|
|
Google is currently rolling out a new feature called ‘Knowledge Graph’ which will enhance our search experience by intelligently guessing what information we are looking for and putting it at the top of the results page. In other words they will be serving information not links. The picture above was released by Google to show how this will look on multiple different devices.
Many of you will have seen vertically focused instantiations of this strategy in your search results already, and if you are like me you will think they are pretty cool. This weekend I was in Munich and when the discussion turned to the relative sizes of different German cities I was able to activate voice search on my Android phone and ask ‘what is the population of [insert city name here]?’ and get a number straight in the results page without having to do any more clicking. Similarly, I like being able to search on ‘weather xx’ and flight numbers and get the results straight on the page.
The Knowledge Graph initiative is an attempt to build publicity around this strategy and broaden it to more areas.
That’s great for the searcher, and it’s great for Google, but it isn’t so good for the owners of other internet sites. Wikipedia is the obvious sufferer from the population examples I mentioned and I bet weather sites are seeing a drop off in their natural search.
There are numerous examples of other initiatives by Google which arguably improve the experience for searchers but disadvantage other sites on the web, particularly if Google is biased in it’s search results. Google+, Google Places, and their acquisitions of Zagat and BeatThatQuote spring to mind.
I think what we are witnessing here is Google acting more and more like a media company rather than a search company. The distinction is important for anyone who wants to build a business that relies on natural search. In recent years many businesses have achieved success with this strategy, but going forward there is an increasing risk that Google will effectively decide to compete with your business. (Note that this risk is in addition to, although not entirely independent of, the way that Google is increasingly demoting the position of thin affiliate sites in search results pages.)
This is another example of the way the internet economy is moving away from an open collection of sites services with tools that allow us to find the best ones for our purposes at any given point to a world where partnerships and playing nice inside someone else’s empire is the best route to success. Today the most important empire owners are Google, Facebook, Apple, and Amazon, with companies like Twitter, Quora, and Pinterest as challengers.
Facebook is picking up a lot of criticism as it heads into its IPO. In particular there are questions over its continued growth and the effectiveness of its advertising formats, largely on the back of General Motors saying its Facebook ads haven’t been paying off. And to top that an increasing number of insiders are deciding to vote with their feet and cash in on the IPO.
I think all this criticism is a function of the valuation rather than a reflection of the quality of the company. As you may have seen Facebook just raised its IPO price and will be worth over $100bn if it goes out at the top of the range. Meanwhile the FT is warning that retail demand risks inflating the IPO.
I think that if I the valuation was considered to be low everyone would instead be focused on what a great business Zuckerberg and his team have built over the last eight years.
Venturebeat gave us a reminder what they have achieved this morning when they wrote that Facebook is breaking three records with its IPO:
- Facebook will be the largest venture backed IPO of all time (c4x what Google did in 2004)
- Facebook has raised $2.2bn in venture capital – more than any other VC backed company in the US and nearly 2x Clearwire, the second placed company (Twitter has raised $1.1bn)
- Facebook has acquired 13 companies, more than any other company in the run up to its IPO. Twitter is second with 11.
Most companies would be happy reaching a value of $2.2bn, let alone raising that much money.
I say hats off to Facebook.
Regular readers will know that I’m a fan of open systems and have been following the battle between Android and Apple’s iOS with interest. For a long time I have thought that the most likely outcome is that Android becomes the dominant player with Apple remaining important with a significant but minority market share. However, as time has passed and events unfolded I think Apple’s chances of prevailing have been increasing, largely due to their total dominance of the tablet market.
Today I saw some detailed data on the extent of fragmentation in the Android community, and whilst in a straight odds bet I would still back Android, I am now thinking that is a pretty close call.
OpenSignalMaps have an Android app that shows the signal strength available from different carriers and they did an analysis of devices accounting for 681,900 of their downloads. The chart below is a graphical breakdown of which phones accounted for the most downloads, and as you can see there are a huge number of tiny rectangles. Overall they counted 4,000 different devices. (The green square with the label GT-I9100 is the Samsung Galaxy SII, and if you click through to the original post you will find an interactive version of the chart with more device level information.)
4,000 devices is an awful lot (even with the caveat that custom ROMs overwriting the android.build.MODEL variable will have inflated the numbers a bit) and a big challenge for developers given the variety in OS’s, screen sizes etc. If you click through to the original post there is more information on both these elements and all I will say here is that homogeneity is thin on the ground and only 8.4% of devices are on the latest version of the OS (Ice Cream Sandwich 4.04).
Fragmentation makes it difficult for developers to deliver a consistent and high quality experience and based on this data I would say that for startups with a choice beginning with iOS is a no-brainer, particularly given the low numbers of phones running Ice Cream Sandwich. This reflects what I see in the companies we work with and talk to, and whilst I have previously thought I detected a shift towards developing first for Android, now I’m not so sure.
Access to the latest apps is important for early adopters and maintaining the iOS’s status as the first choice for app developers will be real boon for Apple.
When I read on Techcrunch this morning that Quora has raised $50m at a $400m valuation I thought “‘wow’, that’s interesting, I haven’t heard much about them recently, I wonder how they are getting on”. First I checked their Alexa graph which seemed to confirm my initial impression that progress has slowed following a meteoric rise in 2010 (or more precisely their Traffic Rank dropped Q2-Q3 2011 and has been rising again since, recently regaining the earlier peak).
Then I took a look at Quora itself.
And ‘puff’ an hour of my life disappeared, just like that. The content on Quora is now of an amazing quality, and much of it is sifted and prioritised for me by my friends asking, answering and following questions and upvoting other people’s answers.
I’m a very task driven person and as a result Quora doesn’t fit very well into my reading habits. Every morning I exhaustively read two feeds I have created on Taptu and the Financial Times mobile app on my Android – a reading habit developed to reliably complete two tasks – staying on top of the news (politics and finance news from the FT, tech news from my tech news feed on Taptu which aggregates The Kernel, Techcrunch, Venturebeat, GigaOM, and a selection of other tech news blogs) and reading everything produced by a small number of my favourite bloggers (most notably Fred Wilson and Brad Feld). Spending time on Quora doesn’t help with either of these tasks which is I think why I don’t go there very often. Rather Quora is good for improving your general understanding of the world in unpredictable ways, which isn’t something I currently make time for in my daily schedule.
But maybe I should. I read some great stuff today on topics like why aren’t VCs lean and scrappy, and why Pinterest took so long to go viral, but the pick of the bunch by far were some of the answers to What are some decisions taken by the “Growth team” at Facebook that helped Facebook reach 500 million users?. Here are some of the highlights:
- they had a ‘fundamental understanding’ of their product and why people use it
- they combined simple frameworks for understanding the business with rigorous analysis to precisely target development efforts on areas which would yield big improvements – e.g. their user framework had four elements – acquisition, retention, engagement and resurrection and they mapped different products and different areas of the site onto each of these elements so they could precisely figure out how to optimise each of them
- they were incredibly thorough – e.g. going the extra mile to support contact imports from a huge variety of services (to maximise virality)
- they have a team of great people and a unique culture
This list reads like a checklist of attributes for a great company generally (I’m mentally checking my portfolio against it now….). The Quora post has much more colour.
Twice in the last week I found myself coaching founders on how to build a financing plan around value creation milestones so I thought I would share what I said here.
The idea of thinking about value creation milestones comes from the observation that the value that investors put on a startup is generally moves in a step function, even when the company itself is making steady progress. There are certain types of events that investors look to as evidence that a company has overcome certain challenges and has therefore reduced risk in the business and become deserving of a higher valuation. For new investors in a business it is often hard to assess the progress being made towards the milestone and hence little credit can be given until the milestone has been achieved. An easy to understand example of this is in the run up to a product launch – whilst the team has been working hard, can see the product coming on in leaps and bounds and becomes increasingly confident that they will launch on time and customers will come rushing to the door, for investors it is very hard know whether the product will work and be well received until it is in the hands of customers. Hence the value of a company is often relatively flat in the run up to product launch and then leaps up once it is out of the door and starts to get traction.
Typical value creation milestones for technology startups include:
- launching a product
- achieving first revenues
- breaking through revenue thresholds – e.g. run-rates of £1m, £5m, £20m and so on
- demonstrating a repeatable sales model (typically telephone sales, direct sales, or channel sales)
- for consumer internet companies, breaking through user based thresholds – e.g. successful public beta with thousands or tens of thousands of users, 1m monthly uniques and so on
- identifying a scalable customer acquisition channel
- demonstrating profitable customer acquisition (usually by analysis of unit economics)
- signing distribution deals
- demonstrating that signed distribution deals are working
- achieving first international sales
Given that the valuation of a startup increases when milestones like these are hit it makes sense to build a financing plan based on when they are likely to come in. Because it takes time to raise money and investors generally want to wait until the milestone has been hit before they will engage seriously in discussions about investing at the higher price point I generally advise raising enough money to last six or nine months after the milestones is expected to drop. That builds in time for delay and time for a leisurely fundraising. In his book The Second Bounce of the Ball, Ronald Cohen, one of the founders of the UK venture capital industry, is even more conservative – he advises raising enough money to get past two milestones.
However, raising lots of money to buy lots of time is great if you can do it, but for a lot of startups it won’t be possible, at least not without excessive dilution. In that situation I still think you need to stick with the discipline of thinking through when the value creation milestones will occur and timing your fundraising accordingly, but look for ways to compress the timescales. If possible bring forward the trigger events for the milestones, and in any case take your plan, including the milestones to investors before you have hit the milestones. Explain to them that you want to raise money quickly once the milestone has been achieved and ask whether you can keep them updated with progress and at what point they would like to enter into serious discussions. Hopefully you will find an investor or two who likes you and your story who will work to get themselves into a position to make an offer as quickly as they can after the milestone has been achieved.
In one of the conversations I had last week we ended up thinking that they should raise twice as much money as they had originally thought now in order to give themselves twice as long to raise the next round (twelve months rather than six). Our thinking was that with twelve months money the following combination of milestones and investor communications will be possible:
Milestones
- July – pilot deal with a major retail partner goes live
- September – marketing support comes online
- November – retail partner decision to roll-out product to rest of stores
- January-March – anticipated strong sales months
Planned investor communication schedule
- October – first conversations on the back of marketing launch and early data from pilot
- December – tee up serious conversations for new year following positive decision to roll-out beyond initial pilot
- Jan-March – work aggressively to close another round in the momentum months
This is an early stage business and hence any number of things could change and/or go wrong but to my mind this is a pretty solid plan. The timelines are tight, but realistic, with two months contingency and the milestones will all be meaningful to investors and if the business performs I think the planned conversations will all be of interest.
Our portfolio company StrikeAd recently released a new video describing what they do. I may be biased, but I think it is a great video, and well worth watching to learn more about the company and the latest trends in mobile advertising or to see a good example of how a well shot video can effectively introduce a complex product and lend gravitas to a business.
StrikeAd video
I’m sorry to post a link rather than an embed. I’ve spent 25mins trying to get the embed to work and I only have 30mins set aside today for blogging.
GigaOM yesterday published an article about How Google is growing up into a real IT company, which talks about how the world’s favourite search engine has switched from releasing lots of ‘science project’ applications like Google Wave and Google Knol, and Google Questions to a fewer number of well thought through and solidly engineered products like Google Drive, and BigQuery.
There are lots of good reasons why Google is making this shift, the most important and obvious of which is the need to focus on a smaller number of things and make them work. The second most important, is maybe less obvious, and that is a desire to strengthen the enterprise side of their business, and this is what made me think about Microsoft.
As I see it the parallels are:
- Both started with a fantastic monopoly business – Microsoft Windows and Google Search
- Both expanded successfully into adjacent areas which leveraged and protected their monopolies – Microsoft Office and Google’s Android
- Both ran into trouble with regulators for abusing their monopoly position to advance their other products – Microsoft with IE and MediaPlayer and Google for using individual’s search data in their other products
- Both were blindsided by a major market evolution which undermines their core franchise – Microsoft by the internet (espescially Google) and Google by social (especially Facebook)
- Both have sought to reduce their dependence on their threatened monopolies by moving into the enterprise
Google has probably been 6-7 years behind Microsoft in each of these steps and 6-7 years ago Microsoft looked about as strong as Google does today. Hopefully Google will hold its current position more effectively than Microsoft was able to.
As you may well have heard Google has been developing self-driving cars for some time now. When they cross the chasm to maintream use they will be a huge force for good in society, reducing accident rates, easing congestion, facilitating car-sharing, enabling elderly people to keep driving longer and allowing millions of people to re-claim time that is currently lost to commuting. That’s a punchy sentence, and I chose each word carefully.
If you have yet to see a video of these cars in action go search for them on Youtube. You will find some incredible videos there including some quite hairy footage of fast autonomous driving around car parks and slower paced shots of cars navigating traffic, stop-lights and pedestrians. The cars are kitted out with laser range finders, radar, cameras, and inertia sensors which connect to an onboard computer which controls the engine and steering wheel. They have driven over 200,000 miles on the roads now without accident, although there has always been a human sat in the driving seat ready to take over if things go wrong.
The news today is that self-driving cars have taken a meaningful step forward towards commercial reality. Cars are currently required by law to have a driver but Nevada has made the first step towards making them legal by creating a license category for driverless cars. The license will only be granted to cars that have passed stringent safety tests and requires two people to be in the car whilst it is driving, including one behind the wheel, and the first license has been granted already. To Google. I have heard that Nevada wanted to the first state to embrace self-driving cars because they believe that companies will then come to Nevada to develop the underlying technologies bringing jobs and investment to the State. Google and other lobbyists for self-driving cars convinced Nevada to act quickly by talking to other states and countries and creating a sense of urgency and competition.
There is still a long way to go before we can tear up our driving licenses, and the key stumbling block is perceptions of safety. Self-driving cars have a better safety record than human driven cars but the image of a computer controlled car going wrong and causing a fatal accident is a powerful one, and one which makes everybody nervous, but the data so far shows these fears have no grounding in reality. It will be interesting to see how quickly Nevada is willing to embrace this technology and legislates to make it a commercial reality. Given the scale of the benefits and the limited downside I think it is a question of when rather than if.
 Search advertising is generally thought to be more powerful than display advertising because when a user searches for something they often have an intent to buy it. Social networks (along with email and other communications services) are amongst the worst places for display advertising on this measure as most users are there to talk with their friends and are not thinking about buying anything. As a result advertisers are only willing to buy ads on these sites at low rates and the general thrust in the industry, led by Facebook, has been to drive up these rates by targeting the ads based on data about the users, and even make them more effective by using information from people’s friends in the ads.
LinkedIn, however, is a little different. As the Financial Times pointed out this morning:
users come to the LinkedIn site with an eye to completing a transaction. They are potential sellers of a good (their professional services) for which buyers (employers) may pay a lot of money.
i.e. they come with intent.
That is why LinkedIn’s first quarter results were so strong. Revenues at $189m were up 101% on the year ago quarter, and net income net income was up 138% at $5m (that’s still low in percentage terms).
I’ve said before I think that sites which leverage social to help us do the things we already do more efficiently will be interesting investments over the next year or two. I stand by that, but would now hold LinkedIn up as an example of a company which has already had success and which fits that investment thesis. I am also refining the thesis to note that some community sites are strong on intent to purchase, which will likely result in superior monetisation.
I think that one of the hardest things about being an entrepreneur is figuring out who to listen to and when to take feedback on board. At the earliest stages there may well be precious few people who understand what you do well enough to provide constructive criticism and there may well be hoards of people (particularly non-techie friends and family) who don’t get what you are doing and advise you to stop what you are doing or radically change course. At these earliest stages most successful entrepreneurs fall back on their conviction and become determined to prove to the world that they were right.
But as your company develops and you build networks of people who get what you do and particularly as you start to get customers then the dynamic changes. At this stage the best entrepreneurs listen well and iterate their plans and business models accordingly. They don’t take all the feedback on board of course (even customers often don’t know what they want) but they start to balance their conviction with the feedback they receive.
Making this shift requires a change in mindset and is often challenging. When I am working with companies I am liberal with my opinions, but I don’t expect all of them to be taken on board. I like to think that they are listened to, and that some of them will be good, but I expect the entrepreneur to balance my feedback with her conviction and other sources of ideas. In fact I look to the skill with which they perform that balancing act as an indicator of likely future success. Listening poorly or listening too much are both red flags.
I’m writing this today because Roger Ehrenberg wrote a great post on this topic earlier this week. This is the best bit:
Having deep conviction around solving a specific problem or engaging users in a meaningful way is the essential element for starting a company. From there, however, art and science diverge. As you build the company, shape the product and spend time working with and trying to acquire customers, you will collect a bunch of data. This data will give you a sense of whether or not you are on the right track and if a broader array of users perceive your product’s value in the way you do. You will also likely observe others in the marketplace, both direct competitors and those whom you aspire to be which will influence your thinking. You might also have mentors and advisers with relevant experience and perspective who will weigh in, filling out the information mosaic.
Distilling relevant input and shaping the product without losing the essence of the vision and mission is the delicate balancing act most founders face. Some founders hit product/market fit just right the first time. But the overwhelming majority do not. They have to synthesize massive amounts of structured and unstructured data and make good decisions. This is the magic of great builders.
Accelerator programmes are sometimes criticised because the startups are buffeted by too much feedback from mentors. I don’t think that too much feedback should ever be a problem per se (although startups must spend time doing things as well as talking with mentors), the issue is more how it is dealt with, as Roger eloquently described. I suspect that most good accelerator programmes already advise companies on how to process feedback. Maybe they could do so a little more.
|
|
