Monday, July 31, 2006
Now Amazon have opened a store there - see http://www.typepad.com/t/trackback/5523140
And this weekend Dell opened a real store in Dallas!!!
More on Social Networks business models
1) MySpace is flying - traffic wise - and has passed Yahoo! and Google in terms of page views. See Jeff Clavier at http://www.typepad.com/t/trackback/5297436 for all the good news there. Although some claim that MySpace's design produces a large number of extra page views - maybe 2-3 times as many as are required
2) There is a lot of uncertainty about how much revenue they are making. Information in press releases from Fox Interactive Media (FIM - the division of News that owns Myspace) is confusing in that it talks about Myspace but then seems to give figures for the whole of FIM.
3) Nonetheless they are making big claims - $1bn in revenue by the end of the decade
4) CPMs are VERY low - reported as 10c average - adserving technology costs on their own are likely to be 1/3 to 1/2 of that (note the extra page views point above which should, in theory at least, have the effect of reducing CPMs by 50-67%)
5) One reason advertising may not work (or CPMs will be low) is that people come to MySpace to visit each other, not to access content
6) On the other hand some rumoured numbers I have heard for Facebook penetration of their vertical and average time spent per day on the site are staggering and lead me to question my whole thesis here that the traffic may not be monetisable
7) People are starting to talk about the business model for MySpace not being traditional advertising but instead a recommendation or collaborative filtering tool
It is easy to see why opinion is sharply divided on whether MySpace was worth the $580m or not. I'm still of the opinion that on a pure DCF basis it will turn out not to have been worth that much. For NewsCorp it may have been worth that much to get a foothold in the new world and start learning. It is only 2.5% of their market cap and I can't shake the suspicion that part of their justification was that being associated with the hype around MySpace will do more than that for their share price over the short to medium term.
Thursday, July 27, 2006
Social network business models and value - Bebo and Myspace
On the one hand they have A WHOLE LOT OF TRAFFIC - so there is unquestionably some value there. The question, is how much?
Now I'm an old fashioned guy when it comes to corporate finance but to me profits and cash flow are important. To really be worth a lot of money you need to have cash. At the risk of being boring; a company is worth the discounted value of its future cash flows.
Sometimes it seems like there are two investment philosophies out there in the London market. 'Follow the hype curve' and 'invest on fundamentals'. I would be a fool if I never sought to capitalise on hype round a hot sector, but my leaning is towards investing on fundamentals. So before I can put a big value on a company I need to understand how it could make a lot of money, even if it isn't doing it yet.
So to Bebo and Myspace. They have traffic in abundance, but how are they monetising that?
My understanding is that they are pursuing advertising models. Fair enough. The two issues they face are that advertisers are wary of advertising next to unknown and potentially offensive content (e.g. soft porn) and that users might find advertising intrusive and leave.
The response is to make "safe areas" where the content on the pages is understood. So they almost become traditional content businesses with a radical new method of sourcing traffic. Both Myspace and Bebo are doing this around music, in the first instance.
They have traffic and a clear way to build content so they can generate revenues (in addition there is the not insubstantial home page advertising revenues).
That leaves the questions of how much? and for how long?
To me there is a lot of uncertainty on both fronts.
The "safe areas" strategy sounds like a good one, but it is unproven. There is no way of knowing how much of these sites' traffic can be monetised in this fashion so there is no way of knowing the quantum of revenues that can be generated.
There is also no way of knowing if the current popularity of these sites will be maintained. They are new, there is no history we can look at to inform us. There are reasons to think it could be sustained - mostly based on the argument that people have invested a lot of time and effort into building their profiles, so they won't leave. There are reasons to think it might not be - the current crop of 18-24 year olds may not need this form of social interaction when they go to college, or grow old and get married, and the next crop who are currently say 12-18 are currently using other social networking sites targeted at their age group (e.g. Stardoll and Sulake) who are presumably planning to hold on to them.
So these sites could generate a lot of revenue for a long time but they might generate a lot for a short time, or a little amount for a long time, or worst of all not very much for not very long.
To my mind it is hard to put a big value on that. Far too much uncertainty.
Valuation is an art though, not a science, and people have different ideas. NewsCorp paid handsomely for MySpace, and good luck to them. But you can't bet on people paying more for your companies than you think they are worth, it happens, but you can't make that your investment thesis.
Wednesday, July 12, 2006
Trust and certificates
In what is starting to be a tradition in this blog lets start by postulating what trust might be (and I'm talking in the very specific context of web trust systems - there is a very interesting wider debate on trust in society which I've blogged about before).
Forms of trust on the web:
- Certificates that say you can trust a site to be something - e.g. VeriSign for security against theft of your credit card information
- Trust systems that let you trust an individual in a given community - e.g. EBay's rating system
I'm trying to answer the question of whether there is an infrastructure play in trust on the web.
In the first category VeriSign has built a business based on a single very specific area of trust but I think security of credit card information is probably the only area where trust is missing on such a broad scale. Other areas where certificates are prominent are child safety (ICRA), accessibility (RNIB and others) and more recently readiness for mobile. There will be a lot of certificates required (at least in the first two areas) which tells me there is a volume of business in administering them, although I'm not sure how profitable it would be. There is an element of natural monopoly though, which might be helpful. The million dollar question (and maybe that should be billion) is whether the list of three certificate areas I gave above is (or soon will be) much shorter than it should be. If it is much too short then there ought to be a play in "trust certificate infrastructure".
In the second category I don't think there is an infrastructure play - the technology to administer an eBay style trust system will be coded on a site by site basis and it doesn't strike me that there is much of a software product opportunity here. Instead, everyone is talking about a service play which allows people to export their rating from e.g. eBay and use it elsewhere. This would be great but it strikes me that implementation will be challenging. There is little in it for e.g. eBay to let this happen, indeed people could use their eBay rating on competing auction sites, and there is the hazard that eBay will not be able to police abuses of eBay ratings outside of their site. The play then morphs to some kind of meta rating system which is probably tied in with the holy grail of detailed individual profiles. It'll come, but it will take a while.
(An aside - I've been saying for a long time now that one day we will all have profiles on the web that we love, nurture and cherish like people used to love nurture and cherish their cars. )
Thursday, July 06, 2006
More on Web2.0 models
All good fun and very exciting. I might have managed to have a few beers on Friday as well. And then a barbecue at Bob Hook's house on Saturday for all the new Esprit'ers. Great pad. Fantastic garden.
So very busy, but I'm going to attempt two posts tonight. This one explores a little more where I think the money might be in web2.0.
In my first post (which was a month ago and this is only number three - I need to speed up a bit) I put web2.0 companies into three categories:
1) Stuff people have always done - simply moved to the net - e.g. ActiveHotels
2) Old stuff done new ways on the net - e.g. LastFM, betfair
3) Stuff people have never done before - e.g. Myspace, Blogger
Now I'm trying to figure out where the money is going to be made in the next few years. To take each area:
1) Stuff People Have Always Done - Simply Moved to the Net
This is the easiest to understand and unsurprisingly the most developed. Opportunities for small companies here are limited to the increasingly small universe of "what hasn't been done before". Furthermore it is the easiest stuff that has been done, meaning the remaining opportunities are the ones with the most execution challenges. Nonetheless as the web develops things that are on the margin of feasible today will become good opportunities tomorrow and the great thing about a lot of companies in this space is that they have transactions at their heart and the business models come naturally.
The features of markets that are ripe for this kind of disruption are:
- Information is far from perfect - i.e. buyers and sellers struggle to find each other
- There is a path to critical mass
- Transaction size is not too large
2) Old Stuff Done New WaysCompanies in this space often seem to start by giving their stuff away for free and they only come to VCs for the big money once they have thousands (or millions - pick it) of "users". And therein lies the rub. What exactly is a user? This is all obvious stuff, but to get it down for completeness users can vary from registered and used once for 10mins, to occasional free user, to regular free user, to paid once user to regularly pays user. Ultimately there is only value (and I mean real value, the sort that can get your company bought for a LOT of money) if it is clear that there will be lots of "regularly pays" users. The problem is that valuation expectations start to leap up when there are lots of non-paying users and the risk that there will never be many regularly paying users remains substantial. I thank Zennstrom and all involved at Skype for putting European VC on the map last year, but people need to understand that emulating them is not a viable investment thesis! So will there be a lot of big companies created here in Europe? My gut is telling me there will be a few. Collaborative filtering has the potential to disrupt a lot of media consumption. It will take a while though - people's behaviour needs to change, people need to start forming more communities, learn to tag etc.
The features of service ripe for this kind of disruption are:
- Large market
- Structural problems with the way things are done currently (e.g. telephony too expensive)
- Ability to launch a basic service at little cost and therefore give it away for free
- Ability to market virally (that is why comms services spread so quickly - from Hotmail to text to Skype to blogging)
3) Stuff people have never done beforeThe growth in blogging is astounding. In March 2005 Technorati announced it was tracking 7.8m blogs and that the blogosphere was doubling every 5 months - in April this year they were up to 37.3m blogs - the growth continues. This is phenomenal. Teen social networks and geek social networks have also had dramatic growth. Ajax personalised home pages may be the next place to go. To my way of thinking this is all stuff that is new - it is true that kids used to hang out together in the street and now that we won't let them outside at night they are doing it online, but there is a personal expression thing going on here that is new. Making money here is not easy either. Most of the business models are long-tail advertising based and unproven. The sustainability of these properties is also more a matter of faith than evidence, to my knowledge at least. It would be wrong not to mention here the online personality extension plays - e.g. sulake and stardoll - they have a revenue model at their core - selling lots of stuff to lots of kids that adds to their identity for pennies a time. I think there will be more opportunities here as well, although they will be for the brave.
I don't think it is useful to be prescriptive here as I have been for the first two categories. Spotting the next thing is a matter of keeping ones' ear to the ground (and maybe keeping a close eye on what my brethren on Sand Hill Road are doing...).
There is a little opportunity in (1), a little more in (2) and a little more still in (3). Which tells me that this is not a useful categorisation for the purpose of identifying where the opportunity will be! But nonetheless, this has given me some kind of framework for thinking about opportunities in this arena.
So it is only going to be one post tonight after all. One to follow shortly on infrastructure/trust.