On March 13th, Bill Simpson wrote an analysis of Nextest (NEXT-OLD), "a small but promising little semiconductor testing outfit." Nextest went public on March 22, generating gross proceeds of $80.9 million. The text of Simpson's writeup follows:
From the S-1:
We design, develop, manufacture, sell and service low-cost, high throughput automated test equipment, or ATE, systems for the semiconductor industry. We address rapidly growing, high volume segments within the semiconductor industry such as the flash memory and flash-based system-on-chip markets.
The big difference between recent IPO Eagle Test Systems (EGLT-OLD) and NEXT is that EGLT does not provide test equipment for consumer flash memory semiconductors, while it is the focus for NEXT. EGLT's focus is mixed signal semiconductor testing; NEXT is digital, specifically digital flash memory.
Flash memory has taken hold the past couple of years as the key internal component in many consumer digital products, the reason being flash memory maintains its data without any external power source and can be easily erased and reprogrammed. Flash is ideal for consumer digital devices and most estimates continue to call for strong growth in consumer related flash memory the next few years. Pricing tends to be very competitive.
These flash memory chips/components need to be tested whether they be in the design, fabrication and/or manufacturing stage and that is where NEXT comes in. NEXT believes their Magnum systems offer the high throughput low cost scalable solution that gives them an advantage in the flash memory testing niche. Highlights of Magnum include the ability to run independent multi-speed parallel tests on flash memory semiconductors and components. Additionally NEXT claims Magnum offers higher speed testing and lower cost per pin. The 5,120 pin configuration of the Magnum can test 640 NAND flash devices in parallel. The 5,120 pin configuration can also simultaneously test 160 NOR flash devices. NEXT claims they have the equipment that offers the lowest cost flash memory tests.
Along with flash memory semiconductor testing, NEXT products also test flash memory components such as integrated circuits, or ICs, microcontrollers, smart cards, and field programmable logic devices, or FPGAs. End products for NEXT semiconductor customers include portable music players, cellular phones, digital cameras, computer notebooks and USB flash drives.
Of note, NEXT test products are designed to test both the stagnant NOR flash memory and the much faster growing NAND flash memory. NEXT believes their recent strong quarterly growth rates are directly attributable to testing NAND flash memory.
In the past 18 months 75% of revenue has been derived from Asia, 25% from the US. NEXT is still relatively small having shipped over 1,500 systems to 60 semiconductor companies since inception. Much like EGLT, the bulk of NEXT's revenue has come from a few customers. Amkor Technology, Atmel Corp., SanDisk, Hynix and Samsung have accounted for the bulk of NEXT revenue the past 3 years.
$4+ per share in cash post-offering, no debt. 2 3/4 X's book value at mid-range pricing. Revenue troughed in 2002 and 2003. Since than NEXT has ridden the wave of flash memory boom avoiding the steep revenue drop-offs of non-flash semi test equipment makers such as recent IPO EGLT. In FY 2004 revenue tripled, while FY 2005(ending 6/30/05) saw revenue climb 10% to $48 million. Fueled by the 2 strongest quarters in NEXT history, revenues for FY '06 are off to a good start. Thus far through first 1/2 of FY '06, revenue is on pace for a 40%+ increase to $70 million. Seasonally the first half of each fiscal year is stronger for NEXT, the $70 million estimate actually forecasts lower revenues the 2nd half of FY '06 for NEXT. Gross margins quarterly have remained in the 45%-50% range. Operating expenses really jumped up in FY '05 as NEXT rolled out their new Magnum product. This led to roughly 25 cents per share loss, coming off a small net gain in FY '04. For FY '06 NEXT has managed to keep SGA in line with FY '05 while growing revenue rapidly as note above. This has led to operating margins through the first 6 months of FY '06 of 13%. I imagine that will contract a little going forward as seasonality kicks in, but I do think NEXT can book net margins for FY '06 of 8%. That would lead to net earnings for FY '06(ending 6/30/06) of 30-35 cents for NEXT. At mid-range pricing of $15, NEXT would be trading at 45 X 2006 earnings. This is roughly the current fiscal year valuation at which EGLT priced their IPO.
The big risk here is the cyclical nature of the semi/semi component test business. While NEXT was able to weather the most recent testing niche downturn in 2004, revenues did take a nosedive in the early 2000's. I don't see this as a near term risk for NEXT as they are focused on the sweetest spot in the semi test niche, flash memory. Robust backlog backs this assessment.
However the downturns in this sector are severe and often come with little warning. Much like most of the small technology companies coming public this decade, NEXT will not be debuting cheap. A strong initial valuation definitely leaves NEXT share price very vulnerable to a change in cycle. This is something that needs to be watched closely, because when these cycles turn they turn hard and ugly. Sandisk (NASDAQ:SNDK) has been a consistently large customer for NEXT test equipment since FY 2003. A good tell going forward for NEXT may be be SNDK's outlook on the strength of their flash end-market. As long as SNDK continues to see strong demand, NEXT should be all right.
I like this NEXT IPO much more than EGLT. It is not coming cheap by any means, but NEXT equipment is focused on testing the fastest growing semiconductor sector of the past few years, consumer flash memory, one too which is forecast to continue strong growth the next 3-4 years. For this reason I like this one in range and even $1-2 above. This is a deal that should work much better than similar EGLT. Yes there are risks here, the topmost being one of initial valuation. However I think demand will exist for this one, I'd take allocations and will be looking to enter on a reasonable aftermarket print.