The snake in Facebook's garden...

Turning away from Bloomberg and Apple for the moment (rest assured, I still have many thoughts to post on both; I am particularly intrigued by Bloomberg's Polarlake acquisition and how it pushes it towards to world of Big Data), some more thoughts on the shifting platforms of Facebook-land.

Some interesting data-points in this arena over the last couple of days.

A million users's isn't cool. Do you know what's cool?

Facebook announced yesterday that as of 14th September it had hit a billion monthly active users. Extrapolating that forward it implies 1,009m MAU's at quarter end, up from 955m in Q2.

What was more interesting however was the underlying in the appendix.

Look closely and there a snake in the garden. Well actually, three of them:

Firstly, although user numbers continue to grow, growth rates continue to slow:

That's not that surprising. It's the inexorable law of large numbers, but its also a reason why large companies tend to attract lower valuations than smaller ones.

The second thing is that they mentioned in the appendix that mobile users hit 600m. This means mobile users are now a stonking 59% of the mix (they didn't disclose the more interesting mobile-only users in this release; for context this was 11% of MAUs at Q2).

Remember mobile users are lower revenue than Facebook, and judging from my experience, Facebook haven't figured out how to drive this up. My wife continues to be spammed daily by adds for Salesforce.com, and I remain unconvinced that a soil mechanics PhD student really needs a mid-market on-demand CRM system.

Thirdly, and most interestingly though Facebook disclosed the median age of its user base as of Q307, Q308, Q310 and now (the points where it hit the 50m, 100m, 500m and 1bn user milestones). I may be wrong, but I think this is new disclosure. What's fascinating though is that if you have the age of the given user base at any point in time, you can figure out the average age of new users:

Median age for Q309 and Q311 not provided, so this is interpolated, weighted by (reported) MAU user growth.

This is curious. I would have thought as Facebook became more mature and extended beyond its campus-origins, the average age of new users would have gone up. Instead it's actually fallen consistently from 26 in the early years to 20.4 in the last twelve months.

Remember Facebook's big challenge is to increase ARPU - but if you're new users are getting younger this is a headwind because younger people tend to have lower disposable income. Yikes.

But there's more.

I suspect the real reason for younger and younger users is that at Facebook reaches maturity in (richer, older) developed markets, new users are coming from (poorer, younger) emerging markets. The table below shows the mix of quarterly net adds by region (and yeah the funny bit on the right is because US & Canada MAUs were negative in Q212):

But there's more.

Are you've no doubt realised, emerging markets are poorer markets. So lets use regional GDP/capita to figure out the average GDP/head for new Facebook MAUs:

Yikes!

So in summary, Facebook's volume growth is slowing, is shifting towards lower ARPU mobile customers and its new customers have less money to spend.

Are you sure this is the kind of stuff you want to issue a press release about???

A zinger from Zynga

The other fun fact yesterday was Zynga's Q3 profits warning. Basically they reported a Q3 revenue and EBITDA miss and cut guidance. The chart below says it all (remember this thing was meant to be a hot growth story):

NB I've estimated Online Game Revenues (the core growth driver) by
assuming that (previously growth) Ad Revenues were flat in Q3.

Remember Zynga accounted for 10% of Facebook's revenues in Q2 (though now highly likely to fall in Q3!) so the presenting issue is that this is (yet) another revenue growth headwind for FB.

However this is only the presenting issue. The underlying issue is a more fundamental problem of Platform Risk. This is something I touched briefly on in August and I want to write more about next week.

Zynga is dependent on the Facebook desktop platform for a large chunk of its revenues. Users (as we saw above) are shifting away to a mobile platform where ARPU is drastically lower. Zynga do not control the platform users are on, and do not control their destiny.

This is an issue for Facebook on two levels. On one level Facebook are also exposed to mix shift to lower ARPU.

But on a deeper level think about it like this:

Facebook is dependent on the Android, iOS and Windows platform for a large chunk of its revenues. Facebook does not control the platform users are on, and does not control its destiny.

:-x

Valuing Bloomberg - The $46bn Question

A little coda to my series on Bloomberg. I've been a bit dis-satisfied with my discussion on valuation - I've basically short-circuited a serious valuation discussion by saying "put it on 4x sales or a market multiple. This is a big company".

I think its worth doing a more thorough job to figure out how much this company is worth. This is partly because as an equity analyst that's kind of my job (and I enjoy flexing my valuation-muscles every now and then). And it's partly because I think people will find it interesting (not least any Bloomberg employees who hold stock!).

Edit (5th Oct): Just noticed that Silicon Alley Insider have put up their list of the top-100 private tech companies, with Bloomberg crashing in at no. 2. Interestingly their $35bn valuation is bang inline with my calculation of Bloomberg's enterprise value, albeit using a much simpler approach (5x trailing revenues). Having said that, they don't take into account Bloomberg's cash pile, which I reckon boosts the value of the equity by another $11bn (and in theory would take them to #1 spot on the list!).

How to value a company

As I've written before, the only true way to value a company is to figure out what you, as an owner, are going to be getting out of it. Either in terms of cash in the bank (a classic Discounted Cashflow analysis), or in terms of getting a return on your investment (Enterprise Value-Added analysis).

A word on ratios

Note, as I've said before, that valuation multiples are only a shorthand to get to valuation. Particularly comparable company analysis, where you say "I think its worth x because company y is worth z". You don't have to be a rocket science to realise there are at least three highly qualitative judgements going into that statement

• "It's worth x"
• "its comparable to company y" 
• and "company y is worth z".

Would you fly on an aeroplane built using that kind of logic?

DCF and its failings

Anyhow, both DCF and EVA are mathematically identical solutions although they differ greatly in their conceptual approaches. DCF is the most straightforward one, beloved of a class which I call "P&L jockeys". Basically it tells you how much cash profit the company is going to make which you, as an equity holder, will ultimately receive. Very simple to understand.

However there are two problems with DCF. The first is that is enormously hostage to your terminal value assumption. Once you get beyond the horizon of forecasted cashflows (normally about ten years), the standard DCF values the remaining lump by capitalising it using the Gordon Growth model. Mathematically this is perfectly reasonable, but you end up with the terminal value accounting for a huge part of the value. In effect you are making one vast assumption: "I think in the year 2512 [remember we are discounting into perpetuity], this company will still be around and generating cash". Now ask yourself, how many firms founded in 1512 are still trading today?

The second problem with DCF is that as it focuses on the P&L it often ignores the balance sheet side of the business - i.e. how much you need to invest in real assets to sustain that juicy revenue growth. In theory it includes working capital and capex assumptions to capture this, but in reality they are often an afterthought. Do note however that this is less of an issue for tech companies as they tend to be less asset intensive (one of the reasons I love tech business models).

The EVA fanzone

EVA takes the opposite approach - it is all about balance sheet. It starts with the book value of your assets and how much it costs to fund them. It them figures out how much your profit is and calculates whether profit exceeds funding cost (think rental yield vs. mortgage payments). If your profits are higher than your costs, you are creating value. This approach addresses both failings of DCF - the terminal value isn't as large as in a DCF, and your are explicitly thinking about your balance sheet costs.

On the downside this approach is conceptually harder to grasp than a straight "how much profit am I making" DCF approach. Also people tend to get a bit tetchy when you start talking about book value of assets. This is particularly true tech companies which (if they haven't made stupid acquisitions) tend to have low or zero assets on their book. Indeed some software companies which take a lot of maintenance payments upfront can actually have negative book value of assets! Then again if using book value as a benchmark is good enough for Warren Buffet, then its good enough for me.

How to value a Bloomberg

So which approach to take - Multiples, DCF or EVA?

As I've said I'm trying to move away from multiples and do some more fundamental valuation work. Ideally I've do a full-fat EVA analysis but unfortunately I've got one big problem - I haven't got any balance sheet data (or much financial data at all, for that matter!). I had brief hopes of excavating some historical net-asset-value data from the Merrill Lynch annual reports (ML used to have a 20% stake in Bloomberg so occasionally reported the value of this stake). However when I turned to the reports there was very little. Prior to 2007 the value of the stake is lumped together with a whole load of other stuff. The 2007 report itself records a $373m carrying value for their stake (p116) which would imply an overall book value of $1,865m for the group's equity. However this sounds suspiciously low (I figure it's about a year's worth of retained earnings), so this looks like a dead end.

So unfortunately we're back to good old DCF analysis for all its failings. Natch.

This is this what I found (click on the pic for a more zoomed-in view):

Looks nice, huh? But as I said don't be fooled by all the spurious accuracy. There is, unfortunately, a vast gap between the data I'd like to have and the data I've got.

But this is where I'm coming from:

• Revenue growth: I'm figuring zero growth this year (not a good year for anyone in the capital markets, but Bloomberg as been pretty resilient in the downturn), going back to up roughly 10% mid-term growth. For context Bloomberg has grown revenues at a 10.1% CAGR over the last five years and a 11.1% CAGR over the last ten. I have a 6.7% CAGR over the next five yeras and 6.4% over the next ten. Note this also includes growth from recent acquisitions (e.g. BNA) and expansion into new verticals (law, sport, politics etc.). If acquisitions are fully funded from the operating cash pile (and the cost is fully reflected in earnings - which it often isn't!) I see little difference between buying and building business. Ultimately its budgeting decision.
• Operating margins: I assume margins take a little hit this year before going back up to 35% mid-term, inline with Bloomberg's historical levels (or as much of them as I can make out). That's about par for a single-digit growth incumbent software/processing business - viz Oracle, Microsoft.
• Depreciation & Capex: Precious little to go on here, but I note that Thomson Reuters Markets division reported Depreciation at 7.4 - 7.8% of sales in 2010 and 2011. So I'm assuming a similar lever (7.5%) for Bloomberg. I assume capex is 110% of depreciation - i.e. Bloomberg is spending more on new assets than it is depreciating old ones. That seems sensible if you are assuming revenues continue to grow.
• Tax rate: The only reported tax rates for Bloomberg (in the mid-2000s) are very low (around 10%). This is probably due to various shenanaghans around its private company/limited partnership status. For valuation though its important to think about what tax rates will be in the future. I don't think this special private partnership structure will continue forever (and low historic tax rates may just reflect taxes deferred to future years), so I'm just whacking in a full "normal" tax rate of 30%.
• Working capital: For inventories I am assuming all users have a physical Bloomberg terminal and Bloomberg needs to replace these once every three years (plus terminals for new sales). I assume they hold two months of physical terminal inventory on hand on the balance sheet. For receivables I assume customers have 90 days to pay up (likely to be lower actually, given Bloomberg charge a monthly fee for their terminals) and so Bloomberg carries 90 days of billings on their balance sheet. I assume zero payables to suppliers (a conservative assumption as any payables would reduce the working capital burden). This gives me working capital equivalent to 34% of revenues on the balance sheet - for my forecast I assume that ratio remains constant.
• Net cash: I assume Bloomberg has net cash of $11.1bn on its balance sheet. This is basically the free cashflow from the last nine years cashflow I have reconstructed, minus $990m paid out in 2011 to acquire BNA. Obviously I add this back to my DCF-derived Enterprise Value to get to an Equity Value.
• Cost of capital: Financial academics have long and involved discussions about the appropriate cost of capital, WACCS, levered and unlevered betas and suchlike. I take the Indiana Jones approach and use 9%. In my experience that's the right cost of capital for a growing tech company (as a rule of thumb below 8% is too low and above 10.5% is getting punchy).

Capiche?

Anyhow this may all sound like financial jargon to you - but this is honest-to-God the sort of thing a Wall Street analyst uses to lay out their valuation (minus the lack of financials of course). And if you want to delve more I've whacked my full Bloomberg model (lol) up on Google Docs at the following link. Feel free to download and play:

Bloomberg Model (for what it's worth...)

A nice price for a Bloomberg

So here is the summary of my model, excerpted from the table above. It gives an Enterprise Value (the value of the ongoing business) of $35.4bn and an equity value (the value of the business + the cash on balance sheet) of $46bn. That's in the same ballpark as the $38-42bn number I put up before, maybe a little higher.

Looking at this with my nuanced analysts eye I would say the DCF number looks sensible, although its perhaps a touch on the high side. This is where I used valuation multiples to cross-check. I also show the EV/Sales, EV/EBITDA and P/E ratios implied by the DCF valuation. Note that we are pretty much all the way through 2012 now so the 2013 and 2014 numbers are the ones you should be focusing on.

As I said before, as a rule of thumb I figure EV/Sales runs at about 10x operating margin. So a 4x+ EV/Sales on a 35% operating margin is a little high. The P/E of 19-22x also looks rich given this isn't a massively high growth company. Note this is going to be distorted by the high cash pile (a quarter of the groups valuation) and the relatively low returned being earned on that cash (I assume 2-3% interest rates rising to 4% in the longer term). Net cash piles tend to inflate P/E ratios (this is a problem AAPL also has).

It's probably better to look at the EV/NOPAT (cash-adjusted P/E) of 15.9x - 14.5x. That's similar to the P/E ratios for other large tech companies; for comparison Oracles trades on a current-year P/E of 12,7x, SAP 18.1x and Microsoft 10.6x and IBM (the world's second-biggest software company dontcha know) of 14.4x. Of course you will find much much racier multiples for some cloud computing companies (and as I've said there's a strong argument that Bloomberg is the world's biggest cloud company), although bear in mind these companies tend to be much smaller and have a much longer "runway" of revenue growth ahead of them. Bloomberg is at a much more mature stage of growth than that.

Or to put it another way, if I was running the IPO of Bloomberg and wanting to put out the seller's view of valuation. I think this is where I'd come out at.

Oh and while I'm on the topic of the IPO, if Michael Bloomberg does have 88% of the equity then that implies 12% is in the hands of employees and other founders (hello Mr Secunda!). I make that $371,000 of equity per employee, or  $248,000 if you exclude Mr Secunda. Lucky chaps!

Why the US Election is already over... (and how to arbitrage it)

A quick post today, which isn't really about technology (hmmm, do trading systems count?) but just something that I thought up this week which you might find interesting. After all, isn't a risk-free arbitrage the holy grail of investing?

The US election is already over

US politics is the world's greatest spectator sport. Big money, a year-round season and you never get lockups. Plus you get the final result in one night, rather than having to wait around for the finalists to slug it out for a seven-game series.

That's the only way I can rationalise the fact that the US takes eighteen months and $6bn to arrive at essentially the same decision that the UK manages for $49m. Given there's no public benefit to be gained from an addition $5,951m of electoral friction cost (unless you are political lobbyist), it must be some hideous example of bread and circuses.

But its a funny sport, because even though the season end is more than a month away, we already know the result.

Barack Obama will be re-elected (probably by a shellacking) on 20th November.

Can we move along now and talk about more important things? (like the fiscal cliff?)

So how are you so sure?

If you want to know why, you just to check the market.

I've been an avid follower of the Iowa Electronic Markets for over ten years. Run by the University of Iowa, its one of the world's longest-running real-money prediction markets. They make markets (For purely research purposes of course!) in a select number of political events - largely US elections and nominations.

So not vastly useful if you're a tech investor, but if you follow the soap opera of American politics its gripping stuff. And its amazingly ahead of the curve. In every US election I can remember its called the winner and called them early (2000, 2004 and 2008). In contested nominations too it showed Barack pulling away from Hillary and Romney sailing away from this year's field long before the Bubblevision experts would admit it. Partly this is actually psychological - if you're a journalist a) you don't want to risk being wrong and b) you want to string the story out as long as possible. In contrast for an investor being wrong's not a problem - as long as you're right more often than you're wrong.

So what's the story this year?

This chart shows the price (in cents) for a futures contrast that pays out $1 if Obama (blue) or Romney (red) wins.

In a nutshell, Romney's toast. And has been toast since, to be honest, around the start of August.

If you see a Bubblevision expert tonight telling you next week's debates are a game changer they're wrong. It's game over.

But here's the cool bit

So prediction markets exist and I think they work. But that's not what's cool.

What's cool (as I found out reading last Friday's FT - yes it takes me that long to write things up) that there are other prediction markets, particularly the one run by Intrade, and they give you different odds on the same event.

The latest price on the IEM for an Obama contract is 79.1c, for Romney 20.2c. After this is, give or take rounding errors, a zero sum gain; either Obama wins or Romney wins (or something so bad happens that the election is called off).

But on Intrade you can buy a contract on Obama for $7.51 and Romney for $2.54, paying out $10 if your man wins.

So you basically have two different markets, making different prices on the same asset!

So here's what you have to do!

• Buy an InTrade contract on Obama for $7.51.
• Buy 10 IEM contracts on Romney for $2.02. Total trading cost - $9.53
• Wait until November 6th.
• When Obama wins you collect $10 on the Intrade contract. Oh and if an act of God occurs and Romney wins you make $10 in the IEM contracts.
• I make that a guaranteed 4.9% gain (55.2% annualised).

Now there's a couple of catches here. Unfortunately IEM imposes a $500 trading limit (it is purely for research purposes after all) so you can't just blow all of grannys inheritance on this one. Plus there is a $5 sign-up fee for the IEM (full details here). Intrade also charges a $4.99 monthly fee (the fee is charged on the 1st of the month, so sign up next Tuesday to minimise your trading costs).

But its still $14.45 of free money!

Assumes you can get current prices on 2nd November when the next Intrade fee cycle starts.

What we have is an lovely example of an inefficient market, where either a lack of liquidity or an inability to quantify value means that an asset is inefficiently priced. What is cute though is that inefficient prices are readily accessible on two separate, consumer-facing markets.

That's all. Now go off and watch Bill Clinton's 2012 Convention Speech. Whatever your political views, its a masterclass in how to make a complex, serious argument in a simple and humourous (but sharp-elbowed) manner.

And it's great entertainment. That's the point isn't it?

What's wrong with Apple Maps (and when they will fix it)

Anatomy of a problem

Since the release of iOS6 last Wednesday it's been a case of Schadenfreude all round. However I've seen little analysis of what exactly is wrong with Apple's mapping apps. What people are missing is it isn't just a case of "Maps is crap", but rather a number of separate issues, some avoidable but others not.

Having covered TomTom and Tele Atlas an an equity analyst on and off since 2005, I think I can offer an informed view on these issues. So what I've done in this post is to simply lay out the problems, identify what's gone wrong and think about what Apple can do next to fix it.

Issue	Impact of issue	Difficulty to fix	Time to fix	How to fix	Example
DATA ISSUES
Poor road detail (TomTom data)	xxx	x	x	Get new data from TomTom	Tickhill
Poor road detail (non-TomTom data)	xxx	xx	xx	Revert to local sources	Lack of Tokyo detail, Grand Cayman land shape, Santo Domingo road names
PoIs in wrong location	xxx	xx	x	Source improved PoI data	Helsinki park, Stadium in wrong place, Starbucks in wrong place
Lack of categories	xxx	x	xxx	Source improved PoI data	Stratford CBD labelled a park
Local businesses not listed	xxx	x	xxx	Crowdsource data, more listings deals	Burger King in church, IBM sock picture
Use old names	xx	xx	xx	Revert to more recent data	Valley Forge State Park name, Holy Ghost Hospital, Aldwych tube
Typos	xx	x	xx	Crowdsource data	Doncaster named Duncaster, Florence Cathedral typo
TEXTURE / IMAGING ISSUES
Weird 3D artefacts	xx	xxx	xxx	Hand edit data, get better mesh	Brooklyn Bridge (OK on Nokia Maps), Severn bridge bad 3D model, End of Oxford Street 3D data
Missing textures	xxx	xx	xx	Get more textures	85 Walmgate, York, England
Low res textures	xx	x	xx	Get newer textures	York, England
Non-matching textures	x	xx	xxx	Manually grade/match textures	York, England
Clouded satellite images	xxx	x	xx	Get newer textures	Colchester
SOFTWARE ISSUES
Poor routing	xxx	xx	xx	Improve / acquire bette algos	Directions to San Jose runways
Poor search	xxx	xx	xx	Crowdsource more searches, improve engine	Antwerp search, Street name gives wrong search, Man Utd FC wrong search

Note: More xxx = Biggest impact / harder to fix / will take longer to fix

What's wrong with Apple's maps? (and where did they go wrong)

As you can see from the table above Apple doesn't just have one issue with maps, it has a number of issues. I've broken these down into three rough categories.

1) Data issues

• Poor road/land data data: Apple maps features a number of cases where roads are non-existent (e.g. the example of Tickill in the UK, flagged in this BBC News article). This also affects coastlines and geographical features (e.g. the shape of Grand Cayman here).
• Missing or inaccurate Points of Interest: PoI is map-speak for a specific location, normally denoting a place of business someone might want to visit. There's a bunch of different problems here. Either they are in the wrong location, wrongly categorised or not there are all.
• Typos in place names: Sometimes Apple seems to fall back on outdated data for place names. Other times there are straight (and embarassing) typos.

I think there are a number of different issues here. For some of the problems with roads Apple seems to simply have poor quality data (for a full list of their sources see this page). This is particularly surprising in developed countries like the UK as they are using TomTom mapping data which should have highly accurate road data. My theory is that they have overlaid data from other local sources (e.g. in the UK they cite the Ordnance Survey as another source) which is less accurate.

In less developed countries the opposite is true. Here they appear to have used TomTom data, not realising that they do not have complete coverage and only include major trunk roads. For example Belgrade in Serbia is missing only has major trunk roads. If you zoom to the city in TomTom's data set you see a similar issue.

Missing or inaccurate Points of Interest comes down to one things - either their business directories (sourced from Acxiom) aren't accurate enough. You can argue about whether this was always going to be the case and they could only crowdsource the data once they were live, or whether they should have gone to more local organisations and get the data. Either way the data either isn't there or hasn't been cleaned up enough.

Typos in names are a subset of this. Like with the first category of road data, I suspect they have been overlaying older less accurate sources onto the TomTom data (which is generally pretty good).

2) Texture/imaging issues

• 3D artefacts: Apple's 3D view, involves overlaying satellite imagery over a 3D mesh. Unfortunately for some objects, particularly for man-made structures like bridges, which leads to some weird 3G artefacting (e.g. on the Brooklyn Bridge). Also where the highest level of 3D coverage (including buildings) ends there is an obvious disconnect.
• Missing textures: For the Hybrid and Satellite views Apple relies on data from Digitalglobe. Unfortunately there are a number of issues with this, in particular missing textures at high levels of zoom, lo-res textures, non-matching areas where different photos are stitched together. Also some textures are obscured by clouds entirely.

Apple's highest-res 3D coverage (which leads to the worst artefacting issues) was acquired with C3, a Swedish company which was also behind Nokia's great 3D city maps. The problem is that the mesh they are using doesn't look high enough resolution to capture man-made objects like bridges, or hasn't been tidied up enough. It is interesting to note that the image of the Brooklyn Bridge shown above the Nokia Maps engine (based on an earlier version of the same tech) renders the bridge correctly - I suspect Nokia's model was tidied up by hand whereas Apple didn't apply that final bit of finish.

The issue with 3D coverage ending is, to be honest, something of a non-issue. C3 coverage is very limited (and I think will always remain so given the requirement to fly over with a plane rather than use generic satellite imagery, and the need to tidy up images by hand). This means there will always be a cutoff.

Missing textures, like missing PoIs, looks again like a straight failure to get adequate imagery. This should be a commodity - I remember viewing top-down imagery on Multimap back in in 2000. There is no excuse for low resolution, missing or clouded textures - either the vendor or Apple screwed up here.

Non-matching areas where photos are stitched together is more forgivable. This is an issue which Google used to have on Google Maps and Google Earth all the time (IIRC they only updated color-graded textures to get round this a year or so ago). Largely a non-issue.

3) Software issues

• Poor routing: Turn by turn driving instructions sometimes take people to the wrong destination (or not quite the right destination - e.g. the runways of San Jose airport).
• Poor search: Searching for an address can give incorrect results. Searching for a fuzzier PoI even more so.

In some ways poor routing algorithms in a v1 product are forgivable. Google has similar problems when its turn-by-turn was first released in 2009. In other ways though this is again something Apple could have nailed from the outset - a quick glance at the number of different turn-by-turn apps (from TomTom to Navteq to Navigon to Sygic) shows that even relatively small companies can get this right. With Apple's resources (and given they were apparently working on iOS maps for many years) you would have thought they could have got this one right.

Similarly getting search right from the off wasn't a given, and this will obviously improve once Apple starts to get a decent amount of user data to work with. But the current algorithm does look particularly crude (whilst searching for Christchurch, Mayfair (London) whilst standing in the Church building I get the Mayfair park in Christchurch New Zealand - surely a simple bit of GPS would have figured out I was searching in London?).

How does Apple fix this?

Just as the problem is really a bunch of different issues, so are the solutions (which I summarised in the table at the start of this post). The good news is there are some easy wins, particularly being smarter with the map data Apple already had. The problem is there are other steps which are harder or will take time. I think its unlikely that Apple Maps will be fit for purpose until well into the New Year. I think some of the bigger problems may not even be fixed until iOS7 hits next year.

1. Be smarter with the assets you have: For the basical map data issues a lot can be improved by being smarter with what they have. The most obvious fix will be to flip to the TomTom data in countries where is is good (basically the US and Western Europe) and prioritise local data vendors elsewhere.
2. Get better assets: As I said stuff like textures and PoIs should have been better on day 1. First thing they need to do is get their chequebook out and secure some higher quality imagery. Similarly with Points of Interest - its clear that Acxiom and Yelp are not enough.
3. Wait the wisdom of crowds: Another obvious route to improvement is to wait for the crowd-sourced data to come in. This will help both underlying data quality and search quality. Apple should act to accelerate this process, for example by making the report-a-problem button a bit more prominent (maybe rename "suggest an improvement") or give it a cooler rebranding (like TomTom MapShare). The problem is this takes time to kick in (although hopefully small business owners will have the sense to register their corner shops ASAP), and Apple will need at least some manual vetting of the data before it goes live.
4. Build bridges: What people often don't realise is that the hardest thing about building a digital map isn't driving all the roads. In fact most digital mapping wasn't originally done with cars (far too expensive) - rather it was done by building relationships with thousands, utility companies, transportation companies and government agencies to get their data. Apple has tried to shortcut this process by going to the existing mapping providers but either the data is not up to scratch or they have failed to integrate it properly. Therefore a longer-term solution will be to go straight to the source. This doesn't mean they have to build their own map from scratch - but rather they need to build those crucial partnerships so they can get the data that matters from all the other stakeholders to build on what they have already.
5. Acquire more mapping assets: Apple need to get better and it needs to get better fast. Now M&A is not (as investment bankers would suggest) the solution to every problem, but it could help Apple here. Jason Perlow at CNet put up a very good post over the weekend outlining some avenues for investment. For my part I would add that TomTom is an obvious fit because it brings Apple the bridges it so desperately needs, and a top-notch turn-by-turn routing engine into the bargain. Yes they currently licence the data but owning it and having deeper access could make a big different to the quality of Apple's offering.

The ball is now in Google's court

After spending a good few weeks getting their head kicked in over the Samsung, trial, Google would have enjoyed the last week.

For them the near-term issue is over what to do with their iOS Google Maps (they obviously have a working app - at worst, the old iOS5 one). To keep it off the iPhone would be fun, but long-term stupid as its clearly in their interests to have the iOS user base going past their web properties and not Apple. My hunch is they let iOS6 users stew for a couple more weeks to appreciate how bad the current version of Maps is, and then release it before Apple has had a chance to make major improvements. That way they get the best of all worlds - they keep the installed base, and leave users with a profoundly negative impression of the Apple's app.

The longer term issue is how to keep iOS users installing Google Apps and using Google Services. I suspect this means they have to start bringing at least some of the features Android users have enjoyed (turn-by-turn, vector-based maps, offline caching) to their iOS app. By finally opening up Google's iOS map to competition, it will force it to improve.

So ironically Apple's failure with Maps may actually help iOS users - by giving them a better Google app.

Bloomberg eats the world

Quick Edit (3rd Oct): As a coda to this series I've put up a more detailed valuation analysis of Bloomberg (if you want the short answer, its $46bn with a wide corridor of uncertainty). For more details check it our here!

Stop. Read. This is important.

Over the past week I've been writing a series on Bloomberg, probably the largest cloud company in the world (and definitely the largest cloud company you've never heard of). If you haven't read them it's worth doing so - if you care about finance, tech and cloud computing you will almost certainly find something in there that will surprise you:

• The biggest cloud computing company you've never heard of: Introducing Bloomberg, its business model and the fact its one of the world's biggest cloud companies.
• The amazing Mr Bloomberg and his magical clouds: Outlining Bloomberg's remarkably successful SaaS and PaaS cloud offerings.
• Mr Bloomberg and his extraordinary moats: Bloomberg has three key competitive moats - its user interface, its messaging platform and its content.
• Inside Bloomberg's secret financials: Uncovering Bloomberg's hitherto secret financial statements, I think this business has an Equity Value of $40bn, making it one of the biggest cloud companies full stop.

In the last post I want to outline where Bloomberg is going to go next. It's increasingly clear they want to move beyond being "just" a terminal provider. That has serious implications for the industry.

For anyone who works in an investment bank, Bloomberg might be about to become your biggest competitor.

Mr Bloomberg has a problem

Overall Bloomberg has had a good crisis. As my analysis of their financials shows, they seem to have been able to grow revenues consistently through the downturn. Even in their worst year (2009) it looks like revenues grew 1%, and terminal volumes were only down 2% (albeit helped by some M&A - stripping out the BusinessWeek acqn presumably revenues would have declined 2009). That's doubly impressive given their business model is almost entirely exposed to Wall Street headcount. By pursuing an aggressive overseas expansion strategy, it looks like they have been able to ride out the worst of the turmoil in their home markets.

Nonetheless compared to competitor Reuters, the trend is impressive. The chart to the right shows Bloomberg's group revenues ($mn) vs. Reuters' Markets Division (NB 2008 is proforma for the Thomson-Reuters merger). While the businesses are not wholly comparable (e.g. Reuters does not operate Bloomberg's physical terminal leasing model), it looks like Bloomberg has taken advantage of the crisis to leapfrog its biggest rival. As a private company Bloomberg has had the luxury of being able to invest heavily into the downturn, at a time when most of its listed competitors are having to cut back in order to hit market earnings targets.

However at the same time the crisis has presented a fundamental challenge to Bloomberg's model. I would characterise Bloomberg's historic model as "money for old rope". Although its terminals supplied a vast array of complex analytic capabilities, in my experience the majority of its users simply used it for 1) news, 2) stock quotes and 3) instant messaging. At the time this was great business for Bloomberg because customers were happy to stump up over $20,000 a year for the privilege. In effect they were buying a Cadillac and then using it to drive to the corner shop.

Of course in a downturn customers get meaner and keener. Volumes naturally fall as they shed staff (Lehman brothers alone used 3,500 Bloomberg terminals), and despite Bloomberg's moats they might - quelle horreur - even consider cut-price alternatives like S&P Capital IQ. If all they are using is news, quotes and messaging, they might be willing to take the competitive hit of moving off the Bloomberg platform in favour of saving a few pennies.

The solution for Bloomberg is obvious. They need to drive deeper into the customer, and get them using deeper functionality which a cut-price competitor like Capital IQ cannot provide. In effect they need to enter new (but adjacent) markets.

The problem is these markets are inhabited by some of Bloomberg's biggest customers.

Bloomberg's solution is to dig deeper

In fact Bloomberg has been pursuing a two-pronged approach to driving revenues. It has been trying to go both deeper and wider into its customers.

• Wider: Attack entirely new markets/verticals, in particular legal research (through its $990m acquisition of legal research house BNA) and government (through the launch of its Bloomberg Government subscription service).
• Deeper: Provide more services in its native finance vertical, such as industry research, corporate access and trading.

This post is less concerned with Bloomberg's move to go wider. That is not to say its unimportant - the BNA move was probably its biggest acquisition ever, and put it into a head-to-head fight with incumbents LexisNexis (owned by Reed Elsevier) and WestLaw (owned by Thomson Reuters). The government move is also fascinating - I am always bemused by the vast sums of money spent on elections in the US (here in the UK we can do a perfectly free and fair general election for thirty million quid). And despite the vast spending the whole Beltway set-up sounds like a bit of a cottage industry (or maybe that's just the West Wing version), presumably a ripe market for a big, data-heavy player like Bloomberg.

However I am more interested in Bloomberg's push to go deeper into the market, as this is where the conflicts are most fascinating (and remember, this blog is all about conflicts). In the past few years I've noticed Bloomberg launching an increasing number of initiatives aimed at increasing its footprint with its financial customers:

• Bloomberg Industries: Over the past few years Bloomberg has been building up a stealth equity research platform, Bloomberg industries. It provides industry and stock research on a wide range of sectors (more than many bulge-bracket investment banks), and has recently gone global. While they do not provide direct Buy/Sell recommendations (to avoid competing against their customers at brokers), the product looks much closer to broker research than conventional industry research (e.g. by IDC and Gartner in the tech space). As they say themselves, the staff are mainly experience buy and sell-side analysts. If you want more of a flavour of the personnel, just do a quick Linkedin search.

Bloomberg Industries research - isn't this what we have Sellside Analysts for?

• GLG tie-up: A prototype of the Bloomberg Industries rollout was its earlier tie-up with Gerson Lehman Group, the (in)famous "expert networks" provider. Again this helps provide customers the sort of industry scuttlebutt previously purveyed by research analysts at banks. In fact given the negative publicity over expert networks coming from the Galleon trial I suspect that Bloomberg have more of this field to themselves, with investment banks rethinking their involvement.
• Corporate access: Corporate access (bringing company management's round on roadshows to current and potential investment) is another staple of the large investment banks, who traditionally act as intermediaries between companies (who are their clients on the investment banking side) and investors (who are their clients on the broking side). Bloomberg are now muscling in on this area. I was particularly interested to see a firm run by an old colleague from Morgan Stanley signing a European alliance with Bloomberg for corporate access.
• Bloomberg Vault: Given the volume of communications which run through its Instant Messaging platform, an archiving and compliance solution was something of a no-brainer. Lo and behold Bloomberg has launched Bloomberg Vault. Having previously covered Autonomy (which had a thriving archiving business with investment banks) prior to its acquisition by HP, I can say this is definitely a great cloud business. It also has the handy advantage of being counter-cyclical - when things get worse the lawsuits mount up and banks suddenly need to cover their ass on the compliance front!
• Bloomberg Tradebook: And of course we shouldn't forget Bloomberg's existing broker-dealer business Tradebook. While this isn't anything new (it apparently had $420m of revenues back in 2007). Given Bloomberg's ability to invest in the business through a downturn without having to worry about public earnings expectations, I'm not surprised to see them making an international push with this business at the moment.

Bloomberg's trading floor - sure looks like a bank to me. Hey they even have Bloombergs!

• Consensus data and financial models: In the last few years as a Bloomberg user I've noticed a marked step-up in the quality of company-specific data which Bloomberg provides. In particularly they moved away from IBES in favour of providing their own consensus estimates. Also the quality of the financial models is much better - they will now give you a spreadsheet where you simply input the ticket of your company and it will download and fully populate a model with full quarterly and annual financial statements. That's invaluable for investment research if you want to do quick-and-dirty analysis on a company you're unfamiliar with - and also short-cuts the normal buyside procedure of pining a friendly broker for their model.

The problem is these moves pit Bloomberg in direct competition against some of its biggest customers.

Bloomberg eat their customers

What's clear about these initiatives is they (with the exception of Bloomberg Vault), they bring Bloomberg into more and more competition with brokers and investment banks.

Note Bloomberg has two breeds of customers. Those on the "Buy-Side" (primarily fund managers) gather assets and invest them. Bloomberg does none of that and doesn't want to. However those on the "Sell-Side" (investment banks and other brokers) exist to provider services to fund managers to facilitate their investment.

Well that's also Bloomberg's job description.

I would imagine Bloomberg's response to this is twofold (and yes I am putting words into their mouth here):

1. We are not directly competing with brokers - note that their Bloomberg Industries analyst do not give buy/sell recommendations like Wall St analysts do. However in reality much of the value of sellside brokers is not tied to their recommendations. Sure some analysts earn a reputation as stock-pickers, but I've seen other number 1 ranked analysts which have been appalling at calling their stocks, but very good at in-depth industry insight. That is precisely what Bloomberg Industries claims to offer (albeit with a slightly more data-driven bias). Plus there's the simple fact that many functions such as Bloomberg Tradebook, roadshows and generating models do compete with what investment banks offer, fair and square.
2. Even if they we are coming after you, actually what we are doing is the mundane maintenance stuff (consensus estimates, maintenance research, roadshows). We are actually freeing your hot-shot analysts from tedious jobs so they can write proper differentiated  research. Unfortunately the reality is that for investment banks it is the mundane maintenance stuff which pays the bills. When I was an analyst I could never quite understand why buy-side firms gave so much of their vote to roadshows - it wasn't particularly hard for the analyst to arrange (it was basically a glorified exercise in logistics) and presumably if you were a big enough fund you could ring up the company and set up the meeting yourself. Maybe it was the convenience factor, but anyhow this is clearly a very high margin line of business for a broker. Similarly with maintenance research - every quarter we would do a bog-standard sector earnings preview which seemed to the epitome of commodity research. But when we were a few days late getting it out clients would ring up asking where was the preview. In effect this is "money for old rope" business.

And Bloomberg want a piece of it.

Taking a step back, the great value of the modern investment bank is twofold. First they provide capital (and therefore liquidity). Secondly they provide a network. They know the corporates, they know the investors they should roadshow to. They know who is selling a stock and they know who would like to buy it.

Now it might just be the times we are living in, but there is a shortage of capital out there. This erodes banks previous advantage - they can no longer leverage themselves to buggery in order to facilitate client business (and/or take bets on the side). But the second advantage - the network - is something that Bloomberg can provide. Through their messaging platform they now reach everyone who matters. If a corporate wants to arrange a roadshow Bloomberg can scour their database of holdings and figure out who owns a company's stock. Add in a dose of predictive analytics and they could probably figure out who would like to buy that company's stock.

As I said, Bloomberg eats its customers.

Maybe the customer's baseball bat isn't as big as it looks

Take one step closer and I'll use it I swear!

Of course the customers still have a great big baseball bat they can hold over Bloomberg. The banks are all big customers of Bloomberg, so in theory if Bloomberg started stepping on their toes too dramatically they could take their business elsewhere. That should - in theory - hold Bloomberg in check.

But then again maybe its not that bad.

Its worth thinking about Bloomberg's exposure to the big banks (who I think are the firms most likely to push bank; smaller firms would probably be quite pleased to see Bloomberg taking chunks out of their bulge-bracket competitors). According to the NY Times, Lehman had c3,500 Bloomberg terminals before the crisis and they were a mid-sized global investment bank. I would imagine other banks would have been larger, but then they would have slimmed down since then. Maybe an average of 3,500 terminals today is a decent proxy. If so then that would imply the nine global banks (Goldman, Morgan Stanley, JPM, Citi, Merrill, UBS, CS, Deutsche, BarCap) have roughly 32,000 terminal between then or just over 10% of Bloomberg's total.

Thinking about it that way, the VAR (Value at Risk) doesn't actually look that bad. After all they couldn't completely cut their spending (Bloomberg is too deeply embedded in their customers to that - remember what I said about their moats). And for Bloomberg while revenues from new offerings might not immediately make up for any shortfall, in the long term it gets them into new markets, and makes their revenues stickier so  (i.e. - it improves earnings quality).

In short this epitomises the Innovator's Dilemma. Bloomberg has offerings which potentially cannibalise its existing business. But there is enormous upside if they could disrupt the market and succeed.

Bloomberg could go even deeper

Actually I think Bloomberg's existing offerings only scratch the surface. I think there's a whole bunch of other areas (both for buy-side and sell-side) they could enter or expand their existing offering.

• CRM: At every bulge bracket I've worked with there's been a clunky, homebrew CRM system where you can click to register you client calls, send voice-mails or blast emails. I've always wondered why everyone doesn't just suck it and move to Salesforce.com. Anyhow if they do outsource then Bloomberg would be the perfect trusted partner to provide this. Heck the majority of client communications probably go through their messaging already.
• Content management: Actually Bloomberg are doing this already (most investors pull pdf notes from Bloomberg rather than from broker websites). But there is definitely room for Bloomberg to bulk up the content management side, add analytics and maybe even get embedded in the publishing workflow.
• Portfolio management and prime brokerage: Offering buyside clients portfolio analysis and risk-managements tools on a SaaS basis is something they do already but could always do more. In particular I was piqued by a story I read this week about Scotiabank expanding into prime brokerage using a third-party IT solution. Prime brokerage (basically outsourced trading, lending, custody for hedge funds) strikes me as precisely the sort of high-margin business Bloomberg would want to get into. With their huge cash resources Bloomberg certainly have the balance sheet to facilitate this, and it sits nicely with their existing broker-dealer business.
• Equity distribution: This is the blue-sky one. Its frequently been noted that underwriting and distributing IPOs is incredibly profitable business for banks. So long as you get the pricing right, there is very little risk (if the stock won't sell you just pull the offering), and historically only the big banks have had the buyside contacts book to find the buyers. Of course there are certain regulatory hurdles here, but its not as off the wall as you think. In my time I've seen a number of forays by bond or money-market dealers into equities (e.g. ICAP, Cantor Fitzgerald). Normally they peter out because the challenger lacks the resources and resolve to stick with it - not a problem for Bloomberg.

Why this matters for investors

Of course as a common-stock equity investor why should you care? After all Bloomberg is private and has always been private. It doesn't affect you right?

I think the reasons should be obvious.

• If you are investing in Thomson Reuters, Reed Elsevier or any company which Bloomberg competes in, you should want to know what their biggest competitor is up to.
• If you are working at an investment bank or at a fund manager you will want to know what Bloomberg is up to, because one day they may be either your biggest supplier or your biggest competitor.
• If you like buying hot tech IPOs you should be doing your research now, because at some point Bloomberg is going to go public. In particular bear in mind that Michael Bloomberg steps down as NYC mayor at the start of 2014. Assuming he's not going to run for the Presidency (he'll be 74 in 2016 - Reagan was only 69 when he was elected) he seems likely to leave public life. This means his Bloomberg stake becomes unlocked and all bets are off. A mid-2014 IPO would be the obvious option.

In short, you should never underestimate Michael Bloomberg.

Edit: One related thought which occurred on the way to the shops this morning. Bloomberg and Linkedin are a match made in heaven. If Bloomberg's pitch is that it can provider a better network than the banks have, then Linkedin gives them the best network in existence. If their strategy is to broaden into other industries, Linkedin gives them that network in any vertical they can name.

Of course its not going to happen at the moment because a) Bloomberg will believe they already have a damn good net work in the financial world (they do) so why pay a take-out premium to replicate it and b) LNKD has a $13bn market cap so would probably cost >$16bn to take out - much more than Bloomberg's cash reserves. But if the stock ever got bombed out (hey, the P/E's only 1050x at present) it would be a mouth-watering deal.

Capricious Apple

Wrong turn

Hmmm. When I wrote about the risk of Apple looking capricious on Monday and said we'd see this more and more, I wasn't expecting it to see new examples of it until at least iOS7.

Seems I was wrong.

I actually got a hint of it last night. I tried to looked up my church (Christchurch Mayfair) on a freshly-iOS6'ed iPhone. While in the building.

Maps then took me to the car park for the Mayfair Swimming Pools.

In Christchurch.

New Zealand.

#ios6apocalypse

Why a "fix" doesn't fix it

Now I'm sure Apple will sort this out over time. Good maps rely on crowd-sourcing for points of interest, so it was always going to be a bit rough at the outset.

But even if Apple "fix" the problem the damage runs deeper than that.

By deliberately downgrading users from the Google Maps to a nakedly inferior offering, Apple have very publically put their corporate interests (get Google off the homescreen) above those of their users (provide them with the best mapping experience).

While that might make excellent business sense, the problem for Apple is their walled garden model is build on an implicit pact with the customer: "I'll let you dictate what I'm allowed to do, because I trust that you are acting my best interests."

Instead Apple are very openly breaking that pact by acting in a cynical and arbitrary manner

In short, they looks capricious.

Facebook - What are you doing??

Boo!

Browsing Facebook late last night, and this was the first thing that splatted up (full sized version here):

Untargeted Advertising

For all Facebook's mis-steps the one thing they should be really good at is targeted advertising. They know more about you then anymore. They know what you like, they know who you like, heck they've even ransacked your address book! (without asking your permission, of course)

Which is why I'm surprised to see them pitching me a Nutri Centre Luxury Goody Bag. Look I'll lay it out straight - I'm an early-thirties middle class male. I like playing video games. I've recently regressed back to childhood and started building model tanks. I'm a (very) occasional roller-hockey player.

I have no idea what the main COLOUR of the label for Wild Rose and Lemon Leaf Deodorant is.

And Facebook know that. But given all that, the best their cutting-edge algorithms can come up with is the fact that a guy I was at uni with over a decade ago likes Tesco (he's a hot-shot PR so he'll probably like anything for a fee), and therefore Tesco should spam me with a generic, utterly untargeted Nutri Centre Luxury Goody Bag?

Now to be fair, it could be Tesco's fault. Maybe they told the Facebook sales rep "Nah we don't want the uber targeted advertising which hits exactly the people who want to buy their products and who live near their stores on the day do their weekly shop. Let's just spam everybody with Nutri-Bags."

And to be fair their ad has good screen coverage (albeit on a low-res 11.6" panel). I calculate that the main ad covers 18.4% of the screen, and the little one on the side covers another 4.6%. In contrast the banner ad at the top of the Daily Telegraph page only covers 6.3% of the screen. But if Facebook's strategy to boost their ARPU is to simply spam four times as much screen real-estate as their competitors, then they have a real problem.

Edit: And my wife also keeps getting adverts for Salesforce.com spamming her mobile Facebook. Why Facebook would think a Soil Mechanics PhD student would be interested in a buying an SME-focused SaaS CRM platform is entirely beyond me... #dataminingfail

Facebook's Innovation Gap

But I think its more likely that Facebook simply haven't got their act together. One problem I identified a while back is Facebook seem to have dropped the ball on the innovation front. Their App Platform has stagnated pretty much since Farmville. And I was always bemused why it took them so long to bring out a (not very good) iPad app.

This feels like the same thing - they should have built a killer kick-ass advertising engine. Maybe they're getting there with the launch of Ad Exchange and a new stealth ad network.

But if this is the current State of the Art for the Facebook Platform, I can see why their ARPU is only a fraction of Google's.

Right back to working on Bloomberg!