Tuesday, we looked at the IPO prospectus for Extreme Networks. When prospectuses choke the in-box, the trick is to have an organized plan of attack so that you can read them quickly and move on to the next deal without wasting much of your time.
So far, we've perused the underwriters, glanced at pictures to get a quick understanding of the company's product or service and looked at the management, the board and who owns what. With Extreme, there were a few minor issues, but nothing to get us to close the book and move on.
Today, we will do the real digging, as an exercise in reviewing the wealth of information that's in the average prospectus.
The next step is to see if there are any "gotchas," meaning any funny stuff with distributors, or the sales channels, or large customers. The best way to do this is to look at customer concentration. This shows up in two places, in the risk factors, and in the management's discussion of financials.
I find that in fiscal 1998,
accounted for 25% and 21% of Extreme Networks' revenue. But the contribution from the two biggest customers fell in the second half of 1998, with Compaq accounting for 17% of revenue,
only 11%, and 3Com falling off the map after a splurge in the first half. And Compaq was a lot smaller in the second half than the first, on a percentage basis.
It turns out that Compaq was both a customer and a reseller, meaning it uses the switches internally in the Compaq enterprise -- but it also resells the switches, so its percentage of revenue is a bit high for good reason. The analysis going forward should focus on how Compaq is doing, because there is a risk it may fall off the map like 3Com did. But all in all, the customer concentration is tolerable.
Business and Competition
Now that we understand the company's structure and general idea of what business it's in, we can dig deeper and actually read the business section a lot more carefully.
Almost all prospectuses are structured in the same way: industry problem, industry solution, company solution, company strategy, products, customers, company structure (sales, marketing, service, manufacturing, etc.) and, finally, competition.
The problem that Extreme's products address is that corporate Local Area Networks, or LANs, are bogging down. This is happening even though replacing one part of the older technology -- the hubs -- with newer-technology Layer 2 switches has helped. Even so, the demand to move large amounts of data
LANs means that the old technology routers have now become the bottleneck.
The industry solution is to replace the routers with Layer 3 switches, which in effect do the routing in hardware. Extreme offers gigabit-speed Layer 3 switches, in a variety of configurations, for the enterprise and therefore can offer end-to-end LAN switching for its customers.
The bad news is that
-- the gorilla in this field -- sells software-based routers and has introduced its own Layer 3 switches, the Catalyst 6000. The good news for Extreme is that, for now, Cisco probably can't get that aggressive in Layer 3 for fear of eating into its own quite profitable and quite valuable router business. So Extreme can get aggressive and grow.
In the front of the prospectus, there is always a summary of the income statement and balance sheet, but it is usually worthless since it only tells you annualized results. Go find the quarterly results, usually found under that same management discussion mentioned above to find customer concentration. I find it on page 24.
Wow, this thing is growing like a weed. Starting in the third quarter of 1997 and ending in the fourth quarter of 1998, sales go from $600,000 to $5.5 million, $7.3 million, $10.1 million, $12.9 million and $17.9 million. And its gross margins, at just under 50%, are healthy.
It is always dangerous to extrapolate, but the analysts will draw some sort of trendline to forecast 1999 and 2000 revenue and earnings. Remember, there are never projections in a prospectus. It is the only legal document to sell the deal, and no lawyer will ever allow projections -- as it is an invitation to get sued if the company misses the forecasts on the downside or the upside.
Instead, the underwriter's analysts provide forecasts verbally. So the institutions call
Morgan Stanley Dean Witter
and ask for analyst George Kelly's numbers, but I can tell you already they are based on the company's guidance -- with very little massaging. If the company misses George's forecast, it can just blame George. Funny game, but that is how it works.
The risk section of prospectuses has gotten so boilerplate that it's tough to get much out of it. But it is great to know it's there. I usually use the risk section to figure something out that was flagged in reading the rest of the prospectus, some issue with the sales channel, or management deal, etc. I'll read the risk section if I have to, but if I have to, it usually means it is a deal to pass on.
While not critical to read, every venture capitalist loves to read the Certain Transactions section, which should explain how private-round investors got their stock and at what price. You always get annoyed when someone paid 33 cents in May 1996 and you are paying $9 to $11 in April 1999, but that is how the great capitalist system works. Occasionally there is some funny stuff in here -- loans to officers or weird reseller arrangements for stock -- but with Extreme, it all looks clean.
Share Count and Valuation
The last thing to do is check out the valuation being asked by the underwriters and figure out if you can make a return off the deal.
The first step in this process is to figure out how many shares are actually outstanding. In the front of the book, right under the section titled "The Offering," it will tell you how many shares are being offered in the deal (in this case, 5 million), and then how many shares are outstanding after this offering (for Extreme, 45.85 million).
Unfortunately, the shares-outstanding number is always wrong because it does not take into account options and warrants and other stuff, so you have to dig. In fact, I had to go to the "Capitalization" page to find there are another 4 million-plus options with strike prices from $1 to $2.55; therefore, they are in the money and should count as "fully diluted" shares outstanding.
So I have to add 4 million to the 45.8 million, plus I have to add the green shoe, the 15% overallotment the underwriters can sell to more efficiently trade the shares in the first few days. That adds another 750,000, so let's just say there are a little more than 50 million shares outstanding. I then look back on the cover and see the offering range is $9 to $11.
Now underwriters are supposed to underprice the deal by 10% to 15% to get institutional investors to buy them, but this is 1999, and the underpricing can be a lot greater than that! So my guess is that if this is a "hot" deal, and oversubscribed by 20 times -- meaning they have orders for 100 million shares and only 5 million to sell -- Morgan Stanley will raise the range to probably $13 to $15 and it may open a lot higher.
In the interests of full disclosure: My firm,
, has an order in for shares of Extreme, which should get priced Thursday night and begin trading Friday.
Andy Kessler is a partner at Velocity Capital and runs a technology and communications fund out of Palo Alto, Calif. This column is not meant as a solicitation for transactions; it is instead meant to provide insight into the methods of venture capital, technology and investing. At time of publication, Kessler's firm, Velocity Capital, has requested an allocation in the Extreme Networks IPO. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. While he cannot provide investment advice or recommendations, Kessler appreciates your feedback at