Zelkova Ventures

In our ongoing effort to provide more transparency in the investing process, I’ve charted an analysis of our dollars deployed per month over the course of a few years.

% of Dollars Deployed by Month

As you see there are peaks and valleys in the deployment of capital and I account the cyclical nature of this to a few things:

• As a small team we typically can only handle so much at once, so when we’re in the process of closing several deals it reduces our time to look at new deals, which creates the pattern.

• The lift in November and December is due to the use of ‘year-end’ as a catalyst to close deals. Specifically in 2010, there was uncertainty as to whether the tax benefits for investing would be extended, so there was a rush to close any deals in the pipeline.

As a company looking for investments, the chart’s peaks and valleys represent opportune times to pitch. Since it generally takes us between 30-60 days to close a deal, the best period to pitch would be during the valleys, typically January, April, June and October as we are most likely to have the bandwidth to provide the needed attention.

Last week I attended a webinar on “How to Drive Marketing and Business Strategy With Social Intelligence,” hosted by Zach Hofter-Shall of Forrester  and Melyssa Plunkett-Gomez. Zach’s presentation covered an overview of social intelligence, challenges of social media and in-house verse outsourcing listening platforms and Melyssa discussed how social intelligence is a massively underutilized opportunity for brands. Melyssa works at Crimson Hexagon, a Zelkova Portfolio Company, which is a leading social intelligence business powered by patent-pending technology developed at Harvard University.

One of Zach’s main discussions revolved around how when properly implemented, social intelligence (as defined by finding information online and well, doing something with it) has the power to drive a company’s marketing and business strategies. While many believe social media is made up of different social networks, Zach disagrees, claiming the internet is becoming increasingly social in nature and social media “is a bit of everything” online. To understand his claim, let’s use his example of The New York Times’ homepage – it’s not just an online version of the publication, but rather a hub of articles, blogs, comments and other interactive components that make it social, yet it isn’t viewed as a traditional social media (such as Facebook or Twitter). This brings up the conversation of whether we can ever be passive readers online, or if we automatically become participants engaging with one another as soon as we open the browser?

There are different levels of internet users which Zach describes in the form of a ladder: the most influential group are the creators who sit at the top rung followed by the conversationalists, critics, collectors, joiners, spectators and finally, the inactive. Zach predicts the inactive category will eventually become obsolete because going online will assume interactions with others in some capacity.

Deciphering the data from users’ engagement online can be an exceptionally valuable for brands, as long as companies are able to determine which information is useful. In conjunction with the internet becoming intrinsically social, the data derived from social media will increase, making listening platforms even more imperative. To understand the relationship of social media information and listening platforms, compare it to garbage and garbage trucks: with less inactive users and more engagement online, the amount of information available to companies will become overwhelmingly large, resulting in a huge pile of garbage. While there will be valuable data, most will be useless, and it’s up to the listening platforms (the garbage trucks) to help weed out the trash. Brands need to decide the most fitting method of gathering and applying data to their business strategies now before the info becomes even more vast.

Zach’s Recommendations

Step 1: Internal Resources
- Evaluate

- Determine capabilities

Step 2: Listening Platform Data
- Evaluate the data process from the listening platforms

- If use listening platform services, evaluate them

Step 3: In-house or outsource listening
- Either way, partner with a listening platform

Step 4: Assign leadership responsibilities

Step 5: Define data distribution channels

Step 6: Prepare for Action

Recently, a team of entrepreneurs I am working with (The Shared Web) reminded me of a post, (http://cdixon.org/2010/07/22/graphs/), by Chris Dixon where he discusses different types of graphs and how his product Hunch is trying to create the interest graph. Hunch is doing this by discerning user’s interests based on their answers to playful questions. Hunch illustrates that users are more interested in making decisions such as what movie to see or book to read based on the suggestions of people with similar interests, whether or not they are your friends. Dixon’s believes that Twitter’s innovation was to create asymmetric relationships where you can “follow” people (of interest to you) and obtain what they broadcast.

The Shared Web (TSW) takes the concept mentioned above and applies it to show users content categorized in topics that they’re interested in curated by people they trust. The Shared Web is a social news product – think of a Hacker News for every topic or a social Reddit. Users follow topics and TSW shows them content in the areas of interest by people who have similar taste. To do this, The Shared Web pulls content from users’ social networks (starting with Twitter) and from content shared on TSW. The product promises the following:

1. That users will stay informed. i.e. they wont miss the big stories of the day (i.e. you won’t overlook things or miss them because you didn’t log on to Twitter at the right time)

2. Users will able to discover content they are interested in but wouldn’t have found otherwise.

3. Users can quickly engage around content that they find interesting by thanking the person who posted it, reposting it to their own network or having a conversation about it.

They are building the interest graph by looking at what topics each individual sbuscribes to and who they trust (endorse) in these different categories. If a service or set of services succeeds in mapping out this graph, they can use that information to help people relevant information that is important to them and customize their experience in different avenues. There are numerous opportunities to monetize the data, which is quite exciting to us!

They have an interesting strategy that is two fold:

1. They pull content from Twitter and rerank it based on how relevant it is to you (by looking at topics it belongs to) and the number of people that have retweeted it + other heuristics in their secret sauce.

2. They let people share on their own network with the underlying assumption that there are many things that people see everyday that they don’t share because there isn’t a dedicated place to do that without flooding your audience. Think of articles on Cassandra, a key value store database, or startups that you find, or the myriad of blogs posts that you see over the course of a day. A key differentiator to all the social news readers currently out there, is that they let you share knowing you wont flood your audience because they have subscribed to the topics they care about.

The Shared Web is a curated experience of content on the web, you subscribe to topics you are interested in and in return you don’t miss the big stories and discover content that you wouldn’t have otherwise found. The mixture of curation and serendipitous discovery as opposed to search is very exciting. It’s a way of creating that  ’AHA’ moment where you think WOW I would have never found that if it wasn’t for this product. It is a very satisfying, addictive experience. Sharing is the new search and they believe that if you can share better you’ll share more.

Here is an exclusive link that lets you get onto their site – which is still in early beta - to get a sneak peak on what they are working on (link expires soon):

http://thesw.com/hR5kYq

There are really innovative features on the way and your feedback can help informwhat they do next, so I would love to hear your thoughts once you try it out. Get onto to the beta quick, the link expires in a few days.

We have received visitors from 204 sources, here is the breakout of the top ones.  It was interesting to see how few have come from Bing over the past year relative to the market share that has been reported.

Daily Visitor Trends

This graph depicts visitor percentage of each day. I was surprised by the similarity of vistors on Monday, Tuesday, Wednesday & Thursday but not surprised by the amount of visitors Friday & Saturday or Sunday. It makes sense the percentage of visitors would decrease Friday & Saturday but increase on Sunday because these numbers are in conjunction with # of emails received on those days.

Top Traffic Sources

We have received visitors from 204 sources, here is the breakout of the top ones.  It was interesting to see how few have come from Bing over the past year relative to the market share that has been reported.

Top Pages

Great to see the top ages (besides the home page) are those entrepreneurs should be interested in such as the team and our portfolio.

Top Browsers

Today TechCrunch announced that Mark Davis has closed $3m in funding for he and Steve Blood’s new startup Kohort.  Along the funding announcement you can also go claim your username prior to launch, go to http://www.kohort.com .

Generally speaking I have been pretty vocal about my disdain for startups to be in ‘stealth mode’, but Mark was able to convince me why limiting the knowledge out there was important for Kohort and was more beneficial than being public.   The team at Kohort is working on something that can have profound impact on many stakeholders and in the process disrupt many existing players, keeping tight lips on it was important in this case.

While Kohort is in ‘stealth mode’ they didn’t walk around asking people to sign NDA’s or acting all royal, he instead engaged the right people at the right time.

The lesson learned here is that ‘stealth mode’ doesn’t have to mean tell nobody, it’s more have a plan on engaging and telling people what you are up to, but also think carefully if stealth mode is right for you and your company.