peHUB » Correlation Ventures Closes $165M Fund That Will Use Predictive Analytics
I don’t get this.
(via jryu)
So these guys contacted me last year and I was completely sketched out. This is as close to “dumb money” as a VC could ever be. They’re a bunch of Wall St quants who think they’ve got a model that “proves” an investment will be successful or not. Frankly, entrepreneurs should demand more from their investors than cash. Their oversubscription, in my opinion, speaks more to the stock market’s volatility and asset managers’ desperation to find alternative asset classes with attractive returns.
(via caterpillarcowboy)
When the VC fund’s name belies the fact that its investment thesis isn’t causal but correlated, and it still raises a boatload of money, it signals that investors are willfully grabbing at straws. If I pick my nose and Facebook goes up on Second Market, can I sell you guys my future boogers?
(via caterpillarcowboy)
Nyquil: Organic Seasonality
I’m sick.
I’ve been coughing and hacking up a lung since New Year’s Eve, and it’s finally starting to break. There’s a bug going around, and I caught this one from Hannah’s father (we’ve traveled a bunch over the holidays, it’s taken its toll on me I guess).
Though rarely sick, when I am, I sleep awfully and I’m a giant grump - total pain. I sought refuge with people who could commiserate with my malaise, and found the above Nyquil mentions via Twitter Search.
Now, I’m not a Nyquil user, but I found this fascinating. Yes, tis the season to be merry, but it’s also the season to be sick as a dog. Also, nighttime is when people feel the insomniac ramifications of the flu bug. The amazing thing is that, unlike most Twitter trending topics that are common event based, this one trended due to its organic seasonality:
- height of flu bug
- during the night
- before the first day back to work after the holidays
- similar treatment via America’s preferred legal knockout drug
The organic building of ephemeral community among individual-yet-similar experiences vs. shared consumption of popular events indicates how deep Twitter’s behavioral rabbit hole goes (and how much value they have yet to extract). Nyquil couldn’t have pulled this off if they tried, although, the idea of over-the-counter medicines advertising “hey, if you’re sick, tell us!” may be the next genius cold remedy marketing campaign - heard it here first. Or alternately, Cambell’s will personally deliver chicken soup to your home if you tell them how sick you are.
2012: the year of preying off the weak and sickly.
Paying for pixels | The Economist (via ninakix)
Oh The Economist, there’s a stark difference between a sales tax and an income tax. In a virtual world, where it’s completely legal to fabricate Ponzi schemes (= freemium around virtual goods + false scarcity + artificial urgency), I wouldn’t be too quick to levy an income tax on a system where most people lose in the gentle dance of joy of gain and fear of loss. Game companies need not go bust for players to claim virtual losses against real taxable income. It could work similar to gambling or trading, which is a messy prospect - you can’t have it both ways governments. Maybe they’ll treat all activities in the virtual realm as unrealized income/paper gains and only tax people when they wish to convert to real currencies. However, the time horizon on your virtual good ownership may then come into play, creating short and long term virtual gains tax rates (where long term virtual gains tax rates are lower because they represent a less speculative interest in the underlying gaming economy). Oh, what a glorious mess.
“Bitcoin, take me away from all this madness!”
(via ninakix)
Quick Thoughts on the Upcoming Zynga IPO
As the world eagerly awaits the latest social web IPO with Zynga, many have focused on the event’s significance as evidence of Facebook’s platform power. However, Zynga’s rapid ascendency marks two milestones that I believe are much more significant.
For one, Zynga has long been the largest Amazon Web Services customer, with an estimated tens of thousands of instances spanning every global AWS region (arguably, Zynga financed the expansion of AWS’ footprint). Yes, Zynga is planning a move over to their own co-located machines as there are gigantic cost savings to be realized, but they are the first company who scaled to IPO 100% within the cloud. Indeed, if we are to compare the platform power of Facebook and Amazon, remember that Facebook only introduced their 30% “platform tax” on transactions in 2010, while Amazon got paid for every byte sent and CPU cycle logged from day zero. Bezos is #winning.
Further, Zynga is the first IPO based upon a pure transaction-based freemium model, creating a social engagement funnel where ~95% of customers are free-riding and subsidized by the rabid virtual commerce activity of the remaining ~5%. How the market treats the merits of free over the long haul has yet to be seen, but there’s a wide array of startups with IPO potential leveraging Zynga’s “freemium in the cloud” playbook.
The Vision Behind Klout (Needs Lasik)
Unlike some, I see nothing nefarious about Klout, and they do offer a means to opt out (disclosure: I’ve received many Klout perks and I do post/tweet about them). We are a people who choose to live in public, and as such, our activities can be aggregated, analyzed, ranked and rewarded accordingly. However, like others, I do not think that score and B.F. Skinner-conditioned-response-informed intermittent perks are value enough to offset the creepy stalker feeling most get from public measurement.
With any analytic/measurement, however grandiose, one must ask the compelling question “now what?”, which Klout has really only answered for target-hungry marketers and not for the measured masses by way of value expressed via transparency, guidance, discovery etc.
In comparison, PageRank delivers value across Google’s multi-sided platform; the searcher seeking information, the underlying measured masses of public web site inventory by way of earned engagement, and the marketer looking for paid engagement. Symbiotic.
What does Klout’s PeopleRank offer me, the measured inventory? In some ways, it’s as if Klout got it backwards by monetizing intent rather than promotion against intent. Ergo, like Google, hide the rankings, let people search (intent) for niche influencers (earned), and push marketers to pay to promote their thought leaders (paid). Seems like a cleaner model in my mind.
I hope my opinions here don’t stop the marketing swag from appearing on my doorstep. I’m conditioned to respond to freebies! ;-)
Some Thoughts About A/B Testing
At lunch today, Andrew reminded me of a story that I had long forgotten.
When he was at USV, he attended a board mtg for a portfolio company that we have in common.
It was the early days of that company and their initial product hadn’t been out that long.
The topic of A/B testing came up and Andrew reminded me that I came out against A/B testing in that meeting.
The sad but true part of todays conversation is that I don’t recall that meeting or that particular conversation.
But I would say the same thing to any early stage company with their initial product.
When you have thousands or even tens of thousands of active users, the product comes from the founders vision and desire. It doesn’t come from a focus group or some survey. And it didn’t come from an a/b test of some landing page or the home page.
It comes from the heart.
a/b testing in the early days is never going to be as good as pursuing the founders vision for making the product everything that it’s supposed to be. Fix those annoying bugs. Make it faster. Make it easier. Make it right.
Now, that doesn’t mean you shouldn’t listen to your users. Their feedback is vital. We all know, as a perfect example, that may of Twitter’s best features came from their users (e.g @mentions and retweets).
But a/b testing is optimizing around the edges around a terribly small group in the early days. It’s not worth the energy in my opinion.
As a web service grows significantly, a/b testing goes from not helpful to critical. Analytics, a/b testing, linking new features to metric driven objectives become an essential tool in the arsenal.
gbattle sez:
I respectfully disagree with you here Bijan, but it may be more in method than spirit.
For one, A/B or split testing is just that, a test of a hypothesis, a scientific assessment of the founder’s ability to express his or her vision through the lens of customer behavior. It is not a focus group. It is not a survey. It is not a conversation. You are watching what people do and quantifying it, not listening to what they say.
Split testing provides data that supports or refutes a hypothesis via the purposeful segmentation of customer experience to generate data. What you do with that data is another matter entirely. One can optimize or overfit to a split tested model just as readily as one can become entrenched in the myopic hubris of his or her heart and gut. The real questions are, how informed was the founder’s final decision and, over time, what is the founder’s predictive capacity with respect to his or her heart/gut/instinct? It is in these two questions that data becomes a key factor, but not an exclusive one.
Maybe I’m missing something here, but I don’t understand how one can be an advocate for listening to early stage qualitative user feedback (comments, suggestions, “get out of the office”, etc.) but not an advocate for tracking and testing those qualitative outcomes for quantitative feedback to fuel the iteration of product, exactly what A/B testing was built to help accomplish.
If I replaced “A/B testing” with the term “analytics” or just “gathering data” in your post, I’m not sure you’d feel as confident about devaluing their importance for early stage startups. The scoreboard doesn’t lie, and founders must choose which metrics to listen to and which to ignore at any stage. Making data collection and data analysis part of your DNA early on isn’t a waste of time. I’d argue it’s easier to form that culture more earlier than later.
I get that you want early stage founders to avoid analysis paralysis, blind optimization and wasting precious cycles and resources with little return value, but I think we’d agree these afflictions can apply to any company independent of growth phase.
Anatomy of a Greg Battle E-Introduction
Over the years, I’ve crafted a style of introducing people via email that I nearly always get compliments on. If you’ve been unlucky enough to receive one, you’ll note that I always follow a similar format whereby I don’t have to think about structure or context and can focus purely on altering the fun parts. I’d love to hear how other people approach this most common e-activity.
Subject: Introduction: Jane Doe <===> John Smith
I think it’s important to get both names in the subject as it makes finding the email that much easier. The arrow symbolizes the connection activity. Straightforward, transparent and to the point.
Hi folks. Jane, this is John. John, meet Jane.
My opening line is the same as I’d use in person verbally. Now the fun part begins.
Jane is a blah blah blah …
John is a blah blah blah …
The creative magic happens in the above area. Using no more than three sentences each, I summarize the relevant aspect of each person’s background, beginning in the most superlative fanboy fashion possible, emphasizing perceived excellence, but ending with a small humorous twist at each one’s expense. The point is to personalize the bios via authentic admiration and humanize them via wry wit. Sell the value of the connection with flair.
When John mentioned blah blah, I immediately thought of Jane’s expertise in blah blah.
A singular sentence anchors why the two should meet now. Solidify the urgency.
Play nice everyone. If any deals comes of this intro, my standard rates apply: don’t let my kid starve.
Best,
Greg
Finish strong and remind everyone that reciprocity matters.
So, how do you handle introduction emails?
![This is a photo of the message I received from David Karp regarding my offer to help him learn vim** and a joke I made about his being photographed taking photographs without his beloved Leica M9 (if you have to ask what it is, you can’t afford such photog swag). The point here is that for all the complaints and suggestions that I know David gets all day every day regarding Tumblr, he’s taken extra care to inject delight into any one-on-one message interaction with Tumblr staff by making a bold statement: “we care, and you’re special.” Lesson being, listen to everything even if you heed no one, but just make it such that when you speak, they damn well stand up and take notice.
[insert tip of hat here]
**You know, a vim editor or better yet vim controlled navigation of posts might be … oh, never mind.](http://29.media.tumblr.com/tumblr_lpn3s7jpXz1qzp5buo1_400.png)
This is a photo of the message I received from David Karp regarding my offer to help him learn vim** and a joke I made about his being photographed taking photographs without his beloved Leica M9 (if you have to ask what it is, you can’t afford such photog swag). The point here is that for all the complaints and suggestions that I know David gets all day every day regarding Tumblr, he’s taken extra care to inject delight into any one-on-one message interaction with Tumblr staff by making a bold statement: “we care, and you’re special.” Lesson being, listen to everything even if you heed no one, but just make it such that when you speak, they damn well stand up and take notice.
[insert tip of hat here]
**You know, a vim editor or better yet vim controlled navigation of posts might be … oh, never mind.
Turntable.fm Revenue Model: Jukebox Hero
(Excuse typos, I’m on my phone on the BoltBus.)
So, you’ve tried Turntable.fm, the new social streaming music service by Billy Chasen & his crew of evil genius elves. Visual, sonic, interactive and synchronous, Turntable.fm runs contrary to the near ubiquitous on-demand autonomous media consumption model. Your inner adolescent screams “this is freakin’ awesome,” but the quiet adult voice starts wondering, amidst all the DMCA/RIAA hornet’s nest, “how the hell will these guys make any money?”
The most obvious (read as boring) path would be simple virtual goods around avatar customization. Bling in 2011 = meh. YAMLM = yet-another-music-locker model? Subscription for a human powered Pandora? Ninja, please. We can do so much better.
Let’s start by analyzing the existing behaviors around DJs. Much has been said about both the polite self-policed/organized DJ queues along with the complete free-for-all quickest finger in the West invective laden blood lust to get a coveted DJ spot. All manners and ethics notwithstanding, it’s clear people want to DJ given the scarcity of available spots and urgency of synchronous enjoyment. There should be both an earned and paid method to do so, and history has already provided us with a peek into the answer.
The lowly jukebox. Make one DJ seat a permanent jukebox seat, available for a one-time purchase between each of the five DJs. All jukebox plays are subject to the same ratings as regular DJs except being kicked.
The jukebox encapsulates the virtues of basic queuing theory - your paid request will be handled in the order in which it was received. Not bad, but we can do even better by making the queue dynamic via auction similar to AdWords, where price then timing determine DJ order.
So, what have we just done here? We have allowed the service to discover each user’s realtime reservation price for playing the next song. Turntable.fm can make sure this price is above the online radio streaming royalty expense and pocket the difference to cover expenses and profit (smells like freemium … because it is). Users can purchase spin-bucks in bulk, can gift them, and can earn them via their reputation. DJ score and spin-bucks could be one and the same or separate, maybe it’s via BitCoin, but these are implementation details.
If you are old enough to remember The Box, which was a televised music video version of the jukebox, you can see the promotional opportunities here for major and indie artists. And if you’re old enough to remember the pay-to-play club days of LA’s Sunset Strip in the 80s, you’ll also see the parallels of paid self-promotion. Imagine the type of auctions you could use to promote your band or DJ taste function (even when offline): I will pay up to 20 spin-bucks to play my band’s cover of Michael Jackson’s “Beat It” in any room right after somebody plays “Thriller” but only when the room has over 20 users. Cuh-ching.
By turning Turntable.fm into a transactional model with dynamic price discovery, it can benefit from the intersection of scarcity, urgency and latent narcissism.
Back to my bus ride … G-Mobile.
Tracking Tumblr Dashboard Activity: “No Can Do”
(FYI: If you embed Javascript in an anchor HREF on Tumblr, it’ll work on the front-end of your site but they attach “denied:” for display on the dashboard. That distinction makes no sense to me. I guess someone at Tumblr is worried about someone running a bookmarklet or something that ruins their pretty, pretty dashboard. All I can ask is that whoever makes “missing e” adds in a script to rip that “denied:” out so that we can alert-box ourselves until we’re blue in the face!)
gbattle sez:
Any and all efforts to track Tumblr dashboard activity will be thwarted - Javascript, images/pixels, etc. Even look at their YouTube & Vimeo implementation, the embed is on-demand only. Though I was initially annoyed by this, I admire Tumblr for being consistent regarding dashboard tracking and recognize that if they ever want to explore monetization strategies, owning this core inventory will be essential. It took a great deal of foresight (and I’d imagine convincing of investors) to see this value from the very beginning. Kudos.
Ten Crack Commandments: Notoriously B.I.G. Entrepreneurship Handbook
After speaking with @aweissman and @EghosaO about the tao of Christopher Wallace a while ago, I decided to put together a little end of the year post regarding how to run your digital social startup business like Biggie Smalls and become notoriously big:
- Never let no one know how much dough you hold. The cheddar breed jealousy. Greed and hubris have turned many a king into a jester. As the head of your business, know that the single greatest determinant on the ceiling for employee salaries is your own salary. Hence, keep the amount low, align your incentives with the business via founder’s equity, use non-dollar ways to reward yourself and staff. Stay modest. Stay hungry. Your employees already believe they can do your job better than you can, so why fuel the fire by flaunting the spoils of foundership?
- Never let ‘em know your next move. Bad boys move in silence or violence. Though the public markets demand guidance and penalize surprises, know that while your business is private, zero guidance and surprise are the strongest tools in your arsenal. Projecting your strategy before you can defend it, let alone capitalize upon it, can undermine you. Case in point, when Facebook launches features like Global Like or aggressive privacy policy changes, they don’t do a press release about what they’re going to do for public opinion, they just do it and go as live and as wide as possible with overwhelming force. The biggest crimes/victories are committed/won in public without warning.
- Never trust nobody. Your moms will set that ass up. Common wisdom for certain (apologies to moms everywhere). Though success favors the bold and vigilant, survival favors the paranoid and fearful. The corollary to this is that the person you should trust the least is yourself- self-absorption, self-delusion, self-destruction - which leads us to the next commandment.
- Never get high on your own supply. I’ll defer to Andrew Weissman’s great take on this. Beware the bubble of your own hype, but further …
- Never sell no crack where you rest at. I don’t care if they want an ounce; tell them bounce. Street pharmaceuticals notwithstanding, if you’re selling vapor goods and services to customers, you can’t also sell vapor to your employees. As aggressive as you may market to the outside world is how authentic you should be with your inner circle. The fake-it-til-you-make-it approach cannot be your go-to-market and corporate culture strategy. And if you bring that BS sensibility into your household to manage expectations in your personal relationships, you’ll soon not have a home at all. You can’t get everyone high on your own supply all the time. Only Steve Jobs has such reality distortion field superpowers.
- That G— damned credit, forget it. Think a crackhead paying you back? In all social circles, there’s earned (non-dollar group-experience-enhancing activity) vs. paid (cash) engagement and the return on that social investment (authority, prominence, digital goods, share of voice, scale of distribution, premium features, etc.). As an experience provider, your job is to make sure that user engagement is matched with commensurate immediate return. When that ROI is neither commensurate nor immediate, you are asking users to act on credit and give without getting. And if you think reversing the equation by giving more experience than what your service gets in return will work, forget that too. Instant gratification economy is short on memory and favors and long on distractions from shiny new things.
- Keep your family and business separated. Money and blood don’t mix. Working with family clouds your better judgement. No one should feel so comfortable with you that their seat at the table can’t be upgraded. Normal workplace politics combined with normal familial dysfunction equals a hot tangled mess of nepotism and jealousy. If you must work with family, make sure they aren’t a direct report to you, and make it clear that their manager has full license to terminate them.
- Never keep weight on you. Them cats who squeeze your guns can hold jobs too. If you are going to build a fortune upon a crime, be smart enough to have underlings carry out your dirty work, preferably ones who would do it for free a.k.a. your customers (Twitter, Facebook, YouTube, Flickr, et. al.). DMCA is your best partner in crime … for now at least.
- If you ain’t getting bags, stay the f—k from police. Snitches may get stitches in the hood, but snitches get grand jury investigations in the business world. In legal matters, it’s not about the truth but the position you can defend, so financial liquidity often defines your success rate. Don’t engage in legal wars without a warchest of cash and time, unless, of course, the expected value of the legal wrangling far surpasses that of focusing on your actual business. Unlikely.
- A strong word called consignment, strictly for live men, not for freshmen. Be careful when the revenue model becomes too complex and riddled with contingencies, exceptions and caveats built upon unclear assumptions. Risk isn’t a single variable, it’s multi-factored and multidimensional. Who benefits most from the risks your business is shouldering? If there’s any doubt (and there should be, see #3), remove a layer or seven from your multi-paged whiz-a-bang i-banker approved spreadsheet, reassess the risks, repeat. The basic math behind your revenue model should probably fit on one index card, max.
Google Claims Google Instant Has No Brand Bias
I wrote back in July that Google Suggest appeared to have a branded commerce bias. This predated the release of Google Instant by a few months and the ensuing blogger fervor around brand biased results. Though I believe Amit Singhal’s claim that the results are purely mathematical models and not paid ad placement, I believe he is only telling half the story.**
Here’s the dirty little secret regarding Google that answers the “why” behind the brand bias. There is a gigantic unheralded use case for Google and you need only watch mainstream people use Google for a day or two. When you treat Google as your homepage, you treat Google’s search box as your browser location bar. Hence, the search box is not necessarily capturing intent as much as it is recalling bookmarks you never created.
I cannot count the number of times I have seen the pattern of people typing “eBay” into Google and then clicking on the http://ebay.com top result to visit eBay rather than simply typing the URL into the location bar. Yes, this seems back-assward and inefficient to the browser literate, but when you consider the utility and ease of auto-suggest/complete to every site you want to visit - no CTRL-D or local bookmarks or Delicious add-on or Quicksilver - all available from one simple box, the so-called lames may have found the fastest and easiest way for ubiquitous access to their favorite sites. The top result for typing “e” in Google Instant is “eBay” due to the above use case and not the aggregation of a bunch of intent based searches like “where to sell my Teletubbies”.
Further, I submit that my explanation for the organic Google Instant/Suggest results in turn drives more SEM revenue into Google that they don’t need to monetize Google Instant directly as it is already monetized indirectly. If my use case moves the needle in Instant, it definitely moves their top line revenue significantly - simply guess the line item SEM budgets of the brands that come up in the results.
** The branded results bias is also something that can be mathematically evidenced using simple statistics. The mathematical/algorithmic defense is hogwash. I’m not implying that the results were engineered, but that the brand bias evidence is rampant.
gbattle sez:
Is there a way to rotate the gaming mechanics of say Cityville, Zynga’s newest release with the most advanced/adaptive social gaming mechanics to date, to solve problems in education, healthcare, unemployment, national security & poverty? To the extent a company can create “something more meaningful” within the interactive entertainment industry (and yes, Groupon is entertainment for the web as much as HSN is entertainment for television), this may happen, though pure entertainment plays will always be the bigger and faster wins.
