by VINOD KHOSLA via TechCrunch
Editor’s note: Legendary investor Vinod Khosla is the founder of Khosla Ventures. You can follow him on Twitter at @vkhosla. All Khosla Ventures investments, as well as ventures related to Vinod Khosla, are italicized.
We are in a whole new world of platforms, a post-PC era, which I’d more aptly describe as the always/everywhere era, finally, and that means a whole new set of opportunities. Add to it the fact that because of a variety of factors too numerous to cover here, the cost of experimentation has gone down dramatically (one can start a web startup or write an Android app with no more than a student credit card!) and raw computing power is taken for granted.
What you get as a result are the recent successes in the Internet/mobile space like Facebook, Twitter, LinkedIn, Zynga, Groupon and others, all of which have reenergized both entrepreneurs and investors. Many of these new startups will be the usual poor clones or feature add-ons to Facebook and Twitter, or poor attempts at doing one feature or another better than Zynga, or applying LinkedIn to a small vertical.
A few will be successful, many will fail, some will be acquired for a piece of technology or for the team (acqui-hires). But that does leave the question: What else new has the potential (nothing is certain!) to be truly disruptive or establish a new category in the domain of consumer Internet/mobile/services (which to me are fast becoming interchangeable)?
The post-PC always/everywhere platform will be defined by the many variations of mobile, always available, silent complementary standbys (like home or personal networks and agents) and more. A new category to me means doing something old differently enough to have it become a large business or have substantial impact among users.
AirBnB and Instagram would be examples of companies whose categories existed prior to their entry, but they are meaningfully different. Likewise, LinkedIn was not the first professional social network but it had substantial new impact and business potential.
Personally, it is hard to see all the areas in which some disruptive or large new segment idea will take off, but it is clear that there are many. So when going fishing for these, I have defined certain pools that are more interesting than others in which to fish. I call them the “unhyped dozen” (to go with my energy investing activities, which I call the “clean dozen”) and am hoping the readers of this post can add another dozen. Treat them as potential fishing ponds rather than predictions. Some of these areas will end up pretty unremarkable and others unmentioned here will surprise us (so, to you entrepreneurs, hang on to your idea I failed to mention).
So here they are, with some examples drawn both from inside and outside of our portfolio to illustrate what I mean:
- Data Reduction or Filters (Siri, Donna, Recorded Future, and many others): “Reducing, filtering and processing data streams to deliver the information or action that is relevant to you.” The web has been expanding what we have access to. It is time for tools (our proxies or agents on the web) to start reducing the amount of information coming at us. News or article feeds, TV shows, YouTube, “must watch” emails — these can all be prioritized for us like our proxy was a virtual assistant who knows our current context and our preferences. Even serendipitous feeds will have a higher probability of being interesting. More of “do what I mean,” instead of “make me spell it out in detail” interfaces, will prevail. Siri is a start, but a great application could go much further and reduce, prioritize, and recommend my to-do list according to the time I had, instead of always making me feel I was behind and driven by external interrupts and priorities. Advertisers could approach our “proxies” directly and the proxies could decide if something is of interest to us—automatically separating “wanted advertising” from spam. Meanwhile, the level of data reduction will be based on our specific context, our current priorities, our social networks and our interest networks. Underlying this is the evolution of A.I. and advanced machine learning — which Idiscussed in a separate post — and even big data analytics.
- Big data or Analytics (Ness, Billguard, The Climate Corporation, Kaggle, Datasift): “Analyzing massive amounts of structured and unstructured data to deliver unique services or analysis.” Perthe OATV website, so many of the most transformative applications today rely on massive cloud databases — often generated by user participation — with meaning extracted from that data by predictive analytics and powerful machine-learning algorithms. We see the back-ends of many of these applications becoming the functional equivalents of subsystems in a kind of internet-scale operating system driving not just the web but mobile devices. Location, social networks, identity, and personalization are just the tip of the iceberg. There will be countless new types of data streams and new ways to make data useful. Fundamental data utilities and applications will be built on these and a lot more information and data reduction will be extracted from this hairy-looking big ball of often-unstructured data. New data sources will become valuable, as will new technology for scaling data, new algorithms, and new ways of connecting people and devices.Bionic software apps (another OATV idea) that supplement or are supplemented by humans in conjunction with big data could become newer versions of Mechanical Turk by Amazon. All these will be part of user utilities, business services, health diagnosis, credit, fraud, risk, education, advertising targeting, user relevant services and much more. The colors and text chosen for menus and buttons will depend upon data analysis, as well as discount offers and a lot more. Much has been written about big data and it and may be getting past the unhyped label! There’s a lot going on already – Tellapart, Factual, Ayasdi, Explorys, Platfora, and Metamarkets come to mind, but there are many more.
- Emotion (Foodspotting, Ness, Instagram): “Services that evoke strong emotions in users,” which is often a component of other categories, can also be enough of a driver to be mentioned separately. I include here the applications that because of their emotional appeal are adopted more rapidly or easily (more pull from users than push to them) as a major component of the application. Some will be useful services (Ness, AirBnB) and in others the emotional appeal itself will be the “product” (Angry Birds, Foodspotting. Fotopedia). The power and leverage of this class of applications are making designers the essential ingredient of a startup’s founding team and “experience” design (instead of just user interface design) a key skill and product offering. I might lump new classes of games into this category though I consider games an established category and I am trying to focus on new, surprising areas in this post. The line is hazy though. One could reasonably put gamification of everything from health to education to training to shopping as a new emotional tool for applications.
- Education 2.0 (it’s early, but Altius, Khan Academy, CK12, Udacity): “Education models that dramatically reduce the cost and increase the availability of quality learning.” The puzzling question is why education has not already changed. My guess is we have not experimented enough with non-linear, rapidly evolving, out-of-the-box approaches but have instead tried to force-fit ‘multi-media textbooks’ and other traditional (often broken) ideas into the “computerized” model. We have also had too much punditry from experts in education instead of just trying hundreds of new ways of doing things. This is starting to change; it makes me optimistic that what has not worked so far can now work, especially given the role gamification can play in increasing student interest and social can play in increasing peer and teacher support and assistance. I believe past attempts have failed because of the specific tactics they adopted instead of the overall strategy of new modes of education. Data analytics on what works can also help here. We can re-imagine credentialing as one of the side outcroppings of online education. For instance, Interview Street expects that a programmer’s performance on their ACM “programming puzzles” could be used as alternative credentials if they never went to an Ivy League school. Add your work on Github and an employer may prefer you to a Princeton computer science grad even if you went to Banaras Hindu University in India. Meanwhile CK12 can get you credit for any time spent during high school helping other students, as part of your application to Harvard, provided you use an online study help group or application. And Altius gives high school students who did not get serious about college until too late a second chance to re-credential themselves and get into over a hundred and forty colleges—basically a second chance to get your grades up! Even more extreme, in a separate post I also examine if we could get to self-driven education in high school and focus on job-targeted skills learning without college degrees (nursing or plumbing anyone?) and more.
- TV 2.0 (Miso, Flingo, Maker Studios, both first and second screen apps as well as content production & sourcing): “TV as an interactive and social experience both on the primary and the second screen.” Most U.S. Internet users, I am told now, have a second screen in front of them when watching TV. Whether it is true or not, it soon will be, and the interaction that is possible will allow for all kinds of creativity and user engagement shows/applications/techniques. More importantly, program production, be it video for TV, audio for radio, or text for next-generation news formats (tomorrow’s “newspapers”?) could be crowdsourced or gamified. This allows for new personal brands to emerge (much like the Drudge Report or Politico or some YouTube channels that are emerging now). Better experiences for users, better targeting for advertisers, more access for programmers and the creative types are all likely. Your proxy or agent prioritizing your viewing or reading queue will be an adjunct area. The big guys and the small guys meanwhile will battle for newer first screen experiences and applications.
- Social Next (intersecting with all the interest graph stuff and verticals like Github, Coursekit, andResearchgate): “Social as a useful and productive part of lives—enabling collaboration and deep community building around the world in specific areas.” I include here new uses of social such as Github to do a cooperative task or the kind of social learning Coursekit and others are trying to encourage. I suspect we will see the power of social harnessed for many applications beyond just the Facebook friends network or the Google+ circles implementation. Beyond the few examples cited above, it is hard to envision all the specific vertical applications, though LinkedIn is just another verticalized social application and I expect to see more of those. Some would say social is a part of most applications/uses, though there is a difference between adding social functionality to e-commerce and applications uniquely enabled and defined by social experience.
- Interest-based networks (where Meebo is pivoting to, Twitter, Snip.it, State): “User driven content that maps to people’s interests both for a better user experience and better targeting.” I was impressed by a post by Naval Ravikant and Adam Rifkin on interest graphs and why they aredifferent from social graphs. I do think a number of startups will either target interest graphs to create a network different from Facebook’s and others will use these graphs as monetization strategies. Social is about friends, while interest is about your interests and the two may or may not intersect. For instance, my daughter’s startup Teethie is trying to do intense conversations around your passions. I see interest-based networks as different from social networks (friends versus interest-focused activities) and I consider interest more easily monetizable and more susceptible to the emergence of innovative new applications.
- Health 2.0 (Jawbone UP, Nike Fuelband, Empatica, BodyMedia, MC10, Fitbit, iBike, Recon, Withings, Alivecor): “Exponentially growing data will yield personalized lifestyle suggestions, improved outcomes, predictive diagnostics and applications we can’t imagine.” Health applications will flourish in many directions and are laid out in a separate post. From more data (especially more baseline, or “healthy” data), “health”-care instead of “sick”-care, more DNA and proteomics data, to mobile-based “second opinions” substituting for doctors and more traditional health systems, we will see an influx of non-sick status data and applications leading to what is called the Quantified Self. You cannot improve what you cannot measure, and these days you can measure just about everything – external factors from pace to distances covered to altitude, and internal factors from heart rate to blood glucose levels to sleep patterns and much more. Alpha geeks have been hacking together solutions to track various types of personal data for years, but with the advent of open source hardware, cheap sensors and smart mobile applications, we believe that there will be a new class of applications unlocking the value of this data. And, in doing so, they will reshape the understanding of our own health and the health care industry as we know it and probably provide a lot of fun, games and motivation along the way. All this data will be complemented by artificial intelligence and machine learning systems.
- Internet of Things/Universal ID/NFC/Smart sensors (a technology with the applications still to emerge): “Sensors and authentication technologies which will interconnect everything and remake our interaction with the world around us.” Sensor networks aren’t just for cargo containers anymore. Sensors have found their way into every aspect of our lives — whether it’s the phone in our pocket, the digital photo frames on our desks or the barcodes embedded with information in our local grocery stores, often complemented by NFC, Bluetooth LE, Wi-Fi and other networks. The network of things is supposedly growing faster than any other network, social or otherwise. We’ve just begun to scratch the surface on the kinds of applications infrastructure needed to harness its full potential. Meanwhile,the management of identity, privacy, security and verification is a huge theme for the next decade and hasn’t yet been addressed much beyond the Facebook login, which makes it exponentially easier to take advantage of people’s online personal data/money/(and privacy which some will complain about). There is probably a completely disruptive way to reinvent online presence and verification, beyond the universal ID system (albeit an offline system with online instantiations) being pioneered in India.
- Personal Collaborative Publishing (Pinterest, Tumblr, storify, Snip.it): “Truly free press with no barriers to entry and personalized interest-based curation.” This trend seems to be moving forward fairly rapidly. I’m not sure if it will become more or less verticalized or what new dimensions will emerge but the potential clearly remains promising. Self-publishing on Amazon is becoming real, removing the gateway of traditional editors and the tax of traditional business models. Where will this lead? Books, especially non-fiction, can become more interactive, crowdsourced (ck12.org), social and collaborative.
- Utility Apps (Siri, Seatme, Ifttt, Uber, and many, many more): “Leverage device ubiquity and context to deliver valuable services.” It goes without saying that we will continue to see more and more utilities that are just plain helpful to us. I do wonder how many major new categories of utilities there will be. Ideas anyone, for major new categories? Utilities will provide personal assistance, productivity and maybe even decentralized work (new clones of Mechanical Turk or Skype!). They will aggregate experts into marketplaces (see below) or crowdsourced services or just plain telemedicine and remote reading of your radiology scans (yes it is hard to separate any of these applications into clean areas – overlap is inevitable). One of the evolutions we will see is that these utilities (and other real/virtual crossover areas like gamification) will require less, not more, input from us as they evolve – as the virtual bleeds into the physical, the enhanced experience will become more seamless and a natural part of activity. In a few years it will seem ridiculous that we had to take out our phones, open an app and type something in order to check into a location in order to activate a daily deal—how awkward! Personal transformation might be a “utility” combination of the Emotion, Education 2.0, and Health 2.0 categories, but we are still trying to tackle the problem of how to use all of these new tools for personal transformation, be it habit forming (e.g. learning, meditating) or physical metamorphosis (e.g. losing weight).
- Marketplaces & Disintermediation (Interview Street, Kaggle, Etsy): “Remove the middle man, increase market efficiency and produce better results, faster“ Marketplaces are about economic efficiency (and active engagement) and more and more of them will keep emerging. My favorite new marketplace is Kaggle, where 13,000 data scientists can compete with their talents and the best ones can benchmark themselves and hopefully get paid accordingly. There will be more of these, unconstrained by the additional friction, the overhead (which is really a tax) or the constraints often placed by traditional gatekeepers. Gatekeepers who collect “taxes”, because of their lock-in, have the most to lose from the freedom that comes from connecting buyers and sellers directly without intermediaries. I wonder when more intermediaries will be disintermediated even beyond marketplaces, and I’m looking forward to it. Why does Tom Freidman need The New York Times to get readers, or can a Washington Post writer just get publishing and distribution services and develop his own loyal audiences as the SBNation bloggers have done? This possibility seems quite plausible in the face of news filters that are tailoring content to user preferences. Deep user preference through big data analytics will reduce the need for the Times’ editorial staff, and editorial work will become a more specialized and more valuable function. By reducing friction and increasing access between producers and consumers of content (and reducing the number of “broker” jobs which are essentially unproductive types of work and friction on the whole system), next-generation marketplaces and disintermediation could expand the rate of development and rate of quality improvement across a wide variety of fields. In the end it means more creative people trying to create – and a much more transparent way to select for excellence and improve on it.
Of course, it is hard to classify any startup into a single category. Kaggle is both big data and a marketplace for data scientists. Ness is at its core a big data analytics play but its appeal is primarily emotional. Many in fact are more often than not enabled by the new mobility and capability in both phones and tablets.
So, what have I left out?
I chose not to define mobile or tablets as a category but it clearly is a major driving force behind much of this innovation; mobile is the theme that underlies the concept of “post-PC” or “always/everywhere.” The emergence of new languages like HTML5 (which I suspect will soon turn into new, hopefully cross-platform standards through the addition of traditional operating system services like inter-process communication) will enable more innovation, which will sell more devices, and drive even more innovation. Other capabilities like sensors around always/everywhere devices will enable health, the Internet of things and other functions. Compasses, GPS sensors, accelerometers, touch interfaces, voice, and image capture all open the door for rich new experiences. I consider all of these enablers rather than categories by themselves.
I ignored areas like cloud computing, because they are not new anymore (though still a source of significant innovation and a source of services that can be drivers of innovation). Given the consumer orientation of this post, I also ignored the changes in enterprise that consumer technologies are driving. That trend I suspect will continue to accelerate and surprise.
There will be both large permanent innovations and categories established as well as passing fads (especially in gaming). I don’t list games as a “new pond” here, though it will continue to grow and surprise us in categories—whatever the next Angry Birds/Farmville phenomenon will be—while gamification will become pervasive in everything from education to health to shopping. Given the similarity to social’s ubiquity, I admit a slight inconsistency given that I’ve included new classes of social applications labeled Social Next. Both gamification (separate from gaming) and social may become basic tools that enable many of the areas I mentioned. Still, I cite some examples of what defines Social Next versus adding social functionality to an application like e-commerce.
I also did not focus on e-commerce given its already substantial popularity. Still, we will continue to see innovations in this area, especially given the different optimizations that are possible on new sized screens like mobile phones. I expect e-commerce to be disrupted by many of these ponds—and the move toward all commerce being e-commerce creates massive potential. For instance, what’s the potential impact on local merchant expertise getting supplanted by mountains of behavior data, curation and social recommendations? I wonder what will happen to local or hyper local products; will that be the domain of the traditional large players or Internet players given their scale and greater access to data versus the local merchants? That question is in the hands of the data analytics and data reduction applications. Will some local merchant expertise get supplanted by disruptive data analytics and reduction in some categories, while other services and certain products from local merchants get enhanced?
Then there are payments; I think it is possible that we are seeing just the tip of the iceberg in a potential rethink on payments. We as investors have seen Square take off at an unprecedented rate (so far) for a payments startup, but in terms of relative scale, even Square is dwarfed by Mpesa — it is 20% of Kenya’s GDP already (using a totally different model than Square). Meanwhile in India, their UID system could remake the concept of “cash”. Feel free to include or exclude payments and even next-generation currency from the “new new” categories that are emerging. Diablo 3, a next-generation role-playing game from Blizzard, will be the first game to have a double-world auction house: you can use real-world or in-game currency to sell and buy items. Is this another sign of new currencies, or is it just a fancy loyalty program which gets you to buy more—the reincarnation of airline miles?
Facebook has validated another category I haven’t mentioned, “Timeline”, and others are looking at “health timelines.” This is a feature that will show up everywhere and, to me, is more of a tool than an application. There will also be technology services like Singly and Dropbox that allow applications with new utility and features to be built. NFC may also just end up as a basic service technology, but I listed it above in the hope that it enables a new class of applications. There are probably other universal features that will initially look like applications.
Then there is the Maker movement. Makers are enthusiasts who hack and modify the world around them in interesting and whimsical ways. Tools and services that used to be inaccessible to all but large manufacturers are now available to everyone. Foreign factories that were impenetrable before are now an email away. Design software costing thousands of dollars per seat is freely available (or very cheap). Hackers are mixing all of these elements together and re-imagining entire industries from the ground up.
As with technology movements before it, the Maker movement has laid the groundwork for what will be the next industrial revolution based around personalized fabrication from one-off runs with 3-D printers to at-scale manufacturing. I am not yet sure that this will be a category in the next 2-3 years, but it will happen sometime. I do want my sofa custom-made for the size of my living room, in custom colors, and maybe a custom design while cutting out all the expensive intermediaries in the supply chain. As an example, I recently came across an intriguing example of crowdsourced design for custom vehicles and automotive parts by Local Motors.
While talking about new tools or technologies I would be remiss if I didn’t mention a category of entrepreneurs I find particularly intriguing: “The under 25” who don’t know what they don’t know, mostly have not worked at what traditionalists would call a “real job” and are not afraid to try new things. They are often most creative in their thinking and willing to try things and tolerate failure. Peter Thiel’s “20 under 20” is an extreme example of this as are the many Y Combinator startups. I’m very excited about what the next few years will bring — the rate of change is accelerating and the possibilities are endless!
