Tuesday, October 30. 2007
After my post yesterday on AT&T's kickbacks to Apple where I estimated the kickback to be around $50 a quarter I have come across other people coming to the same conclusion.
A Gizmodo article today talks about another calculation of teh kickback being $18 per month. At $54 per quarter, this is very close to my $50 calculation which leads me to believe that this is right in the ballpark of the actual figure. They also link back to a previous article of theirs I hadn't seen, which calculates the cost of goods for the 8GB iPhone to be around $280. This is not the true cost price, apples cost of sale averages to 25% over COGS, putting the true cost at around $350.
So with this verification and extra information, the profit from a customer that unlocks a phone is aabout $50, while the profit from a customer that signs with AT&T is around $450-$480. If Apple's actions cause them to lose less than 9 sales for every person that goes AT&T instead of unlocking then they come out ahead (financially at least). This doesn't count any lost sales from Brand damage which Apple probably (correctly) assumes will be negligible.
Gizmodo makes a small error in claiming that the data on the kickback was
According to the Financial Accounting Standards Board.... The FASB just made the rules which forced Apple to report its revenue from the iPhone in a certain way. The calculation was done by an analyst from Piper Jaffray as detailed in the NY Times article Gizmodo sources.
Monday, October 29. 2007
It has been common knowledge that Apple must be getting some level of kickback from AT&T for iPhone usage, why else would there be an exclusive deal? This was made 100% certain this week when after having tried to cripple unlockers with their latest code update, Apple took steps to limit resale opportunities for the iPhone by accepting credit card only and limiting puchase to 2 per customer. It is an extreme step to put restrictions on the sale of your product when that is how you make your money. The only reason for doing so is to enable you to make more money.
With Apple estimating that 18% (250K out of 1.4M) of phones purchased are unlocked, if they hit their target of selling 10 million phones within the year, they are trying to not sell 1.8Million units. At the current price Apple may not be making a huge amount of profit on their phones, similar phones sell for around $150 more. However raw profits on high end phones are expected to be higher. Nokia makes an average 22.6% profit on its mobile phones, but higher end phones like the n95 are likely to be closer to 50% or even more. Whatever the real numbers are, I expect that Apple is still making some profit from their $399 price.
To see if I could get some sense of what level of kickback Apple might be getting I turned to Q4 sales data which nicely splits out iPhone revenue. The number here is stated to include accessories and carrier agreements, so kickback proved, but not how much. The numbers on this table suggest that the average revenue per iPhone is only $105. While this doesn't make much sense, we need to look at the more detailed financial statements. On the last page you can see that Apple has deferred revenue across at least the next two financial periods. While the numbers are a bit contaminated by Apple TV figures, my back of the envelope calculations give me an minimum revenue per iPhone of around $525.
To work out what the direct revenue is entails some guesswork. Many 8GB sold for $599 while the rest sold at $399. All of the 4GB sold at $399. A portion of the early purchasers, although probably not all, would have claimed the $100 refund as well. Also taking into account the 18% that never signed to AT&T, my estimate is that the average sell price will probably come to around $475, but to be safe lets call it $500. This suggests that at least $25 per phone sold came from AT&T. While I have no way of knowing the details of the contract, I would presume that AT&T would try and avoid signing bonuses and pay on contract revenue. The iPhone has been on sale for one quarter at the time of these financial statements so this quarters figures contain almost all the payments. The economics trick to calculate a progressive payment over a growing install base is to take half the growth, in this case half the install base, as the base figure. Without going into boring sums, this translates to the actual revenue per phone from AT&T calculating to $50 per phone.
Considering that the average plan price for AT&T is $80 per month, $50 per quarter to Apple does not seem an unfeasible sum, a 20% kickback. This is also a bigger number than the maximum amount of sales Apple could lose by curtailing unlocked phones. Realistically they know they will reduce but not eliminate this, and taking action will keep AT&T on side. Even if they lose 10% of their sales revenue, the 20%+ kickback from AT&T is more valuable.
Now researching this made me wonder why Apple is deferring revenue for the Apple TV? The reason to defer revenue is if you have not delivered the entire product or service, i.e. you have an obligation in the future that is created by the sale. Often a warranty portion of revenue is deferred in order to match it with the cost of that warranty, however the Apple TV has the same warranty as the iPod which does not defer revenue. Another common reason is if the product/service involves a subscription or future service that must be delivered. From the descriptions of the product this is not the case. There is a YouTube access portion that may have a license cost they need to cover, however this didn't start until the end of June and there is deferred revenue for the quarters ending March and June. I have no answer to this, but my curiosity is piqued.
The audited financial statements for the year will probably not be released until the end of December, and those should give a lot more information than the interim quarter results statements do. These may not go to enough detail to achieve certainty about these figures, but I will be looking at these as soon as they come out. I'll update the figures from this post then.
Thursday, October 25. 2007
In case you hadn't seen seen it yet, Apples stock price continued its steady climb to pass $187 per share, which in market capitalisation terms makes it worth around $162 Billion. Unsurprisingly there has been some noise about this in the Apple blogs (iphone-tools, tuaw.com, B&T Wireless) talking mainly about how they are now worth more than IBM and Intel and in fact any other PC maker. This is all well and good, and the analysts are impressed. Mac Daily News even ranks the new analyst target prices, encouraging the support of the $243p/s target price from Bear Stearns. It is always tempting to think of numbers like this as proof of superiority however this actually shows the value of hype over substance.
I should say that I like and use Apple products, and although I wouldn't use an iPhone (here's why) but there is an iTouch and a new MacBook Pro in my future. They make good products (arguably better than IBM) they market well and have great design. They do not have the numbers to make them worth more than IBM and Intel though.
The above table is based on each companies last reported quarter. If we look at the raw numbers of the companies we see that Apple are dwarfed by both Intel and IBM, if we ignore revenue and look just at profit both are more than double Apple's size. While Apple has a larger growth rate for the latest quarter, there would need to be an argument that this rate will be constant over the next 15 years to justify the price.
The mobile phone market is not going to do it, Nokia has quarterly revenues from phone sales of US$8.5B and they get this by having global coverage and hundreds of models in different price brackets. The iPod market is already saturated by Apple. While the PC market will continue to grow for them, Apple has no Enterprise products or penetration and are not geared to compete for the business IT spend.
Looking at a different set of numbers still doesn't show anything that would rate Apple as a $200+ stock. They aren't as heavily leveraged as IBM, but are more reliant on debt than Intel. They have good cash flow at the moment, and there is an investment war chest, but not so huge that it would change the story without looking more deeply into the cash situation (which I will not do here). For a company that makes its money from selling manufactured goods, a P/E ratio of 47.35 is high. Even if you outsource the infrastructure, shifting to follow market trends and launching new product ranges is capital and time intensive, and risky.
I cannot see any good reason to value Apple this highly, hype is great for a company but is always transitory. A single hiccup can remove a lot of exuberance from the market and Apple's market cap will be re-evaluated by its shareholders. While it may look good to some Apple fans for the price to be so high, it puts a great pressure on the management of the company to make no mistakes.
All financial data gathered from SEC filings by the relevant companies sourced from their investor websites.
Monday, October 22. 2007
I was reading the Mashable blog and was drawn to Peter Cashmore's article on internal linking. I myself rarely link to previous articles in my postings, mainly because there seemed to be a negative stigma about blogs that do so. It was good to see the perspective of a much more experienced blogger on how he deals with the issue.
This problem stems in large part from the lack of good statistics coupled with the amount of money that is starting to appear in the blogging space. If we look at the demand side of web advertising, very large amounts of impressions are being recorded, which if even remotely true are a potential windfall of marketing dollars available for blogs to attract.
Without access to the server logs it is not possible to get accurate numbers of how many visitors a site actually has. And with RSS readers it is hard to know whether your subscribers actually read you, or if they more likely only read articles that interest them. In this situation the good marketers will take effectiveness into their own hands. Running small test campaigns that track the actions rather than the 'impressions' allows them to build up a picture of the effectiveness of a blog in reaching people. But this does not help the blogger as they don't see this information.
We then see how Google can be so successful. They can offer a sales method where the payment can be based on an action rather than a nebulous viewing of an ad. They can aggregate demand and still put some context into the ad placement by applying their search capabilities. While this is great for Google though, marketers would be better served in having greater control of their ad placements, and bloggers would be better paid if advertisers dealt more directly with them.
This is not going to happen anytime soon, the technical problems are hard to solve and the vested interests are too great. I'm going to feel more relaxed about linking internally though, but only if there is a point. I posted in August about how unreliable my web-stats were.
Monday, October 15. 2007
I have been getting familiar with Mahalo recently, as the possibilities of moderated search are interesting. Lead by CEO Jason Calacanis the tagline of the site is “We’re here to help”. This sums up completely the appeal of a moderated search engine, having someone to clear out the extraneous sites for me. The other important benefit that can be added by Mahalo is the provision of context for the results of a search. An editor can provide a level of summary about the intent and feel of a discovered site that can not be achieved with robot generated search results.
To analogise this, it is like finding a hotel in a new city you are intending to visit. You could get a copy of the Yellow Pages business directory for the city and look up hotels. A better solution is to get a Lonely Planet style guidebook and find a short list of recommended hotels with a summary of the feel and style as well as the features. The latter method not only makes it easier to make a decision, it also allows you to be more certain about that decision once made.
These key differences dictate to Mahalo the direction they should take to offer most benefit to users, and this is the area that I do not think the team has put enough thought into. While my information on their business plan is limited to what I see in the Mahalo and Calacanis blogs, what I see when I go to the site leads me to my conclusion. Let me talk about where they should be and then look at why they are a little off track. I will do this by looking at the two key benefits of moderated search and where these best fit
Summary of best sites: In most cases not being able to find relevant results in an Internet search is due to specificity. If I cannot produce a descriptive enough search string then I will get too many, mostly useless results, however there may be very good reasons why I can’t. If I am not knowledgeable enough about either the search methodology, or the topic I am searching then searching on the Internet will be either frustrating, fruitless, time consuming or any mix of these. A moderated search can help me here, but the areas where this is likely to be relevant are specific.
To illustrate, if I have just purchased my first SLR camera and want some tips on how to best us it. A Google search turns up 123 Million results. A lot of the results on the first page are probably relevant but not certain. A Mahalo guide by contrast gives me a good set of starting sites categorised so I know which are likely relevant to me.
Context: What the Mahalo photography guide gave me that was more important was context. It groups and labels the results so that I know what to expect from the link before I follow it, and I have a level of confidence in this information because it is the editor, not the site owner that has provided the contextual information. In the how to section I can see an “amateur” label which gives me direction to a beginners site, and in the “sharing communities” section I can see where to see examples of others work.
This directs Mahalo to concentrate towards subjective subjects. If my question has a concrete answer, then a regular search engine will do as well. If, however the answer is relative to my environment, point in time or intent then providing context helps me. This is the more powerful hook of the two advantages of moderated search.
As to why I am uncertain whether Mahalo is capitalising on through looking at their site. If I look at the front page it is dominated with pop culture, current events/people and zeitgeist. These topics have wide coverage with little variation so finding them through alternative means is trivial. For example, if I want to see the video of Andrew Meyer being “tase’d” Mahalo offers no greater relevance or speed to my search than Google, Yahoo, Live, etc.
The front page appears to be determined by popularity of searches, and if so this would match with what the Top 50 search terms show, which is mostly Halo searches. While Mahalo cannot directly control what its users search for, if they are going to succeed it will be if search terms they offer extra value on become more dominant. If Mahalo do their job well and give that value, then the top search terms will change as it attracts the users that derive value.
Jason has stated that his main goal is to work out how to achieve breadth of coverage. My suggestion would be to work out the subject areas and topics where Mahalo offers the most benefit over standard search and concentrate on getting that the best it can be, then expand to further topic areas that make sense. On the site and in Jason’s blog the stated goal is to cover the “top 20,000 search terms”, but popularity of search term is not the prime selection criteria if the goal is to help.
Needless to say I am very interested in how Mahalo develops over the coming months.
Scott Carp on Publishing 2.0 posted a piece on Facebook’s lack of potential as a business application. I agree with his point inasmuch as large corporations are not going to embrace the Facebook service as it stands. The potential of the technology as it stands, and in fact many Web 2.0 technologies, is very high within business. In fact, I would contend that this is a major reason why Microsoft is so interested.
One of the challenges for businesses as they grow is to maintain communication flows within the organisation. It is no wonder that email has become one of the most important applications to most business’, it is the primary tool for communication between staff. I would safely say that every company I work with would like to have better knowledge sharing, which is why Knowledge Management applications are a growth segment of the software market.
A Facebook style interface would work fantastically as a front end to corporate communications and knowledge management.
So the same way that Facebook maps out your social connections, a Facebook style front end could map out your intra-company connections, and could probably be extended into partner organisations. It could provide an easy to use interface on disparate information systems, and this could help to increase the adoption of their use. This type of system also provides a tangible map of the companies information network. The business can also track what type of information is being sought in a better way and increase communication and interaction with line staff.
Even just looking at some of the applications that are available on Facebook at the moment. I’ve stated before that I have yet to see any that I would classify as a ‘killer app’. If I look at their potential in a business context though, the opinion changes. If you are on Facebook, check out one of the many TV show fan applications (e.g. “Addicted to The Simpsons). These apps have tabs for news, forum, quotes, trivia and a home-page. This format would translate fantastically to a product training page. News and forum stay the same, quotes changes to success stories, customer quotes, etc. and the trivia page could easily be reworked to a training page.
Other applications, like some of the Myers Briggs style personality tests could be used to map things like knowledge levels, opinions on products and strategy, or varied things of HR interest across the organisation. The nature of the application allows this data to be easily collated across different networks and groups within the company. Like any knowledge management or communication tool, the results are defendant on the level of adoption, however this is not a unique problem to this style of application. The ease of use, and a few social and/or fun apps being allowed in could help a Facebook style app to have better adoption rates.
In short I can see a clear way that Facebook could be used in a business setting, which is one reason I am confident Microsoft won’t blow any money they spend on them. I briefly thought about how I could build and sell this type of application myself, then realized that Microsoft already have one that is close – Sharepoint. Sharepoint has most of the individual components, but in my opinion is missing the glue that makes it really usable. Could this be why MS are so interested? I would confidently predict a Facebook style front end to Sharepoint in the very near future. If Steve Ballmer has his way, they will build it rather than buy. Even if they don’t use the Facebook code though, a business relationship will be very helpful in negotiating those tricky IP licensing issues.
Wednesday, October 10. 2007
If you are a GNC reader you would have seen Todd's link to this article already, if you haven't and are interested in the music biz it is a great read. It is also a great example of a well put together presentation. Presentation skills are one of the best business skills you can have. Even if you are the most brilliant person in your company, club or society and have the idea that will solve world hunger, without the ability to inform others you will not suceed.
What Ian did really well with this presentation
- Had a simple and well defined message.
- Only included points that built to his message
- Did not include more content than the audience could digest, usually no more than 3 ideas
- Simple slides without exposition that summarised what he was saying
- Graphic examples to highlight particular points
- Could have included a lot more points, but limited himself to the key points that built to his specific conclusion
- Ended with a call to action.
- He knew his audience well and tailored his message to them, including using points that were most likely to get their attention (revenue)
A strong introduction that states the conclusion is the one structural element missing that would ideally be there. The best way to get your main point across is called the 'three tells'. Tell them what you are going to tell them, tell them, then tell them what you told them. Giving the conclusion at the beginning causes the audience to frame their own conclusions in the context of yours. Without this step you lose some control of how your message is interpreted. When you then deliver your conclusion, audience members that have followed your logic to a different end point will feel conflict with your conclusion and devalue your points by association. You can limit this by stating your conclusion at the start.
I wouldn't mark Ian down for missing it in this case though. Part of understanding the rules is knowing when to break them. Considering the conclusion of his presentation (in the bluntest of terms) is "You are a pack of idiots and I don't want to work with you until you come to your senses", stating this up front may have been worse than any logical drift in the audience. If you know your audience well you can build a better presentation.
Some key points when building a presentation
- The three tells
- Start with the conclusion, what is the statement you want your audience to walk away with
- No more than 3 key takeaways, the audience won't remember any more
- Simple slides, as few words as possible, they are meant to highlight the key words not give your audience reading material.
- The right pictures are often better than words
- Throw away anything that does not directly support or build to your conclusion
- Know your audience and include points that are relevant to them
- Do a course, Rogen, Toastmasters, or a variety of others can help you build your skills and point out problems you didn't know you had.
A couple of web resources
Presentation skills on Managementhelp
Thursday, October 4. 2007
It feels like a slow news week. While there has been no slowdown on my RSS feed, I have struggled to find something that sparks my interest enough to research. I have a story half written on Capitol v Thomas, but seeing as the trial will probably wrap up this week I will wait until the result is known before finishing it off. Making the best of a slow time I started work on porting the site over to another blogging software. Need I say that this has turned out to be much harder than I expected and I wasn't expecting easy.
My goal was that if I moved, I wanted to keep comments and trackbacks, and to ensure that all the entries kept their time stamps. The time stamps were no problem, but the rest was. This drove me to do some searching on how other people were administering their serendipity sites to see what my other options were. In doing so I found an upgrade version with new features, and a raft of extra plugins that I hadn't seen before.
There are now 2 upgrades to the site. 'Karma' is a post rating system that allows readers to rate stories up or down. I am sure you are all very familiar with this style of technology. I am hoping this will increase the feedback I get so I can tailor the posting style to suit my growing readership. The second one literally saved Serendipity, an autosave feature for the entry screen. This allow me to use the embedded entry tool again without fear of post loss. Writing externally, then cutting, pasting and editing was getting old fast. I have also found an XMLRPC plugin that should let me post from BlogJet, which would be my preference once I get it working.
Next task is to find a better statistics system.
Monday, October 1. 2007
There has been a lot of press about Microsoft's plans to buy a 5% stake in Facebook for between $300M and $500M US. One particularly good analysis of the potential deal can be found at ITPro. This deal would value Facebook at around $10B, and apparently Mark Zuckerberg is hanging out for a $15B figure. Most of the analysis I have seen uses relative modeling to come up with a position on the reasonableness of $15B. This invariably comes up short, especially when no-one is looking to pay anything like as much for Facebook’s rivals. MySpace was bought outright for $580M about 1 year ago, and other web 2.0 sites have not received speculative valuations anywhere near this.
In an absolute valuation it can be hard to see how a company with $150M in revenue, and a guesstimated $30M in profit can be worth $15B. I have been running some ‘back of the envelope’ numbers though, and I am not convinced Mark and Microsoft are insane. If I use a rule of thumb, and say for an unproven technology company I wouldn’t want to see a price/earnings ratio (P/E) of more than 15. This means that the company will make its value in profit over 15 years. Using 15 also makes it easy to work out that for a $15B valuation, Facebook would need to average $1B in profit each year.
This is probably not making anyone feel any better about the valuation yet, until the miracle of compound growth is applied. To get an average of $1B profit over 15 years, facebook needs to grow at an average of 44%pa. If they could maintain that rate, in 15 years Facebook would be generating $5B profit per year (non discounted) and would have averaged the $1B per year. 44% is not a high target for a technology company, there have been many technology companies that have achieved this for sustained periods: Dell, Google, Apple, Compaq, etc. But let me now drop back to relative valuation to test whether 44% growth and a P/E of 15 are reasonable assumptions.
I choose Google as a comparison for two key reasons, it has a demonstrated track record of what is possible, and like facebook its revenue predominantly comes from advertising. Google has currently 9 years old, which is not quite the 15 year window I have chosen for Facebook, but it is already generating around $4B per year in earnings. This proves that $5B a year after 15 years is possible. At this level, Google is still growing at 58% per year, which definitely suggests that 44% is more than attainable. As for whether a 15 P/E is reasonable, Google is currently valued at $177B and has a P/E of 48. This is not a directly valid comparison as Google has many years of proven growth and performance, making it a safer bet. However, a P/E of 15 does not look unreasonable.
Given all this it is clear that it is possible for Facebook to achieve the growth rate necessary to justify an asking price of $15B. It is still a very large gamble for a purchaser to take though, as there is no track record yet to prove that this business has legs and is not just a fad. Facebook has also to prove that they have a business model to capitalise on their popularity, although making a profit is a very good first step. Om Malik puts it very well in his recent posting where he talks about Facebook needing to control their own ads. To get into ad sales they need people, which means they need money. Microsoft don’t necessarily care too much about this though.
Companies like Microsoft buy other companies for many reasons, they only invest in them for one, competitive advantage. They should not be concerned with making a profit from selling their minority stake in the future, that would make them an investment house, and this is not the way companies should use their money. Microsoft want either:
If there are specific technologies in Facebook that Microsoft likes the look of, but they are not yet certain of, then investing gives them a chance to get them preferentially in the future when they make up their mind. If the deal values Facebook at a risky level (like $15B) then this disincents other companies from investing, or buying out Facebook before the technologies are proven. This still risks the $500M they pay, and even though this is only a fraction of Microsoft’s money it is not chump change. There must also be contingency plans for the risk of losing their money. That is probably advertising. If this deal goes ahead, I would expect Facebook to sign up to Microsoft’s advertising system.
Facebook have revenues of about $150M per year, all of this from ad revenue. Even given a very good deal, Google adwords is currently getting at least twice this. Presuming that the Google cut is the same $150M, you can see how Microsoft getting this revenue could pay for their stake in Facebook. With even moderate growth of 30% Microsoft could make $600M in ad revenue in 3 years, not to mention any extra business they could get from marketing their relationship to Facebook, and on Facebook.
So given all this, while a $15B valuation seems high for Facebook, it might not be, and Microsoft's investment is really a quite safe bet. They are almost certain to make their investment back in extra revenue, block others from investing or buying Facebook out, and possibly net a nice windfall if the company succeeds.
(Page 1 of 1, totaling 9 entries)
Syndicate This Blog
Geek News Central
Tuesday, March 28. 2017
Tuesday, March 28. 2017
Monday, March 27. 2017
Monday, March 27. 2017
Sunday, March 26. 2017
Sunday, March 26. 2017
Sunday, March 26. 2017
Friday, March 24. 2017
Tuesday, March 21. 2017
Tuesday, March 21. 2017
Last entry: 2008-08-14 05:17
59 entries written
44 comments have been made
2156 visitor(s) this month
1 visitor(s) online