Atomic Energy

Most businesses – and non-profit and government organizations, for that matter – fiercely protect their turf. In the case of business, it usually means not doing anything that might jeopardize the core income stream. For example, Microsoft has been loathe, at various stages in its history, to move towards the Web, cloud computing, or anything that might put a dent in its core operating system, desktop software and business server businesses. This is understandable. If your business made $58 BN in revenue from these core products in the last year, anything that would reduce dependence on your products, and thus that number, by providing more cost-effective solutions to your customers, would feel like suicide to you. The beauty of our market-based capitalist system, is that even if the incumbent players or regulators cannot (or do not want to) see a better way, someone else can, and can successfully sell it. It is probably fair to say that the only reason Microsoft released any such products is due to competitive threats, at various times, from Netscape, Yahoo, Google and others.

Clayton Christensen, in his brilliant Innovators series of books, calls this disruption of the market. In simple (and overly simplified) terms, you cannot beat a large incumbent at their own game, but if you can disrupt them in a way that changes the rules of the game, provides compelling benefits to customers, and moves in a way that the large player cannot – due to perceived gross margins, cultural inability to shift, existing relationships, or protection of a core market – the underdog can and often will win. Christensen recommends that incumbents can avoid this trap by disrupting themselves out of business. Essentially, they need to become their own competitors. In business terms, we call this a type of cannibalism: you cannibalize your own market-share by selling a product or service that is better for customers than the existing ones, by creating your own internal (or external) disruptive start-up. Of course, the devil is in the details, and few firms can succeed at doing so.

Every now and then I come across a business, large or small, that successfully cannibalizes its own business and grows because of it. Yesterday, I met a small Web design shop, just a handful of employees, that nonetheless cannibalized its own business, at least partially, and impressed me.

In most respects, Web Design Insight is like every other quality custom Web design shop out there. They have design talent, programmers, and provide high-quality service. However, unlike most such shops, they saw a pattern, saw a competitive opportunity, and decided to grab it themselves. WDI does a lot of work with synagogues. Synagogues are, as most readers know, Jewish houses of worship that also function as community centers. In general, they use the Web to publish information about themselves, whether static “we are Congregation So-and-So,” or more dynamic weekly updates. In addition, they have the usual activities of most non-profits, especially religious ones: keeping in touch with members, raising funds and membership dues, etc. Finally, they have the unique needs of a Jewish house of worship: language, as most Jewish ritual is in Hebrew and English; calendaring issues, as Jewish ritual revolves around the Hebrew calendar, distinct from the Gregorian one in common civil usage; unique lifecycle events, such as birthdays on the Hebrew calendar, bar/bat mitzvahs, etc.; times of prayer that vary weekly based on season and location; and many others. Thus, what a “normal” non-profit might be able to purchase for $15-20k, a full-service Web site for a synagogue would require significant amounts of custom work, boosting prices to multiples of that, perhaps as high as $50-60k. (These prices are my estimates; I have no actual pricing information from WDI). For a Web design shop, this is a book: lots of custom work means lots of consulting, with gross margins on each hour sold. Yet the husband-wife proprietors of WDI decided to cannibalize their own business. They built a product called Synagogue Launcher, which likely sells for a fraction of what a custom solution would. They took the common work in all of the synagogues, did it once in a generic fashion, and sell it as a synagogue platform. In doing so, they have robbed themselves of their own consulting revenue, i.e. cannibalized their core business, to create a new one. They disrupted their own business, took a big risk, and hopefully are reaping the benefits. They deserve credit… and if only the larger businesses could do so as well. At the same time, that would leave little opportunity for all of the entrepreneurs out there, so we should be grateful.

On the reverse side, I heard Esther Dyson speak 3-4 years ago. Esther is a fixture in the tech community. She also sits on the board of the advertising conglomerate WPP Group. As part of her address, Esther noted that WPP had recently acquired a small, brash interactive media player (whose name escapes me at the moment). When asked about the rationale for the acquisition – to be blunt, large companies tend to smother the creativity of smaller dynamic firms that they acquire, making acquiring a small firm whose success depends on its culture highly questionable, see in dictionary: IBM and _____ (you fill in the blank) – she said openly that they acquired the firm for it to be a catalyst for change in WPP. The world was moving rapidly towards dynamic, interactive, hyper-local advertising, and WPP needed to change. She (and assumedly the WPP board) viewed this acquisition as the kernel of change, the grain of sand that irritates the oyster leading to the pearl. In essence, WPP was trying to acquire a disruptor, rather than create one, but in doing so, it actually wanted to disrupt. I found two points very notable:

• WPP recognized it needed cultural change, and openly admitted it could not do so on its own. Whether or not it had tried and failed, they didn’t say. This public openness is refreshing, and is good.
• WPP’s board is incredibly naive in believing that some small, recently acquired group could effectively change the culture of a company with 140,000 employees across 107 countries. This naivete is bad.

So we have three examples:

• Microsoft, who refuses to disrupt itself until forced to, and poorly at that, and only then after it fails to crush the disruptors.
• WPP, who recognizes the need for change, yet tries to buy it from the outside, and shows naivete.
• WDI, who openly and willingly cannibalizes its own young business to serve its customers.

The world is always changing. The smart players are, like Gretzky, going to where the puck is going to be, not where it is.

Last week, Ars Technica had a great in-depth article on ENUM. For those who do not know, ENUM is a standardized method to use Internet technologies to translate your old, staid, telephone number (e.g. +12125551212 or +442076555555) into an Internet device connection. That connection could be simple voice, such as allowing you to use a VoIP connection rather than going through the PSTN (public switched telephone network) that we all love/hate, or even identifying a Skype user ID or email address based on the telephone number.

ENUM really has two purposes, which are related but are not exactly the same. The first is evolutionary, the second more radical.

1. Free from the carriers: If you enter a telephone number, you are essentially telling your phone company or carrier, “go to this country, this area code, lookup this number, find out which carrier holds the number, then connect to them and through them to a particular phone.” You are connecting through carriers, the same way you always have done. But now, you have VoIP phones. You are no longer directly bound to a traditional carrier; you may not even be bound to a carrier at all. There are well-defined ways of publishing your VoIP phone contact info over the Internet. ENUM allows you to publish a number and say, “hey, connect to me through the Internet at this VoIP address, which is how you would get to me.” You are setting yourself free from the carriers (at least to some extent). In the (almost) words of Sting, “if you love some(ph)one, set them free.” This is evolutionary, it assumes phones are pretty much the same as they always were and serve the same purpose, but we want to have a better, freer, cheaper route to get to them.
2. Universal identifier: Somewhere down the line, Internet people began to realize that we have a lot of identifiers. We have one or more email addresses, Skype IDs, Gmail IDs, Yahoo IDs, the list goes on and on. Wouldn’t it be great to have a single identifier that is me but can access all of the other parts? In essence, we are looking for something like a really smart email list. Just like I can set up an email list mygroup@atomicinc.com and have it send the email to 5 different email addresses, I want a single identifier that will allow anyone to reach me, and is smart enough to say, “I am trying to reach you via email, tell me which way to go,” or “I am trying via Skype, which way?” Since everyone has a phone number, the theory goes, why not use the phone number as that unique identifier, and have a system, let’s call it something really catchy, say, “ENUM,” to translate from the person’s unique identifier to the various methods of reaching them.

The first method is slowly gaining some ground, but not very much. e164.org, probably the largest and most well-known Public ENUM provider, says it has just under 47MM allocations. Considering that the total possible numbers under current numbering schemes is just under 100BN, and the US alone is estimated to have close to 300MM phone lines, these numbers really are tiny. e164.org does not publish how many are actually in use. There are two great challenges to adoption, one from the user side, one from the carrier side. From the user side, it is useful for dialing in, terrible for dialing out. Let’s explain why. If I am calling you, I don’t need to publish my ENUM, I just need yours. In having yours, I can get a much cheaper route to you. But you gain nothing from it. Similarly, when you try to call me, it is really useful to you if I have published my ENUM, but I gain nothing; you are the one who saves on all the carrier interchange fees. The net result is that the only real incentive to publish ENUM is, well, none. Who is left? The carriers. The carriers, who probably provided you with a telephone number, like these interchange fees. Sure, it would be nice if someone contacted them directly, but in the interim, everyone is getting a small piece of the pie, with the whole pie provided by the call originator. All in all, not a great recipe for them to publish ENUM either. Finally, as pointed out in the Ars Technica article, ENUM explicitly puts control of the number in the hands of the end-user, i.e. the phone subscriber. Needless to say, the carriers are not exactly going to be thrilled about relinquishing control.

From a universal identifier perspective, I do not believe ENUM will take off, due to sheer usability and control issues. From the control perspective, most numbers are still controlled by carriers. Unless and until it becomes easily possible for someone to purchase (not lease, rent or otherwise) a telephone number and directly control it, carriers will be the gateway for telephone numbers, which means users will not view them as their own. Yes, you can buy low-cost DIDs from many forwarding and ITSPs, but these are still owned by them and they remain the gateways. As long as this holds true, people will not view them as their unique identifier, but rather tied to whatever phone line they have, not to themselves. Additionally, in many countries and especially the United States, people do not want to be globally identified by a number, with apologies (but not too many) to former New York State Governor Eliot Spitzer. People would prefer obscure email addresses over even more obscure telephone numbers. Finally, it is easier to remember steve@apple.com than some long-winded telephone number. And while it is true that .com implies (but does not insist) on a US focus, +16055551212 is very clearly a US number. People are more global nowadays. On the other hand, steve@apple.com is a work address, and people change a lot. Further, people don’t want some material going through work, whether because it is a waste of work time, personally sensitive or any other reason.

The real question, then, is whether any unique global identifier would work. The answer, in my opinion, is a qualified yes. I think many people would be happy to have a unique identifier, that I can simply give someone my ID, whatever it is, most likely in email format, and they can use it to get everything about me. It is qualified, because I believe people want to maintain control. I don’t want every spammer to have my email address, I want business associates to have my work email and phone but not my mobile, and I may not want my home contractor to have my work number. I believe that if someone came up with a unique identifier system that could be used for real-time lookups across systems, while providing reasonable and non-burdensome access control for the owner of the data, it would likely be successful. Whether or not it can make money, and how it fits with existing social networks, is an entirely different topic.

Ask VCs, most will tell you that Information Security is, well, old. It has been around for a long time, and a lot of money is flowing into newer, hotter areas. While I cannot fault investors for looking for areas that will have the best combination of future follow-on investment, rapid growth and a high-value relatively quick exit, I believe InfoSec is still overlooked.

When following information security, I feel like I am listening to the scientist who said, at the turn of the century (right around the time that an obscure patent clerk in Bern was writing obscure papers that might have a minor impact on physics), that “Physics is Dead.” Perhaps it is similar to Francis Fukuyama’s “End of History” proclamation.

Information Security is not dead, it is live, it is hot, and it is a great sector to be in. InfoSec is great for the same reason that rifle design or tank design is great: the other side is always getting better. InfoSec is the combat divisions in a war, which, if you want to win, need have the best intelligence, be well-equipped, and always be ready, or the enemy at the doorstep is likely to conquer. Every organization in the world now has significant online activity, including funds and banking. Every one has levels of security behind which hide very sensitive and valuable data. If Willie Sutton were alive today, he wouldn’t go to the banks because, “that’s where the money is;” he would go to the laptop. Given that access is just as easy for a 14-yr-old from a low-level-law-enforcement country as it is for a 20-year veteran in the United States, and that the 14-yr-old likely has a far less developed sense of scruples or morals, the list of people willing to become the “digital enemy” is almost unlimited. Thus, the defenses and weapons (and thus budgets) that legitimate organizations need to deploy is constant and growing. And all this is before one takes into account the various compliance requirements such as HIPAA, PCI-DSS, SEC, Sarbox, BASEL II, etc.

I believe several areas will be particularly hot in the coming years:

• Authentication: Mail accounts, Web logins, cell phones, credit cards, the list of methods of identity theft is growing by the day. I believe that organizations, by desire or by force, will look for much better methods to authenticate each individual and partner than the standards available nowadays. The challenge, of course, is that most better methods available are intrusive and expensive, e.g. authentication tokens like RSA, or sidechannel SMS one-time codes. I find the keyboard typing pattern firms, like BioPassword, interesting. Nonetheless, they can still be spoofed, and are too sensitive to the person’s behaviour, such as typing while eating a burger, or using the left hand because of a cast. I am awaiting a new, better two-factor authentication that is harder to spoof, less sensitive to legitimate divergent behaviour, and relatively inexpensive.
• Transactions: Transaction changing, known as man-in-the-browser (MitB) attacks, (why are these called man-in-the-browser? I am fairly neutral on this issue, but if we are all equal, then it cuts both ways) are particularly nasty and hard to beat. In essence, all of the work we do to secure our transactions – encryption, digital signatures, etc. – all rely on some complex mathematical functions or relatively secured memory tokens like cookies. These are far too complex to perform in our heads, so we rely on our computers (and browsers) to do them for us. MitB attacks is like having the bad guys already in your house; the alarms and bars on doors and windows won’t do you much good then. A better method of securing the browser is one alternative, while authenticating each transaction – which is back to the difficult and expensive if we don’t trust the browser to do the complex work – is another. Neither of these, in its current implementations, is going to take off anywhere near as quickly as the evil they fight, namely MitB.
• Authorization: In small organizations, authorization is easy: either I am allowed to do something or I am not. In large, consumer-facing organizations, it is not that much harder: each person has an account and can do anything to that account, and only that account, that any person can to their own account. Internally to midsize to large organizations, however, authorization is particularly messy. Permissions can be controlled by who you are, your job, your relationship to a particular piece of information, your relationship to someone else, the list goes on. Role-Based Access Control (RBAC) has helped a lot, but still has limitations. These limitations by definition mean organizations put in place workarounds, which open up security holes. I believe we are going to see growth in better user authorization management that is both easier to use and more secure.
• Cloud Security: Cloud services are great: you basically can buy resources on demand and use them. However, they open up security holes that need to be addressed. While some services sweep the issues under the rug (bad), and others tell you what they are (better), for a firm with data and transactions that really need to be secure, whether because of their own internal needs or due to compliance, cloud is not an option. For example, right now, you cannot be PCI-DSS compliant on most cloud infrastructure. I have done lots of PCI-DSS work, and it just is not possible. I foresee much work on security in the cloud, both from the providers like Amazon and Google, as well as third-party providers, to provide compliant security in the cloud. I do not know if this means a new secure cloud provider, or security wrappers to the existing providers, but this area will grow.
• Trust the Untrusted Providers: Right now, if you use Google, they have access to your data; ditto for Yahoo, Hotmail, and all of the others. The same holds for Software-as-a-Service (SaaS) providers. For many people, “trust us” is sufficient, or perhaps a privacy agreement, or some other form of protection. For many, however, there is the same problem as general cloud infrastructure services: they are not willing to depend on the current statements or good will of the service provider, and thus are locked out of these services. I believe that new, secure services will start to appear to allow those organizations to take advantage of the general services like Gmail, Salesforce.com, etc. Once again, I do not know if these will be additional services from the existing providers (more likely) or third-party providers (less likely, especially in the case of those who make their money from knowing the content, like Gmail, but still possible), but there are too many organizations out there who are still doing Exchange/Lotus to ignore.

In sum, InfoSec is still a great area, and, in my opinion, will continue to be for the foreseeable future.

Everyone appears to have analyzed the Google and China issue to death, both from the China perspective and the Google perspective. One of the best comments I have seen is from Fred Wilson’s “A VC” blog, where he focuses on the reported fierce disagreements between Eric Schmidt, the hired gun non-founder CEO, who wanted to continue doing business in China and sweep the issues under the rug, and Sergei Brin, the founder born under Communism, who insisted on pulling out. Fred goes on to expand to the general differences between founder CEOs and non-founder CEOs. I suspect the late great Akio Morita of Sony fame would agree. When I was in business school at Duke, we looked at Sony under the wave of MBAs who ruled the roost after Morita-san, and the lack of innovation therein, in a company that invented and successfully marketed many of the great innovations of the second half of the 20th century.

I suspect, however, that there is more to the story than either the realists or idealists (or Jacksonites vs. Wilsonians, if you prefer foreign policy) may be seeing.

In its early years, Google was not only a much better, faster and more accurate search engine than Yahoo (or MSN, AskJeeves, Excite, AltaVista, and many of the competitors in the digital dustbin). Its motto was “Don’t be evil.” After some of the DoubleClick and other privacy debacles – some of which are laughable today – as well as the debate in the United States about government intrusion into calling records from major telecoms providers, there was great hunger for a company whose entire personality seemed to be, well, “behave.” Google’s foray into China, with its self-censorship, has definitely damaged its image in that respect. However, Google is too big and dominant to care that much.

Now, however, Google’s growth has slowed, other competitors are moving into the market, and, most importantly, mobile is the hot new area. Google wants its android, and possibly its HTC-partnered phone, to be major elements in its future growth plans. However, privacy in the old Web 1.0 or even 2.0 is child’s play compared with privacy in mobile. Your phone operator – and likely operating system provider, especially an online one like Google – knows who you called, where you are, where you were. It could probably even figure out if you were linked via Bluetooth to a car and the speed and route the car took; any local law enforcement would love to be able to boost revenues issuing tickets that way.

If Google really wants to dominate in the mobile area the way it has on the Web, it is desperately in need of regaining its “Don’t be evil” image. While I have great respect for Brin, and he undoubtedly does remember life under the Communists, Google’s plans for future growth are likely a major factor in its China decision.

This article is a short follow-up to our discussion of the low profitability of Israeli firms. I recently had the pleasure of sitting with the Chairman of a very well-established Israeli firm, not one of those discussed explicitly in the previous article. I raised the issue of Israeli profitability with him. His perception was that Israel, being a small country, is subject to an excess of information. You could almost name it the “too-efficient market” hypothesis. Because each customer knows how the costs of each supplier, including salaries and margins, customer squeeze suppliers down to bare-minimum gross margins. As he put it, “we have a pricing problem.”

I take issue with these comments for several reasons, but specifically note the insight it provides into at least part of the Israeli business mentality.

1. There is no such thing as too much information. Buying Dell servers in Israel is no more transparent than buying consulting services in Palo Alto. Any good executive knows, or can quickly find out, the variable and overhead costs of each and every supplier. If anything, the US market is actually more transparent than the Israeli one, since the size of it provides many more opportunities and a broader data set. Outliers are less likely to dominate or skew.
2. Many of the companies that have higher margins – and the only ones I can directly analyze without inside information – are publicly traded. These companies issue annual reports, and thus have all of their costs, at least at a high level, publicly available.
3. Many of the Israeli companies with poor margins actually do the majority of their business internationally. Thus, their customers are the exact same customers, with the same pricing negotiations, as their US or European competitors with higher margins.

What I find most notable, however, is the insight this provides into the mentality of at least part of the Israeli business market. Many Israelis, due to their culture, are natural salesmen, doers, dealmakers. As such, they focus on constantly making a sale, and have the salesman’s mentality. This is great for volume and top-line revenues. Unfortunately, if unrestrained by a CEO with an eye on profits, can lead to many low gross-margin deals. This focus on sales at the (possibly unconscious) expense of profits leads to higher revenues, low profitability, and poor return on investment for investors.

Israeli companies appear to be trapped in the mindset of “deal at any price.” In order to grow into profitably, Israeli firms need to learn the investor’s profit mantra, how to measure deals based not on their revenue but their margin, and learn to differentiate their products and services sufficiently, at least in their customers’ views, so that they can charge rates with sufficiently high gross margins. Their problem is one of both marketing and sales.

In the prior article, we analyzed the likelihood of Oracle meeting their comment, at their fourth quarter 2008 results conference, of having Sun add $1.5BN in operating profit to their bottom line within one year of closing the acquisition, essentially in fiscal 2009. The conclusion we reached was that it was a highly aggressive target, and unlikely to be reached by half. Nonetheless, if Oracle could get Sun halfway to Oracle’s gross margins and fixed costs as a percentage of revenues by 12 months from acquisition, they would get fairly close.

In this short follow-up, we will look at the political drivers behind Oracle’s $1.5BN announcement.

According to the April 20, 2009, announcement, Oracle agreed to pay $9.50 per share for Sun’s stock. As of this writing, the total number of shares issued and outstanding for Sun is 746.25MM, for a total value of the transaction as $7.1BN, which differs slightly from Sun’s announcement that the value is $7.4BN. We will ignore that discrepancy, as it is not material to this article. After accounting for Sun’s debt and cash, the net cost to Oracle is $5.6BN.

Sun, like Oracle, is in a fairly risky business. In its latest annual report, Sun ranges the estimated weighted average cost of capital, and hence discount rate for in-process R&D (IPRD) for its acquisitions, as between 12% and 22%. As some of these are much riskier projects than Sun as a whole, we will stay in the middle and assume a discount rate of 17%. Using a basic perpetuity schedule, for Oracle to break even on its $5.6BN investment in Sun, and ignoring the actual cost of closing the deal – legal fees, filing/regulatory fees, investment banker fees – all of which can add up to well in excess of 10%, as well as the operational costs of merging the businesses, Sun will need to add almost $1.0BN ($952MM to be precise) to Oracle’s bottom line.

Now, let us add 10% for closing costs, and another 10% for operational costs, for a total deal cost of $7.1BN: 10% of the transaction cost of $7.4BN (not the net cost of $5.6BN – those investment bankers are in a very good place), another 10% for operational costs, gives about $1.5BN in additional costs. Add that to the net costs of $5.6BN for $7.1BN in total cost to Oracle in the first year. Returning to our 17% discount rate calculation, and the Sun acquisition needs to $1.2BN to Oracle’s bottom line. Of course, Oracle’s managers do not want to have taken all of this risk in the deal, just to get the same return as if they had done nothing. This is especially true during an earnings report where they did not do quite as well as they would like. As we recall, the headlines indicated that Oracle was “not immune” to the recession. Clearly, many had expected it to be. Thus, it was critically important that Oracle’s managers show not only that the Sun deal would break even, and in one year, no less, but that it would actually do 25% ($300MM on top of the $1.2BN break even is 3/12 = 25%) better than break even, and thus return higher than its cost of capital.

To be fair, it would be reasonable to also use Oracle’s weighted average cost of capital and discount rate, and those that apply to the company as a whole, or perhaps its internal projects, rather than its acquisitions. The only discount rates provided in Oracle’s latest annual report is for its BEA acquisition, where it uses discount rates of 7% to 17%. For a tech company, this is quite low. The Corporate Executive Board, in its November 17, 2008, report, indicated that the typical company expects to see WACC rise to between 9% and 12%. A tech company such as Oracle or Sun should expect a good few percentage points higher, given the volatility and riskiness of the business; anyone remember DEC? when Dell was the undisputed leader? when Yahoo ran search and there was nothing new in that space? Thus, our 17% rate will suffice.

Conclusions

Oracle should, and possibly could, achieve significant improvements in Sun Microsystems’ bottom line, by bringing its margins more in line with Oracle’s. This is in addition to any reductions in cost due to redundant activities, as well as additional revenue due to cross-selling opportunities. Its $1.5BN is challenging but not absurd. However, it is highly likely that its selection of this number target was driven as much by public/investor/political relations incentives as actual financial benefits.

A colleague of mine in Israel, Yishai Boasson, recently wrote an article analyzing the perceived and actual value of a “special promotion.” This article, available here, is only in Hebrew. As it is an interesting analysis and an excellent example of the importance of distinguishing between sales and profits, it is worth translating here for the larger English-speaking audience. It is also, in itself, an example of how numbers can trip you up.

Original Analysis

In honour of Israel’s Independence Day, which this year was on Wednesday, April 29, 2009, the large supermarket chain Rami-Levy decided to have a special promotion of a half-kilogram (about 1 pound) of Kebab plus one kilogram (about 2 pounds) of hummous for just one New Israeli Shekel (NIS). The exchange rate has been fluctuating, but we can roughly estimate it at 4NIS=1USD, which means you could but a pound of meat and two pounds of hummous for $0.25. The apparent value of the kebab plus hummous special is around 30NIS. It is unclear from the original article Yishai quotes whether this is the retail price, i.e. the value of the gift includes foregone profit, or if this is just the variable cost, i.e. the value is just the direct cost to Rami-Levy. Either way, the supermarket chain is believed to have brought in 100,000 customers, each of whom spend about 200NIS, for total revenues of 20MM NIS.

As a general rule, supermarket chains run on very thin margins. Yishai decided to be very generous and assume they are operating at 10% margins. If so, then the profit from this little promotion was 10%*20MM NIS = 2MM NIS. Of course, Rami-Levy also gave up 100,000 customers * 30NIS per special package = 3MM NIS. The net result is that this little promotion brought in 2MM NIS in profit, at a cost of 3MM in marketing.

First rule of thumb in marketing: if your marketing costs exceed the profit generated by that marketing, it was a bad idea. Spending 3MM NIS to get 2MM NIS in profit is a good way to go out of business fast. Let’s only hope they don’t decide to “make it up in volume.”

Digging even deeper, the picture gets worse:

1. It is not at all clear that the 100,000 customers are actually new. How many of them would have come to Rami-Levy anyways, even without the promotion?
2. How many customers did Rami-Levy lose? While many people do come in for special promotions, many others are unwilling to deal with the mob, and are just fine paying the extra 30NIS (or $7.50) for their kebab and hummous, rather than fight the madhouse. Worse, those people who are willing to forgo the special to avoid the lines are precisely the customers that are most likely to spend more money at the supermarket.
3. Rami-Levy’s margins are undoubtedly below 10%, with some analyses putting them at around 6.5% or lower.

This analysis correctly highlights that the value of any investment is not the sales return from the investment but the profit return from it. Spending 1MM to get 10MM in sales may or may not be a good idea… depending on the gross margins of those 10MM, and alternative uses of cash.

Other Benefits

Several commenters on the original article pointed out that Rami-Levy may have had larger goals, that may have justified the net 1MM (or more) loss. These include: savings on general marketing or growth of a certain department. Additional values may have been: savings on launching costs for a new store or series of stores, and clearing out inventory. Although margins are low, inventory that might have been close to end of shelf life are worthless unless they can be cleared. A 5NIS loaf of bread is worth 0.5NIS to the bottom line at 10% gross margins… but it is worth a full 5NIS if the alternative is to completely discard it.

The Flaw in the Numbers

Last, but not least, there is one strong flaw in the analysis: the low margins are operating, not gross. While supermarkets do not normally have very high gross margins either, they tend to be much higher than 5% or 6%. We can analyze Publix, a large US-based supermarket chain, Publix, which is a public company and therefore has a publicly available 10K. In 2008, it had $24MM in revenue and $17.5MM in COGS, for an operating margin of 27%, over 4 times the 6.5% operating margins of Rami-Levy. Publix operates on even thinner operating margins than Rami-Levy, despite being a much larger chain, and thus likely has lower gross margins as well. Thus, it is not impossible that Rami-Levy actually lost quite a bit less, or broke even on the promotion, depending on what its gross margins are.

Conclusion

In sum, we see that numbers are very powerful but seductive. In our case study, we needed to watch for two things:

1. Don’t confuse revenues with profits when calculating return on investment.
2. Don’t confuse operating margins with gross margins when calculating profitability.

As any manager of a (successful) business will tell you, the two most important factors are:

• Cash – as in “cash is king”. You need cash to pay your payables. If you do not have enough cash, you cannot meet basic obligations, let alone invest in expansion. Cash generally comes from sales, particularly from converting receivables to cash. Never confuse the accounts receivable asset with the cash asset. Although there are receivable-based financing options, most of which are quite expensive, you normally cannot pay your bills with a receivable. Try telling your engineer that, this week, instead of direct deposit of her paycheck, she will receive a receivable from customer X! Additionally, cash can come from investors, to support expansion or growth in a time of negative profitability, until profitability is reached.
• Gross margins – gross margins provide the excess over costs of sales (for services) or costs of goods sold (for products) to cover fixed costs, not to mention create operating profit for the owner(s).

Norm Brodsky, in his columns in Inc magazine as well as his book “The Knack” hammers home, again and again, how important gross margins are. As an aside, one of my favourite lines is the story of the manager who, upon being told that each product is being sold at a loss, i.e. negative gross margins, says, “no problem, we will make it up in volume!”

The two major factors that go into COGS (in the rest of this article, we will use COGS and COS interchangeably) are:

• Product costs – materials, labour, transportation, etc.
• Sales – commissions, channel costs, etc.

Assume Linksys sells a router for $100. If the cost of materials (plastic, chips, antenna, power supply, shipping, manufacturing) is $50, and they pay a 30% channel commission, i.e. $30, then gross margins are $20. Obviously, besides increasing prices, the two primary methods of increasing gross margins are:

1. Reduce manufacturing costs, through one or more of: more efficient processes, lower-cost materials, better economies of scale, etc.
2. Reduce sales costs, either through having each sales unit (e.g. salesperson) on a fixed cost sell more product, or reduce commissions to sales units, i.e. reduce the commission rates.

Clearly, the choice of sales channel, and its costs, has a significant impact on gross margins and, hence profitability.

This insight is very important when analyzing how to structure a new business. We will analyze two potential similar businesses, both based on actual business plan case studies, to see how these costs impact the choice of channel.

Business Overviews

Both of our businesses sell security software. To protect their privacy, we cannot disclose exactly what are of security, but suffice it to say that it is software that is usable only to businesses. The software is very similar, except that company S sells to small-to-midsize businesses (SMB), while company E sells to enterprises. S’s SMB targets have average revenues of $20-100MM; E’s targets have average revenues of $250MM to $BN+. SMB businesses are likely to have, at most, 1,000 employees. Enterprises will have, at minimum, twice that, and most will have 10,000 or more. Given the value of security at the larger scales, the minimum price point of E’s sales at an enterprise is at least $100k, and maximum sale price of $250k. By contrast, S’s minimum price point is $10k, and maximum sale price of $25k. Given the different market sizes and complexity of usage, it is not surprising that S’s minimum and maximum sales are 10% that of E. Both S and E wish to follow the traditional salesperson model. A salesperson, in both cases, makes direct sales calls, and closes each deal.

Enterprise

Given the complexity of E’s customers – both organizationally and in terms of deployment – it is unsurprising that E prefers the inside sales model. This model actually works quite well for them. Following our assumption of the minimum price point of $100k, our salesperson is earning a 20% commission. In the real world, some earn more, some earn less, but this number works for our purposes, and is close to the average assumed in the actual business studied. Our salesperson, being a star, wishes to earn at least $200k per year. Given that his commission is 20%, that means he needs to bring in $1MM per year in sales. Since the minimum sale size is $100k, he needs to make 10 minimum size sales per year, or just under one per month. For a good salesperson, this is eminently reasonable. 160 hours (4 weeks * 40 hours per week), is a reasonable amount of time to spend on closing a large deal. Additionally, since the maximum sale size is $250k, this salesperson likely will earn significantly more.

Enterprise Summary: Required minimum of $1MM per year in sales, for 10 sales per year.

SMB

S actually would prefer to avoid using a salesperson. S seriously investigated using distributors and other channels. However, these channels generally take 30% (sometimes more), and, more importantly, tend not to stock a product until there is significant customer demand. Given the higher COS (and hence lower gross margins and profitability), as well as the chicken-and-egg problem of distributors not taking product until there is demand, but no ability to generate demand without some sales and reasonable cash flow for marketing, S falls back on the same strategy as E: sales staff.

We follow our above assumption that the minimum price point is $10k. As with E, our salesperson earns a commission of 20%. Our salesperson, of course, also wants to earn at least $200k per year, which means at least $1MM per sales each year. However, since the minimum price point is $10k, our salesperson must make not 10 but 100 sales per year. With 52 weeks in the year, she must sell two products every single week. This gives her 20 hours to sell each product. This is extremely difficult, if not impossible, to do as a salesperson reaching out to customers. Even with the maximum sale at $25k, she would have to make just under a sale a week, if every single sale were at the maximum.

SMB Summary: Required minimum of $1MM per year in sales, for 100 sales per year, an unachievable target.

Alternate Paths

Clearly, our SMB is in a bind. If it uses distributors, it has no viable channel until customers demand the products. If it uses sales staff, it can either (a) increase commission rates to retain staff, thus killing gross margins and profitability or (b) expect to lose sales the staff, with a high turnover rate and an inability to attract savvy salespersons, as a quick analysis shows it to be physically impossible to make enough sales to earn the desired amount of income.

What does the SMB do, then? It has several possibilities:

• Raise investment funds, which will allow it to unprofitably invest heavily in marketing and inside sales for the first one to two years, at a loss. At that point, sufficient demand should exist in the marketplace to allow it to use other channels that will be more profitable than a dedicated sales staff.
• Increase prices significantly, essentially moving into E’s turf, or somewhere around the top of its market and the bottom of E’s.
• Use Web distribution channels. This idea was raised by Mike Baird, author of “Engineering You Start-Up“. Mike’s argument is that in order to reach SMBs, you need to engineer and simplify your product sufficiently that it can be explained – value and installation – on the Web, and thus marketed and sold through Internet channels.

Assuming the SMB does not want to change markets because, for example, it has identified a large enough and strong enough niche, or the E space is already too competitive, and it does not want to or cannot raise investment funds, then the third choice is the best. Additionally, from a market product perspective, this is the best choice. Enterprises have large and complex environments, with advanced and highly-paid IT staff. Enterprises are the ones that need and can use complex products. SMBs rarely have such large staff, and normally have a fairly simple environment. The product from S must meet all of the needs of the customers, but must do so as simply as possible. Thus, not just for sales and distribution reasons, but also for core market reasons, Mike’s point is accurate.

Conclusion

Inside sales is rarely the best choice for a sales channel, unless all of the following are true:

1. The price-point is at least $100k per sales, and preferably more.
2. The organization to which you are selling is complex and requires a lot of handholding to get the deal done.
3. The environment into which your product will be installed is complex technically, with advanced staff, and requires customization and/or advanced features for proper installation.

No, this article will definitely not be a Bible sermon. However, it will look at the success levels of two fairly well to-do Israeli companies, and try to get a bit of understanding as to why the land of the Jewish Prophets is not as successful as it might be with Israeli Profits.

Before we begin, it is important to note that this article will not delve into macroeconomic policy. Policy leads to politics, which is absolutely necessary, but writing about it, in this context, is not good for business. We will focus solely on two public Israeli companies, as well as certain macroeconomic data about the State of Israel.

First, the overall view.

• Israel has a population, as of the end of 2008, of 7.4MM people (Israel Central Bureau of Statistics).
• Israel has a GDP of $202BN in 2008 (IMF).
• Israel has a GDP per capita of $28,365 (IMF). Yes, these numbers don’t quite add up, due to a slight variance in how the IMF calculates population, and its averages over the year, as opposed to point-in-time census.

By comparison, the United States has a GDP per capita of $46,859, and the UK has a GDP per capita of $43,785. On the other hand, Israel far exceeds its neighbours’ GDP per capita, with Egypt at $2,161, Syria at $2,757, Jordan at $3,421, and only Saudi Arabia coming somewhat closer, largely due to oil revenues, at $19,345.

The interesting question is why Israel, with its highly educated workforce, intense innovation and entrepreneurial drive, and multilingual and multicultural workforce, as well as a network of expats around the world, has a GDP of only slightly more than half that of the US and the UK. Interestingly, even in highly educated industries and roles, for example C-level executives, engineers and top academics, the income scales are also normally one third to one half of those for comparable roles in the United States, ignoring the very highly paid executives of American multinationals, the equivalent of which largely does not exist in Israel, due to its small population size, approximately 1/28th that of the United States or the Euro zone.

Rather than delving into macroeconomic management (or mismanagement), although much of that does exist here, we will take the microeconomic view. Many of the larger Israeli conglomerates, like Koor and Klal, are vestiges of the statist/socialist days of the Israeli economy, prior to the economic liberalizations of the 1980s and 1990s. Much of the growth engine of Israel in the last two decades has been due to innovative firms in what is called “high-tech” in Israel, the software, security, hardware, Web, biotechnology and energy sectors. As such, let us investigate two firms that are distinctly “high-tech”, have been successful enough to live on their own revenues and even go public, analyze their results, and determine what insights are available.

The two companies we will explore are AudioCodes, manufacturer of Voice over IP, or VoIP, hardware and software, and NICE, manufacturer of enterprise and security information technology products.

For the years 2005 through 2007, the last year for which publicly released financial information is available, AudioCodes had revenues that grew from $116M through $147MM to $158MM, an annual growth rate of 26% in 2006 and 7.5% in 2007. These are undoubtedly respectable growth numbers, especially in a sector that has major players in the market, such as Cisco, Nortel and other technology behemoths. Now, let us turn to operating profit. In 2005, operating income was $12.5MM, or 11%. In 2006, operating income was $4.3MM, or 2.9%. In 2007, operating income was ($4.2MM), or -2.7%. Immediately, we can see that even before the recession, as AudioCodes grew its revenue impressively, its operating margins shrunk and then turned to a loss. Further, even before the shrinkage, in 2005, operating margins were at 11%. By contrast, in 2005, competitor Cisco had operating margins of $5.7BN on $21BN in revenues, or 27%. While one can argue that VoIP products are a small subset of Cisco’s business, it is highly unlikely that Cisco would go into any business where it could not foresee maintaining its margins.

If we look at NICE, we see a similar pattern. NICE is much larger than AudioCodes, with 2005 through 2007 sales of $311MM, $410MM and $517MM, respectively. Once again, however, its net income is $32.1MM (10.3%), $17.1MM (4.2%) or, excluding a one-time in-process R&D write-off, $30MM (7.3%), and $33MM (6.4%). Like AudioCodes, despite healthy and growing revenues, and a scale that is sufficient to be solidly profitable, NICE is stuck in low operating margin territory.

Two questions arise from this analysis:

1. Why are these two companies operating with such terribly low margins?
2. What impact does this state of affairs have on the health and robustness of the Israeli economy and GDP per capita?

The answer to the first question, of course, is that it takes some real time on the inside of the company – top-down financial and operations analysis and bottom-up walking the floor, talking to staff and customers, and seeing what is going on inside – to understand why. Nonetheless, there is no reason for these companies to be operating at such a low-level of profitability.

The answer to the second question is, “a lot.”

• GDP is defined as the total income of the economy. If these businesses, engines of economic growth in the most dynamic sector, are having such low income, then GDP is directly reduced. With a constant population base, GDP per capita is reduced.
• With such anemic returns from even successful companies, investors will be hesitant to invest in them, or other similar companies. This reduction in investment directly leads to lost growth opportunities and reduced GDP. With the same population base, GDP per capita, again, is reduced.
• With very low returns for the same staff size, company profitability per employee (i.e. productivity) is much lower. The employees themselves may or may not be less productive than their US counterparts. However, for the company as a whole, profit per employee is low, and hence both the willingness to invest in more employees and the desire to pay them more, is reduced, directly reducing the human component of GDP and GDP per capita.

As we can see, despite the miracle of Israeli growth over the last two decades, and the brilliance of its entrepreneurs, many of its more mature companies are focused too intensely on revenues, at the direct expense of profitability. This is not a shortcoming unique to Israeli managers; most entrepreneurs, owners and CEOs are top-line driven rather than bottom-line focused. It is for this purpose that COOs exist. However, the pattern is likely having a negative impact on Israeli GDP, GDP per capita, investment and salary levels. Improving these numbers using real operations expertise is crucial.

In the years 2006-2008, Google earned, respectively, $10.5BN, $16.4BN and $21.1BN in advertising revenues. With net income of $3-4BN each of those three years, it is hardly surprising that “free supported by advertising” became the model of choice for many an aspiring startup, as well as many non-startup businesses. During the latest mini-bubble of venture investing, it was nearly impossible for an online company to get an investor’s meeting if your business plan was not based on the presumption that your services would be free and paid for by advertising.

In the 12-18 months since the investing bubble collapsed, and investors went into a deep freeze not that different from what occurred in 2001-2002, the value of the advertising model has been severely questioned. Whether using CPC (cost per click) or CPM (cost per thousand impressions), the rate for each of these has gone down. In some cases it actually has gone down; in others, the assumptions of CPMs and impressions (or CPC and click-rates) was dramatically overstated. In January, AdAge reported that a study from the Interactive Advertising Bureau and Bain Capital was showing CPM rates on ad networks of $0.60 to $1.10, with their own direct sales of $6-10. The report indicated that (a) those 10x multiples for direct sales were likely overstated and (b) online publishers were only able to unload around 1/3 of their ad space inventory directly anyways.

Assuming the high CPM of even $1.00, an online services company that wants, say, a small $1MM in revenue, needs to have 1BN impressions per year ($1.00 CPM means 1,000 impressions makes $1.00, times 1MM to make 1BN impressions). Over 12 months, that is 83MM impressions per month. If we assume that each unique visitor visits once per week, and sees 10 pages, and each page has 3 ads, then each unique visitor sees 120 impressions. For 83MM impressions per month, this company needs nearly 700,000 unique visitors each and every month, just to clear $1MM per year in sales. Of course, these numbers change if:

• Like most sites, visitors see fewer pages per visit, reducing revenue.
• Visitors visit more or less frequently than weekly.
• The site has some strong unique niche, like ClubMom, that can command higher CPMs.
• The site does not have a strong unique niche, and thus gets much lower CPMs.
• The site does not have 700,000 unique visitors per month, a fairly high number.

Essentially, just to clear $1MM in annual revenues, the site needs to have many regular, loyal, high-usage visitors.

Conversely, if the site managed to sell to each of these unique visitors at just $5 per month (the model used by backup services, like Mozy), then it would generate $3.5MM per month in sales with the same number of visitors. Those kinds of numbers could even attract professional investors. Clearly, if you can sell instead of using a three-way (service provider, user, advertiser) model, it is far more revenue-intense and, likely, profitable. Of course, paying customers demand a whole different level of service, but $42MM per year can pay for a lot of customer service.

Two interesting case studies came out in the last several months that illustrate this concept.

• YouTube vs. iPhone vs. Kindle: On June 15, 2009, Business Insider published a report by Goldman Sachs analyst James Mitchell. In it, he showed how YouTube revenues were essentially on par with iPhone App Store and Amazon Kindle eBook sales in 2008, while eBooks should double YouTube revenues by 2010, while iPhone App Store revenues will be quadruple YouTube revenues in the same year. It is important to note that YouTube has 71MM unique monthly visitors, according to its own information, and has been around for over 4 years. The iPhone, by contrast, has been around for 2 years, with the App Store somewhat less, and certainly nowhere near open enough to encourage heavy usage until into 2008. The Kindle launched a few months after the iPhone, and is thus even younger. Essentially, one of the hottest properties on the Internet, with an enormous number of visitors, cannot match direct e-sales products half its age. Interestingly, both iPhone Apps and Kindle eBooks are of the same essential format as YouTube videos: 3rd-party content. In the case of YouTube, it is user-generated videos; in the case of the App Store, it is overwhelmingly 3rd-party developers; in the case of the Kindle, it is 100% publisher-author provided books. If we compare users, as stated, YouTube has 71MM unique monthly visitors, while the total number of iPhones sold by Apple is slightly over 20MM, and the number of Kindles is almost certainly below 1MM. With fewer than 1/3 the users, and half the life, iPhone App Store and Kindle eBooks are handily putting YouTube in its place. Finally, notice that we are discussing revenues, not profits. Video is notoriously expensive. It requires a lot of bandwidth, even with advanced encoding such as mp4, and a lot of processing and memory. There is no doubt that the fixed and variable costs for YouTube far exceed those of the App Store and Kindle eBooks.
• iPhone App advertising: A great company, based in New York, is Pinchmedia. Pinchmedia is one of several companies (another excellent one is Medialets) that provide in-application advertising and metrics (akin to analytics such as Google’s Analytics) for iPhone applications. Earlier this year, Greg Yardley, the founder and CEO of Pinchmedia released a fascinating study based on the analytics his service has provided in 30MM downloaded applications. The report is available here, and is worth reading. It is a masterpiece of data analysis. The net result of his analysis is the following. Free applications, on average, are used 6.6 times more than paid ones. Since the minimum price is $0.99, and Apple keeps 30%, at minimum a paid application returns $0.70 per user. But since free applications are used at most 80 times per user, based on Pinchmedia’s analytics, then to make advertising worth it, 80 usages would have to generate at least $0.70 per user. $0.70 in 80 sessions is an equivalent of $8.75 CPM, if one ad is shown per session. Yet, Pinchmedia is seeing CPMs more on the order of $0.50 to, at the most, $2.00. Essentially, unless you have a very high usage application, one in which there are lots of advertising opportunities, i.e. users stay in the application a lot and return to the application far in excess of the average 80 usages, or, if the application has a much stronger than 6.6:1 ratio of free:paid, because of the inherent nature of the application, you are unlikely to recoup anything near the revenue of selling the application.

Summarizing, selling directly to the end-user beneficiary is far normally much better for your revenue than advertising, Google’s search business notwithstanding. There are instances in which advertising-supported free services do, indeed, make a lot of sense, and need to be examined on a case-by-case basis. This brings back to mind the late 90s Internet bubble, when revenues did not matter, all that investors and managers looked at were “new economy” metrics, primarily “eyeballs.” Once again, the market has reasserted itself. There are businesses wherein advertising-based works on the Internet, just as it does in the real world. Clearly, however, in most cases, free can be very expensive indeed.