John...Here is an interesting contrast to Arrowhead.
Steve
Andy Kessler: New Year's Revelations, Part 1
By Andy Kessler
Special to TheStreet.com
12/22/98 1:51 PM ET
I'm not one to believe in New Year's resolutions. Frankly, I forget them by the time the
Fiesta Bowl is over. But the new year always brings its revelations.
Here is the first half of a list of 10 of them, some more out of the box than others. Each
is intended to help frame the technology investments you have now or will have in 1999.
A capital-spending boom will start in April after Y2K testing concludes in February.
That's right, you read that correctly. The Y2K problem ends in February 1999. Well, it
doesn't end, but most corporate testing of systems for Y2K compliance ends in
February. I have heard this now from several sources. There could still be problems that
need to be solved, but most of the heavy lifting will be completed.
The good news is that new projects and new purchases that had been postponed until
the completion of Y2K testing can now be unleashed. That is why I think you will see a
capital-spending boom beginning in April. Most of the spending will be in Web-enabling
systems and software, but my guess is that it will be bigger than anyone thinks. This
includes servers, routers, switches, caching software, network equipment for load
balancing, enterprise software, self-service software, application server software and
video editing and delivery systems.
The consensus on the Street is that this capital-spending boom won't start until April
2000 or April 2001. Look for it at least 12 months sooner.
A rip-roaring semiconductor cycle will ensue.
Shortages of DRAMs are already starting, and I think we will see this expand to many
other segments, including programmable logic and analog products. Part of this is based
on the coming capital-spending cycle mentioned above, and part based on the huge
inventory reduction that took place from April through October 1998. No one ordered
parts until October, and then only to populate boards or systems that needed to ship.
Order lead times will stretch, prices will go up and semiconductor stocks will double
and triple. It happens every time.
The .com stocks will roll over by March, despite the discovery that shopping is elastic.
It turns out that shopping is elastic, with the elasticity coming not from pricing but from
convenience. As Wal-Mart (WMT:NYSE) and Price Club (COST:Nasdaq)
discovered in the '80s and early '90s, if you make shopping more convenient, people
will buy more.
Now the Web has added the ultimate in convenience (but not yet great prices), and lo
and behold, people buy more, even stuff they don't really need. Admit you bought
something you didn't need on the Web just to see if it really worked. I am still explaining
to my wife the Jimi Hendrix greatest hits album, the Motorola (MOT:NYSE) radios with
five-mile range, the 1,700 Dirty Limericks book and the case of '86 Bordeaux.
However, the .com stocks are gonna roll over. This is based on many factors (and I
won't even count valuation), not the least of which is that first-quarter revenue will be
decline. Retailing is seasonal, and the Web hasn't been, so far. Investors are expecting
growth -- and seasonality will be a bit of a shock.
Add to this three more factors:
The novelty of Web buying having worn off.
An oversupply of stocks from the IPO boomlet I expect between Jan. 4 and March 31,
which will saturate investors' appetite.
As mentioned above, a capital-spending boom returning attention to companies with
growth and earnings.
The Street will realize voice is in.
An increasing number of call centers and help desks will include interactive voice
response and voice recognition. Add to that Web-enabled voice services, like Portico
and Webley. While these are specialized services creating a personal information
manager in the sky, new vertical applications for voice response and voice recognition
will take hold, increasing the total available market for Web services from just those
with a PC to those with phones as well.
Scalability will win out as the ultimate differentiator.
The PC went from zero to 100 million units sold over the last 20 years despite being an
incomplete product that requires a huge infrastructure -- including both a value-added
reseller channel and tons of tech support inside corporations -- to implement real
functionality and perhaps even productivity.
In other words, it really wasn't scalable, but because of its continual compatibility (even
back to the first PC), a support infrastucture scaled around it. Well, those days are
over. From now on, new applications and new services must be rolled out without
support. Amazon (AMZN:Nasdaq) is a good first example. No real handholding was
required to get customers to execute transactions. Instead, a scalable infrastructure of
servers and routers and caches and load balances that is still being augmented helped
Amazon go from zero to 5 million customers in record time. Look for things that scale in
'99, not things that are backward-compatible to the first PC.
Look for more revelations tomorrow.
Andy Kessler is a partner at Velocity Capital and runs a technology and
communications fund out of Palo Alto, Calif. At the time of publication, the fund owned
General Magic, which runs Portico, although positions can change at any time. 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. Under no
circumstances does the information in this column represent a recommendation to buy or
sell stocks. Kessler appreciates your feedback. |
