally developed service that can be considered successful in my mind is Yahoo! Answers.

This reminds me of an article about Yahoo and product innovation where Caterina Fake, co-founder of Flickr said:

“There are tons of amazing ideas in big companies, and no innovation deficit. But the obstacle to getting things built is mostly process. There is one kind of process developed for building and maintaining large-scale products… And the development processes for that are very different from what it takes to build a new product in a short amount of time.”

Caterina tried to foster the start-up process in an internal incubation lab called the Brickhouse which was recently shut down because it just didn't work. Internal innovation labs are tough because they can never really foster a meaningful lineup of entrepreneurs which I believe to be 'the' critical success factor in a start-up. Additionally, it's very difficult to provide a lot of 'skin in the game' for internal entrepreneurs. So, for the many big companies that are unable to create a culture of innovation and unable to adopt new processes, perhaps they should focus most of their innovation energy outside their four walls by fostering/investing in external development.

There are many ways to do external innovation and I'd be interested in what you've seen as successful and not so successful practices.

Legal Nirvana for Startups

If you're like most entrepreneurs, dealing with legal firms for issues like company formation, financing and business contracts is like visiting your dentist. You cringe just thinking about it because you know it's going to hurt. I'm talking about your pocketbook. Hiring big law firms these days will cost you $400+ per hour and you're often handed a junior attorney that's learning the ropes. But entrepreneurs in the know have found great alternatives in a new breed of firms like Paragon Legal.

Paragon provides startups with senior level attorneys (10+ years experience) who have worked at the best law firms in the country at 50%+ less than the average associate from a large law firm. How do they do it? Paragon has striped out the overhead. Instead of operating under the traditional law-firm pyramid, they have removed the bulk of the costs associated with that management structure. This allows them to offer services at greatly reduced rates but still attract the best attorneys. Like many internet businesses they are just stripping out most of the middleman costs.

For most startups looking at cutting every cost possible, you'll want to look at Paragon.

*For full-disclosure I have to admit that my SO is the managing attorney at Paragon..but this has only made me more aware of the issue and alternatives;)

The #1 Under-served Audience on the Web?

 With all the services being launched on the Web today you'd think every segment has been saturated. But  one segment has been grossly under-served; namely the aging population. For years the discussion has always centered around the fact that the older population did not use the web and avoided technology. Things are changing rapidly.

According to the TNS Compete and the Consumer Electronics Association joint study, Greying Gadgets: How Older Americans Shop for and Use Consumer Electronics, the age segments of 50-somethings, 60-somethings and 70+ use many technologies at or near comparable rates as younger age segments.
For example, eighty percent of 60-somethings used a cell phone in the past week, nearly equal the usage rates of 18-34 year olds. Additionally, 71 percent of 60-somethings and 52 percent of 70-somethings used a search engine in the past week, compared to 77 percent of 18-34 year olds.

The Older Segment is a Growing Market

Not only is the Older segment becoming more inclined to technology but it's also growing. Over the next 10 years, the Older Americans segment (50+) is projected to grow 23%. By 2020, this market segment should be nearing 119 million people and represent 35% of the total population, as provided by the US Census.

They are Also a Growing Market On the Web

Here are some basic stats from the Greying Gadgets report that tell the story:

• 78% of 50-54 year olds are online
• 45% of 70-75 year olds are online
• 67% of 70-somethings use a cell phone on a weekly basis
• Older Americans are 27% more likely to visit travel websites than the average internet user, and 98% more likely to visit health sites
• 77% of 50-somethings, 71% percent of 60-somethings and 52% of 70-somethings report having used a search engine in the past week
• 20% of 50-somethings have used a social networking site in the past week
• 24% of those 70+ have watched an online video in the past week

Simplicity Sells...You May Need a Modified Offering

As might be expected, sixty percent of consumers aged 50 and older indicated that a product having too many features was a main reason for being frustrated with technology, compared to 39 percent of consumers aged 18-49. So, if you're creating a solution for this segment you'll need to keep this in mind.

So, if you're looking for that killer audience...you might just want to think Older;)

Other interesting articles

The Older Audience Is Looking Better Than Ever, New York Times, 4/19/2009

Demos that Stick!

You're looking to change the world. But, you need to get people to take notice and inspire them into telling others. Whether you're talking at TechCrunch, DEMO, to investors or customers you need a demo that sticks.

If you're looking for basic tactics behind great demos you'll want to read GuyKawasaki's blog, 'How to be A Demo God' or Michael Arrington's TechCrunch post, 'How to Demo your Startup'.

But a great demo is more than just tactics, it's an artform that when done well leaves your audience in awe. Don't get me wrong, tactics are incredibly important. In fact, one of the most important tactics is often screwed up. Namely, people do the demo last when it should be done upfront. Jeff Bonforte outlines some great thoughts on the necessities of doing a demo upfront on his blog post, 'Boom, Demo First'.

But to really reach your audience, your idea needs to stick. With this in mind, I'd suggest taking a read of Made to Stick because the same principles Chip and Dan Heath discuss apply to demoing your product. Here are their 6 principles:

PRINCIPLE 1: SIMPLICITY - How do you find the essential core of your ideas?

“It’s hard to make ideas stick in a noisy, unpredictable, chaotic environment. If we’re to succeed, the first step is this: Be simple. Not simple in terms of ‘dumbing down’ or ’sound bites.’ What we mean by ’simple’ is finding the core of the idea. ‘Finding the core’ means stripping an idea down to its most critical essence.” (pgs. 27, 28)

PRINCIPLE 2: UNEXPECTEDNESS - How do you get our audience to pay attention to our ideas, and how do you maintain their interest when you need time to get the ideas across?

“The most basic way to get someone’s attention is this: Break a pattern. Humans adapt incredibly quickly to consistent patterns. Figure out what is counterintuitive about the message-i.e., What are the unexpected implications of your core message? Communicate your message in a way that breaks your audiences’ guessing machines.” (pgs. 64, 72)

PRINCIPLE 3: CONCRETENESS - How do you make your ideas clear?  

“Abstraction makes it harder to understand an idea and to remember it. It also makes it harder to coordinate our activities with others, who may interpret the abstraction in very different ways. Concreteness helps us avoid these problems.” (pg. 100)

PRINCIPLE 4: CREDIBILITY - How do you make people believe your ideas?

“How do we get people to believe our ideas? We’ve got to find a source of credibility to draw on. A person’s knowledge of details is often a good proxy for her expertise. Think of how a history buff can quickly establish her credibility by telling an interesting Civil War anecdote. But concrete details don’t just lend credibility to the authorities who provide them; they lend credibility to the idea itself.” (pgs. 138, 163)

PRINCIPLE 5: EMOTIONS - How do you get people to care about your ideas?

“How can we make people care about our ideas? We get them to take off their Analytical Hats. We create empathy for specific individuals. We show how our ideas are associated with things that people already care about. We appeal to their self-interest, but we also appeal to their identities-not only to the people they are right now but also to the people they would like to be.” (pg. 203)

PRINCIPLE 6: STORIES - How do you get people to act on your ideas?

“A story is powerful because it provides the context missing from abstract prose. This is the role that stories play-putting knowledge into a framework that is more lifelike, more true to our day-to-day existence. Stories are almost always CONCRETE. Most of them have EMOTIONAL and UNEXPECTED elements. The hardest part of using stories effectively is make sure they’re SIMPLE-that they reflect your core message. It’s not enough to tell a great story; the story has to reflect your agenda.” (pgs. 214, 237)

So, now all you have to do is pull all of these principles together;) If you're looking for someone who has mastered the skill look at some of the demo's by Steve Jobs...a master in my mind. Here is his demo of the IPhone (IPod capabilities) at MacWorld. Notice how he captures his audience by sticking to most of these principles.

Ad Based Startups -- Bad Idea?

CPM rates have been falling. That's no surprise. What's been a surprise is the depth of the decline for remnant inventory.

An August study from the Interactive Advertising Bureau and Bain Capital found on average, CPMs on remnant ads ranged from $0.60 to $1.10. In my own recent informal poll, CPMs seemed to have cratered. Untargeted, non-search CPMs are around $0.10-0.15 and targeted ads are around $0.15-0.25.

If you're a startup focused on ad based revenue streams this is shocking news. A quick back of the envelope says it all:

Suppose you have a userbase of 1 mil users that visit your site 5 times a month and view 10 pages per visit. Even if a publisher were to sell 100% of it's inventory at $0.15 CPM the publisher would earn $90k per year in revenue. Maybe enough to pay 1 FTE.

What Factors Determine Your Adverstising CPM Rates?

Andrew Chen wrote a great blog post about 5 Factors that Determine your Advertising CPM Rates...here is the big takeaway:

Sites that are easier to monetize have high click thru rates because people are in a "transactional mode". Characteristics of these sites include an evergreen userbase, focus on a particular category, US based and significant traffic through SEO.

What about More Targeted Ads with Unique Formats?

Remnant inventory has been hit hard but have advertisers fled to quality? The IAB report suggests that better targeted ads have not been as hard hit as remnant ads. This means high quality vertical ad networks like Martini Media (which is an affluent based ad network) may be where some of the flight to quality action ends up. If you're a publisher you'll want to look at vertical ad networks that align with your userbase...they could provide a significant lift.

Ad Based Startup?

With the supply of remnant inventory so high it's hard to see rates increasing in the foreseeable future. So, unless you have a quality userbase that fit's Andrew Chen's mold you really need to look at alternative business models. In a nutshell, most ad based startups are bad ideas in today's market.