司亞樂 (SWIR) 2005 Q2 法說會逐字稿

Thank you for holding for the Sierra Wireless second quarter 2005 results conference call. I'd like to introduce your speaker, David Sutcliffe. Please go ahead.

Thank you, operator. I'm David Sutcliffe, the President and CEO of Sierra Wireless, and with me here for the conference are Dave McLennan, the CFO, and Jason Cohenour, the COO.

I'm going to turn it over to Dave in a moment to do the forward-looking statements disclaimer and cover Q2 financial performance. Jason will then address business highlights and product update for the quarter. We'll go to financial guidance and then summarize and open the call for your questions. And with that, I'll turn it over to Dave McLennan.

Thanks, David. Our disclaimer is as follows. Forward-looking statements are not guarantees but are only predictions that relate to future events or our future performance or state other forward-looking information and are subject to substantial risks and uncertainties that could cause our actual results, performance or achievements to differ materially from those expressed, anticipated or implied by forward-looking statements.

These forward-looking statements relate to, among other things, our revenue, earnings and other financial guidance for the third quarter of fiscal 2005, plans, objectives and timing for the introduction or enhancement of our services and products, statements concerning strategies, development, statements about future market conditions, supply conditions, channel and end-customer demand conditions, projected or future revenues, gross margins, operating expenses, profits and other statements of expectations, intentions, objectives and plans that are not statements of historical facts. Forward-looking statements reflect our current expectations.

The risks and uncertainties that may affect our actual results, performance or achievements are many and include, among others, our ability to develop, manufacture, supply and market new products that we do not produce today and that may not gain commercial acceptance, our reliance on the deployment of next generation networks by major wireless operators and increased competition. These risk factors and others are discussed in our regulatory filings with the securities commissions in the U.S. and Canada. These factors should be reviewed carefully and you should not place undue reliance on any forward-looking statements.

On to our results. Our results are reported in U.S. dollars and are in accordance with U.S. GAAP.

For the second quarter of 2005, our revenue was $21.9 million. Gross margin was negative $5.9 million. Operating expenses were $21.2 million and our net loss was $26.7 million or a loss per share of $1.05.

As reported on June 7th, 2005, we are exiting the Voq Professional Phone initiative. We are incurring restructuring costs and other expenses and we have shifted some of the Voq resources to our core PC card and embedded modules business. This has resulted in substantial restructuring costs being incurred in the quarter of approximately $13.5 million related to Voq.

In addition, we have implemented some non-Voq-related reductions in our operating expenses and assets. Total restructuring and other charges and legal provisions during the quarter amounted to $19.2 million and include inventory write-downs, severance costs, impairment of fixed, intangible and deferred tax assets, provisions for facilities restructuring commitments, other costs related to the restructuring and future legal costs associated with litigation matters.

Excluding restructuring and other charges and the provision for future legal costs of $19.2 million in the second quarter, our revenue was $21.9 million. Gross margin was $6.9 million or 31.2% of revenue. Operating expenses were $15.3 million and our net loss was $7.5 million or a loss of $0.30 per share.

A summary of the $19.2 million of charges taken and the impact on our financial results is as follows. $12.8 million is included in cost of goods sold and include Voq components and finished goods inventory of approximately $8.6 million, contractual royalty commitments of $2.7 million, fixed asset impairment and commitments of $1.4 million and $100,000 of reduction-- work force reduction costs were also included in cost of goods sold.

$4.9 million of restructuring and other charges in OpEx include fixed and intangible asset impairments of $2.6 million, work force reduction charges of $1.5 million related to severance costs associated with 32 employees terminated during the second quarter of 2005.

In total, we are reducing our work force from 321 employees to 270 employees. Of the 51 employees terminated, 32 left during the second quarter of 2005 and 19 are on working notice that ends during the third quarter of 2005.

Related to the reduction in our work force, we increased our facilities provision by $700,000 and incurred other charges of $200,000.

$500,000 was charged to tax expense relating to an increase in our deferred tax asset valuation allowance and we also recorded a provision of $1 million for future legal costs associated with litigation matters, which is included in G&A in administration expense.

So that gives you a breakdown of the $19.2 million of charges in the quarter.

We expect that the costs of the restructuring will total approximately $5.9 million, of which $5.2 million is expected to be disbursed in the third quarter of 2005.

We expect these actions will lower our operating expense run rate relative to our second quarter by approximately $2 million per quarter.

Results for the second quarter of 2005 relative to company guidance, which we provided on April 21st, excluding the impact of restructuring and other charges and legal provisions of $19.2 million, were as follows. Revenue of $21.9 million was better than our guidance range of approximately $20 to $21 million. Gross margin percentage was 31.2%, lower than our guidance of 34%. Operating expenses were $15.3 million, better than our guidance range of $16 to $16.2 million. Our net loss of $7.5 million or a loss per share of $0.30 was better than our guidance of a net loss of approximately $8.5 million or a loss per share of $0.33. Our cash flow from operations was negative $12 million, consistent with our guidance of negative cash flow.

Looking at Q2 2005 results, excluding the impact of restructuring and other charges and legal provisions of $19.2 million compared sequentially to the first quarter of 2005, revenue increased to $21.9 million from $20.2 million, an increase of 9%. Gross margin percentage decreased to 31.2% from 35.3%. Operating expenses remained virtually unchanged at $15.3 million compared to $15.2 million in Q1. Similarly, our net loss of $7.5 million or $0.30 per share in Q2 was virtually the same as the $7.6 million net loss in Q1.

Again, sequentially, revenue increased primarily due to the increased sales of EVDO and edge PC cards across a well-diversified group of customers and geographical regions, offset by decreases in sales of our legacy 2.5G products. Gross margin percentage declined, reflecting the effect of lower selling prices, partially offset by favorable product mix. Included in Q2 gross margin were non-recurring supply chain costs, which negatively impacted Q2 margin by 2%. Operating expenses remained flat, reflecting the continued focus on cost management.

Our Q2 balance sheet compared to Q1 balance sheet shows that our cash and short-term investments decreased by $15 million during the quarter to $105.7 million, reflecting the impact of negative cash flow from operations, as well as capital expenditures of approximately $2.6 million. Trade accounts receivable DSO was 51 days, favorably below our target of 60 days. Inventory, excluding Voq-related amounts, declined favorably during the quarter by approximately $2 million.

On a year-over-year basis, our Q2 '05, excluding the impact of restructuring, other charges and legal provisions of $19.2 million compared to Q2 '04, revenue decreased by $57.5 million to $21.9 million in 2005 from $51.6 million for the same period in 2004. Gross margin percentage decreased to 31.2% compared to 40.5%. Operating expenses were $15.3 million in the second quarter of '05 compared to $13.5 million for the same period in 2004. And net loss for the second quarter of '05 was $7.5 million or a loss per share of $0.30 compared to net earnings of $6 million or diluted earnings per share of $0.23 in Q2 '04.

Again, on a year-over-year basis, PC card revenue was lower, primarily due to the impact of increased competition in the CDMA EVDO market and lower sales of legacy 2.5G products. Embedded module revenue was lower due to the completion of CDMA embedded module shipments at the end of '04 to Palm One for the Trio 600.

Gross margin percentage declined, reflecting the effect of lower selling prices, partially offset by favorably product mix. Included in the Q2 '05 gross margin were non-recurring supply chain costs, which negatively impacted margin by 2% in the quarter. Operating expenses increased primarily as a result of new product development initiatives.

I'd now like to pass it off to Jason Cohenour for some business and product comments.

Thanks, Dave. I'll begin with segmented financial performance. First, revenues by product line. Revenue from PC cards represented 75% of our revenue in Q2 compared to 65% in Q1. OEM represented 7% of revenue in Q2 versus 11% in Q1. MP represented 15% of revenue in Q2 versus 21% in Q1 and we had revenue from other products of 3%.

In summary, PC card sales were up considerably over Q1. Embedded module sales were down slightly as we drive to new product introductions during this half and revenue from MPs was strong for the second quarter in a row.

Revenues by distribution channel -- carriers represented 42%, resellers 49%, OEMs 8% and direct and other 1%.

Moving to revenues by technology, revenue from GSM products, including edge, accounted for 44% in Q2 versus 31% in Q1. Revenue from CDMA products represented 53% versus 66% in Q1. Revenue from other products represented 3% during Q2.

During Q2 we saw significant growth in our GSM business, driven by increased sales of our AirCard 775 edge PC card to carriers and channel partners around the world. Revenue from new products -- that is, products introduced in the past 24 months -- represented 79% of our revenue in Q2.

Moving to revenues by geography, the Americas represented 58% of our revenue in Q2 versus 64% in Q1. Europe represented 14% of our revenue in Q2 versus 12% in Q1. Asia-Pacific represented 28% of our revenue in Q2 versus 24% in Q1. Revenue in the Americas was flat quarter-over-quarter and our growth in the quarter was driven by increased sales in Europe and Asia-Pacific, which together accounted for 42% of our Q2 revenue.

We expect product channel technology and geography segment percentages to fluctuate quarter-to-quarter based on mix and new product introductions.

Moving to business updates by product category and starting with Voq, in June we announced our decision to exit our Voq initiative and shift some of the Voq resources to our core PC card and embedded modules business. Following our announcement, we engaged an investment banker to assist us in selling either the entire Voq program or some of its components.

We've conducted a number of meetings with companies expressing an interest in Voq and continue to have dialogue with a few interested parties. At this time, however, our assessment is that we will not have a material monetary recovery from the Voq program or its assets.

Including in the restructuring, other charges and provision for legal costs of $19.2 million incurred in the second quarter are inventory and fixed asset write-downs, provisions for contractual obligations and employee severance costs associated with Voq totaling $13.5 million. Of the 55 people we had working on the Voq program, 24 were transferred to our PC card and embedded modules business and 31 were either terminated during the quarter or on short-term working notice.

Moving to an update on PC cards and starting with activity in currently shipping products, in EVDO the EVDO PC card market continues to experience increasing competition, particularly in North America. Notwithstanding the increase in competitive activity, we continue to make solid progress with our EVDO PC card business. During Q2 we completed network certification of the AirCard 580 Telus Mobility in Canada and commenced commercial shipments. Our EVDO PC card products are now certified on 9 networks around the world.

We also continued to ramp up our AirCard 580 shipments to Sprint and are encouraged with the level of sales activity and early reported sell-through. We believe that our first-to-launch position at Sprint is helping us capture strong channel share.

AirCard 580 sales to Telstra in Australia and Telecom New Zealand also continued to be strong during the quarter.

New order activity for our EVDO PC cards was solid during the quarter and we carried a significant backlog of orders into Q3 from several customers, including Sprint and Verizon.

Moving to edge, our AirCard 775 sales to Cingular continued to be solid in Q2. Sales of the AirCard 775 were particularly strong in Asia, where we launched the product with PLDT in the Philippines and Guandong Mobile in China.

In Europe we saw-- we also saw solid progress with the AirCard 775. As you may recall, we announced a planned launch of the AirCard 775 with Bouygues Telecom in France back in February. During Q2 we completed network certification with Bouygues and commenced commercial shipments.

The AirCard 775 also launched by-- was also launched by Telford in the Netherlands and in the Middle East with Brightpoint.

Overall, we obtained network certification for the AirCard 775 from 9 additional carriers in Q2. The AirCard 775 is now certified on 19 networks around the world.

Sales of our legacy 2.5G PC cards also continued to produce revenue, although we expect revenue from these products to decline over time.

PC card inventory reported to us by our channels continued to decline during the quarter.

Overall PC card bookings were up considerably in the quarter, including orders for current products and new products that we expect to introduce by the end of the year. We're going into Q3 with a stronger backlog-- order backlog than in the prior 2 quarters.

Moving to new development initiatives in PC cards, development of our new HSDPA PC card products continues to be on track. We were the first company to provide a live demonstration of an HSDPA PC card earlier in the year at 3GSM. Since then, we've continued to work closely with infrastructure providers on IOT testing, technical and market trials and co-marketing activities. IOT is now underway with Ericsson, Huawei, Lucent, Motorola, Nokia, Nortel and Siemens.

Sample units of our HSDPA PC cards have been supplied to a number of tier-1 carriers around the world. Some of these tier-1 carriers have selected us to supply them with HSDPA PC cards and we have received initial purchase orders.

We expect to commence commercial volume shipments of our HSDPA PC cards in Q4 of this year. Our goal is to launch our HSDPA cards with leading European, Asian and North American carriers starting in Q4 and throughout 2006.

In CDMA, our development of the CDMA EVDO Release A PC card is well underway. We expect this-- to introduce this PC card in the second half of 2006, concurrent with the commercial deployment by carriers of Release A.

Moving to embedded modules, we commenced shipment of our new EM5625 EVDO embedded module to customers outside of North America during Q2. The product is currently in certification with major North American carriers and we expect to begin shipments in North America during Q3. We believe the EM5625 will be the first EVDO module available in North America.

During the quarter we received new design wins for the EM5625 from MobileAria, a division of Delphi, and AirLink. These design wins are in addition to wins we announced earlier this year from Panasonic, Itronics, AtRoad and Falcom.

We also continued to invest significantly in mini PCI Express modules for laptops and are making strong business development progress in this area. Earlier in the year we reported design wins with 2 major laptop OEMs for mini PCI Express modules. Both of these wins were for EVDO modules. During Q2 we earned another design win with one of these laptop OEM customers for an HSDPA module. In addition, an OEM customer has expanded the number of laptop platforms that will offer embedded wide area wireless capability using our modules.

We are currently providing sample mini PCI Express product to our OEM customers for integration and carrier certification. We expect to begin shipping EVDO mini PCI Express modules by the end of the year, followed by HSDPA mini PCI Express modules in early 2006.

We have long believed that major laptop OEMs would eventually embed wide area wireless into their platforms. This now appears to be well underway. We view this trend as a key strategic opportunity for the company and believe that we are well positioned to lead in this space.

As you can see from my comments, we have some compelling new product initiatives happening in the fourth quarter in both PC-- in both the PC card space with the launch of our HSDPA PC card and the laptop OEM space with the launch of EVDO mini PCI Express modules. We expect these new products will make a moderate revenue contribution to our 4th quarter results, which reflects the typical timing risk related to the completion, certification and launch of new products, launch timing that often results in less than a full quarter's worth of revenue contribution and normal new product production ramp constraints.

I'll now pass it back to Dave for Q3 guidance.

Thanks, Jason. We are providing guidance for the third quarter ending September 30th, 2005, which reflects our current business indicators and expectations. Inherent in this guidance are risk factors that are described in detail in our regulatory filings. Our actual results could differ materially from these-- those presented below. All figures are estimates based on management's current beliefs and assumptions and are subject to change.

With the decision to exit Voq and the associated restructuring, we have a very focused 3G product development plan for key initiatives in the PC card and embedded modules for the laptop market. We expect the first products associated with these initiatives to begin shipping by year end.

We expect Q3 to show sequentially improvement over Q1 and Q2. Accordingly, we are providing financial guidance for Q3 2005 as follows -- revenue of $24 million, gross margin of 31%, operating expenses in the range of $13.7 to $13.9 million, a net loss of minus $5.7 million or minus $0.22 per share and cash flow from operations, negative.

With that, I'd like to pass it over to David to sum up.

All right. Thank you. Just by way of summary for Q2 first, I think the second quarter was a strategic turning point for our business. We made the difficult decision to exit the Voq initiative, we've reduced our cost structure and we've refocused the company on our core PC card and embedded modules businesses.

Revenues for the 2nd quarter were up 9% sequentially to $21.9 million and I'm particularly pleased with securing our first HSDPA design win with a major laptop OEM and with receiving initial purchase orders from carriers for our new HSDPA PC card.

Looking at Q3 and the future, our focus for the remainder of this year is to continue tight OpEx management while developing and launching new products. This focus includes a series of new products across new air links that are expected to launch, commencing in Q4 and throughout 2006, including UMTS and HSDPA PC cards, EVDO Release A PC cards and both EVDO and HSDPA embedded modules for laptops. We expect these initiatives to provide the foundation for continued sequential growth and for our planned return to profitability.

And with that, I'll turn it over to our callers. Operator, we're ready to take questions.

[OPERATOR INSTRUCTIONS] Andrew Lee, TD Newcrest.

The restructuring benefit of $2 million per quarter, will that be fully reflected in your Q3 guidance already?

There will be some minor charges associated-- or minor expenses, I should say, associated with some of the people that are on working notice, Andrew, so most of that $2 million will be in the Q3, but a small amount of approximately $400,000 will be-- will still be left over from the Q2 restructuring.

Sorry, I don't think I was clear. I'm referring more to the operating expense reduction, the benefit of the restructuring. Is that what you were addressing there?

Yes, I was.

OK. So in Q3, then, you're talking about $1.6 million will be reduced because of this. Is that how I think of it?

Correct.

OK. And the-- the guidance-- or for the next several-- when do you guys forecast break-even, if you're willing to put that out there? And what is the gross margin you need to reach break-even under your current operating model?

Andrew, we're not forecasting a break-even point yet. As you know, our practice is to give financial guidance 1 quarter ahead and we-- you've seen the Q3 guidance. It doesn't include break-even. To the second half of your question, at what gross margin level do we reach break-even, that's-- break-even's going to be both a function of top-line growth and gross margin improvement, while we maintain the OpEx controls that-- and reduced operating expenses that Dave's just described.

All right. So, Dave, deep in the gross margins there's a comment about 2-- sorry, 2 percentage points being impacted by one-time items and yet your guidance for the next quarter is still relatively weaker than what you've looked for in the past. What's happening with that?

In the next quarter we're anticipating product mix to move the gross margin down a couple points.

OK. And in the past you've talked about driving towards 40%. Do we assume, given the pricing pressure and the competition today that that is something that you're going to have revisit your long-term model?

Well, we're still committed to our long-term model and we have to improve gross margins over the 31% or 33% levels in order to get to the kind of performance that we want to get.

OK. Last question, then, on the-- on the embedded module market, the margins you were getting for the Palm business and the other handset business, do we think of that business as being as price competitive or even more so as you move towards tier-1 OEMs?

I think, Andrew, as we've stated before, our OEM business tends to be very volume sensitive, regardless of who the customer is. So if there's a high-volume customer, like a Palm, or, potentially, like large laptop OEMs, we would put those in the same category in terms of price sensitivity and resulting gross margin.

Can you give us a band of margins? Are we talking low 20s here?

Well, I think the-- what we've provided in the past in terms of gross margin for our overall embedded modules business is somewhere between 20% and 30% -- again, highly volume sensitive. So you would expect higher volume customers to be at the lower end of that range.

Paul Coster, JP Morgan.

You talked about the gross margin decline and attributed most of it to a change in the product mix. Is that correct? ASPs are down. To what extent is it a function of just simply price-based competition?

Well, we have a few things contributing to gross margin being lower in the 2nd quarter. That includes competitive pressures on selling prices. It also includes a reduced volume efficiency in our manufacturing operations and it included favorable product mix, skewed towards PC cards. So those three factors were the key contributors, two negative and one positive, to the particular gross margin mix we reported for Q2.

OK. The-- the percentage of revenues that came from the card business grew quite substantially. And the card business, if I've got this correct, grew sequentially by about 25%. Is that correct?

That's correct.

Yes, that's right. We were up 26% in dollar terms.

Right and that's not a function of pricing, that's a function of volume.

In large part, that's a function of volume and if you assume some ASP erosion, our unit growth was probably higher than 26%.

And to what extent were you fighting a foreign exchange head wind?

Foreign exchange wasn't helpful to us in Q2. We took some negative foreign exchange effect there, but that's a fluctuation we have to deal with one way or another in any given quarter and it wasn't hugely material.

OK. Last question from me. The OEM design wins sound encouraging, but do you have any kind of binding commitments from the end customer and/or do you have any sort of evidence of what percentage of-- percentage of design wins translates into subsequent revenues? Thank you.

Well, perhaps starting with the last part of your question. In our experience, the majority, a very strong majority, of design wins result in an eventual product shipment and revenues and we haven't had a lot of experience with announcing design wins and then not having the resulting product go to market.

We are not providing any new-- any contract details at this point in time for any of our design wins, however, they are sufficiently strong for us in the opportunity that we assess that we're committing significant resources and, in fact, moved resources out of the Voq initiative and into the embedded modules arena to strengthen our development teams there.

Jeff Kvaal, Lehman Brothers.

My first question comes on-- regarding Verizon. It seems as though there are now 4 vendors at Verizon and I'm-- it seems to be a pretty-- a higher number than we've seen at other carriers. Should we expect that to be the name of the game at other carriers, as well, or is Verizon a special case?

Well, Jeff, it's very hard to say what other carriers will do relative to Verizon. In our experience, carriers like to have as wide a range of devices as they can for their customers to choose from, and, on the other hand, they like to minimize the number of devices they carry because each new device they introduce adds significant internal training and call center support costs. So as we've seen in the past, each carrier looks at their own way and how other carriers will come out is very hard to predict based on one carrier's actions.

OK. Switching gears for a second, on the OEM business, the gross margin may be lower, but I would imagine that their-- the operating margin structure may not parallel the gross margin structure. Should we think-- how should we think about the operating margins in the OEM business, either historically or in the future?

Well, it's easiest for me to comment on a historical basis. And I will say that it's our view that the contribution margin for our OEM or embedded modules business is similar to the contribution margin for our PC card business, which has a notionally higher gross margin. So-- And the reason for that is simply that when we sell PC cards we have a higher sales and marketing costs and higher support costs than when we sell higher volume embedded module business to an OEM who takes on most of the sales and marketing effort, the brand spend and the post-sale customer service.

So-- so we-- while we-- while we report gross margins and we focus on them and we believe you focus on them, the contribution margin would be another key measure and in our historical assessment contribution margins on embedded modules have been just as attractive as those on PC cards.

John Bright, Avondale Partners.

Gentlemen, what would-- would you give us an update on the CEO search, maybe a timeframe?

Yes, I'm happy to do that. It's David speaking and just to recap for other callers, at our annual general meeting last April I announced my plans to retire as CEO around the end of the year and to continue with the company as an outside director. The company engaged at that time Spencer Stuart and formed a search committee on the board.

That process is well underway and we continue to expect it to be completed and my successor to be named around the end of this year. We are looking at both internal and external candidates in the course of the search and there's good interest.

And finally, since that announcement and until whatever time my replacement is named, I remain fully engaged as the President and CEO, active in the business and I plan to do so.

OK. A followup question. David, I think you mentioned that the focus, certainly, on this year's OpEx is a management focus. Does that mean that there are potential additional cuts possible beyond what you've already described?

Our current plan is to make the restructuring and other cost adjustments that we signalled on June the 7th with the decision to exit Voq and are providing the details on today and I expect from here on in we have the development and sales and marketing programs staffed the way we want to have them staffed. We're focused very hard on bringing those new products through development and into market launch and I would expect our resource level to remain stable for the balance of the year.

OK. On a top-line question, what-- would you name the-- as you see it, the top-- maybe the top 2 or 3 attractive tier-1 carrier opportunities that have not yet been decided that make take place, you would think, within the next 6 months to a year?

Yes, I think I don't fully understand your question. I don't believe you're asking me to name individual carriers, so could you clarify your question?

Is it-- Yes, I was actually thinking about naming individual carriers that you might think might put out-- think that might be the best opportunity that you're going to be going after that you haven't already made inroads in at this point?

All right. OK, well, I'm going to stop short of actually naming prospective carriers, but I will say that the company already does business with many of the largest wireless service providers, not only in North America, but also in Europe and in Asia and our biggest opportunities over the next 12 months or so with new carriers are likely to come as a result of the introduction of new products, specifically HSDPA PC cards, which will open the door for us to go into UMTS and HSDPA carriers who-- some of whom we don't currently serve. Others we currently serve already, either with GPRS or with edge products.

So you're primarily talking about European carriers at this juncture?

You'd certainly think that at first blush, but it's not just European. We're very focused on getting products out that will address the needs of both European and North American HSDPA network operators and that's an integral part of our go-to-market plan for that new product.

Sure. Sure. Last question, then, on a-- with your partial departure, the exit of Voq, thoughts on a diversification strategy?

Well, just having come off a quarter in which we got rid of a diversification strategy and refocused on our core business, I'll have to duck that question for at least one more quarter.

OK. Any thoughts on your direct competitor Novatel's Ovation strategy?

Well, we're familiar with the strategy from-- from an observer's perspective. It's not one that we're pursuing directly. I will say that it's a broadband wireless access strategy and one which we do address indirectly through OEM partners. That is, we sell embedded modules to OEMs who are focusing in that market, one of whom we made an announcement with earlier this year, Falcom Technologies.

So I don't think you should be anticipating that we're going to enter that space directly, but you should understand we're already playing in it indirectly through our OEM customers. It's a very-- we see it, as far as direct participation, as a very crowded space. Some of our market research indicates that there's somewhere between 50 and 100 players who are either already shipping product into that space or who have announced plans to do so. So it's not an area that we want to go into directly ourselves.

Amit Kapur, Piper Jaffray.

I was wondering if, just getting quickly back to the pricing environment for PC cards, can you indicate whether the degree of pricing pressure is different for your edge products versus the CDMA/EVDO stuff?

Amit, this is Jason. It's hard to say if pricing pressure is any more intense on either edge or EVDO. Typically what we see is after we launch a product and get it in the market and the number of competitors increases over the life of that product, you get pricing pressure. And I think that's-- that's an environment that stands for both our EVDO products and our edge products. And as we've explained in the past, as we do new product introductions that tends to alleviate, for some period of time, ASP pressure, particularly if you're early to market with those products.

OK, great. And then just quickly on your Q3 guidance, are there any particular product lines you see driving more of that growth than others or is it pretty much sequential growth across the board?

Yes, we think we expect to have another good performance out of PC cards and we would expect embedded modules or OEM mix to increase during the quarter.

Great. And probably the final question for me, just in terms of, how do you see the competitive environment shaping up for HSDPA? Cingular mentioned this morning they're trialing 3 PC cards. What do you view as your competitive advantages once the networks are up and running?

Well, we expect to be early to market. We expect to have a wide range of interoperability and carrier certifications and we expect to be differentiated with products targeted at the North American and European markets. We expect a fast roll, a technology evolution from the 1.8 megabits speed that these networks will be introduced at through 3.6 and 7.2 as HSDPA is deployed and as performance enhancements are rolled out on infrastructure.

And so, in short, we expect to be early to market and a first mover on technology and other product differentiation points.

Gus Papageorgiou, Scotia Capital.

Dave McLennan, sorry, you said that the-- some supply chain issues gave you a one-time gross margin hit in the quarter of 2%. Could you just clarify exactly what that was?

It was a one-time specific thing in the quarter, Gus, but it's not something that's outside of the normal bounds of the manufacturing environment. So a bit of turbulence in the quarter there on the supply chain.

I'll add to that that it was a function of our recent reduction in overall unit volume that we had to renegotiate certain supply chain commitments and there's a-- there's some cost associated with doing that.

OK. Just a question on the model side for PCs. There obviously seems to be some activity there. What are the PC OEMs waiting for? Are they waiting for a broad HSDPA video coverage to launch these products? Are they waiting for the modules? What's the-- what's the big gating factor there?

Well, the PC OEMs have a new product cycle just as we do and they've-- after what we would view as years of waiting have decided that the market conditions are right or about to be right for them to introduce product models that have embedded wide are functionality and so now we're dealing with their cycle of going through their product cycle and incorporating these new products into their planned products.

So once one of these companies makes a decision to go ahead, there's a significant timeline to get a product to market. They have to plan the-- they have to select vendors, do engineering to integrate the device, deal with FCC or equivalent approvals, go through interoperability and carrier testing and bring their new product to market with our or a competitors embedded module inside of it. And that's the case for both EVDO and HSDPA embedded modules going into laptops.

Specifically, in the case of HSDPA, they also have to time the introduction of their models that include HSDPA capability for when HSDPA service will be commercially available. Hopefully that gives you a flavor of the process that they and we are going through together.

OK. I guess just a final question here is, on that trend, from a carrier-- from a PC OEM perspective, I mean, what you're doing here is you're adding a module which is just making your PC more expensive and since you don't have the recurring revenue stream that the carrier has to kind of subsidize that module, why wouldn't they just leave it to the carrier to sell them an AirCard? I don't understand the business case for the PC OEM.

Well, there's economics for us. There's economics for the service provider and there's economics for the OEM and the service provider offers to its existing channels significant channel incentives or economics for signing up new subscribers. So that's as much as I could make an observation from, because I'm neither the service provider, nor the OEM, but clearly there are-- there's an existing business model where carriers provide channel incentives to get new subscribers, particularly high value-added sticky subscribers like data customers.

Pat Svallo (ph), Merrill Lynch.

Just a real clarification first. So am I correct in saying the total cost from the restructuring is going to be $2 million for a quarter and that will essentially be complete by the end of Q4?

Let me just paraphrase that back. The total operating savings per quarter will be $2 million. We're getting $1.6 of that right away in Q3 and after Q3 we expect to be at the new run rate.

And that would be, at least for now, your trough operating expense level?

That's correct. And then you should expect operating expense in any given quarter will fluctuate based on new product introduction cycles and other things like that.

Right. OK, got it. Next question, I just wanted to focus a little bit more on HSDPA. Now you talked about sort of commercial volumes in Q4, but given sort of, I guess, the limited coverage, at least from a network perspective, on HSDPA, would those commercial volumes be at-- at a substantially reduced rate versus or compared to your more typical products? Is that sort of a fair statement?

Well, I think there's really two parts to the answer there. One is yes and one is no. Yes, HSDPA sales in the first quarter that we start shipping the product are likely to be at a lower level than a product that's already shipping and has a run rate.

Three reasons for that. You have timing risk on exactly when you introduce the product. The second thing is you're introducing it during a quarter so you don't get a full quarter of run rate and the third factor is that you have to ramp that product up. Production doesn't go from 0 to 60 miles an hour on the first day of production. You have to go through a ramp process.

So for those reasons it's prudent to have moderate expectations around revenue contribution in the first quarter a new product is introduced.

Having said that, the other thing that I think is very important to call to your attention is the product we're introducing in HSDPA, the PC card product, is backwards compatible to UMTS, to edge and to GPRS and it's not only possible for UMTS and edge and GPRS service providers to offer that product before they have HSDPA deployed or before it's widely deployed, but not only that, from many points of view it's desirable for them to do that, because it allows them to future-proof their end customers from having to upgrade hardware at expense to both the end customer and the operator.

And so it should be attractive to UMTS service providers, in particular, to start offering HSDPA products as soon as they're available. And we saw that experience on EVDO with at least 2 of our significant network operator partners, Sprint and Telus, both of whom introduced EVDO PC cards in advance of broad EVDO footprints

OK. And, David, as a followup to that, given that potentially the HSDPA card volumes will be a little bit more moderate in that it is backwards compatible, what is the risk, then, to maybe seeing a bumpy quarter or two from any cannibalization from carriers opting to wait for the full HSDPA card instead of ordering just the regular cards without the HSDPA.

Right. Well that's going to be a very interesting question for the companies that are shipping UMTS PC cards and don't have HSDPA PC cards ready when we start shipping HSDPA. It could well create some significant -- well, I guess I can't call it cannibalization, but competitive gains.

And with respect to our edge product line, we could see some cannibalization from edge to HSDPA. And I think that'll be moderated by the fact that many of our edge network operators -- and Jason gave an update on the number of new certifications on edge and the total certifications around the world on edge -- many of our edge operators are some distance away from deploying HSDPA and are pretty committed on edge right now.

So we may see some cannibalization there. I think the biggest effect's going to be on UMTS PC card sales for network operators who have UMTS and are deploying HSDPA.

And just one last question. Just if you could give any color on the pricing of the HSDPA cards relative to the current UMTS cards, I mean, if you have any insights? Is it going to be sort of similar? Higher or lower? What do you expect on the profitability front for those? That's all I have.

Pat, this is Jason. I'll answer that question. I mentioned earlier that generally when-- in new product introduction cycles -- for myriad reasons, a couple include the increased capability of the card and also, typically, a higher cost to a new technology-- a card that incorporates a new technology -- you have a higher ASP earlier in the product life cycle. So I would anticipate that HSDPA PC cards would be priced higher initially than typical UMTS PC cards.

And from a profitability standpoint, certainly our goal is to make sure that our HSDPA products, PC cards, are as profitable as the rest of our PC cards in the lineup.

Deepak Chopra, National Bank Financial.

I was wondering in terms of the HSDPA pricing, if I take another stab at it, how would you compare as it starts in the early part of the life cycle versus previous-- previous products that started, in terms of like, for example, EVDO cards, the early part of their life cycle? How would you compare the pricing now versus back then? Are we materially lower or are we looking at the same type of pricing and margin structure?

You're probably looking at approximately the same pricing and margin structure.

So it's not going to be materially lower than what we've historically started at?

No, I don't anticipate that.

OK, fair enough. And just on the EVDO-- on the module side, given that it is a mini PC Express module, would it be-- is it fair to assume that the laptop vendors can switch out vendors relatively easy, given it is a standard interface protocol?

Yes. It would be fair to think that, at first blush, but the practical-- there's some very significant practical difficulties with that. I mentioned the product development process they have to go through, attaining things like FCC and other certifications. And those certifications are invalidated and have to be redone each time you swap out the underlying module.

So, in practice -- there's a theory and then there's a practice -- in practice and in our experience most OEMs will put a module in for the full product life of the product they've designed it into and their swap-out decision is usually made when they assess their next round of new product modules, rather than making a swap-out decision mid-life on an existing product and forcing themselves all back through regulatory and interoperability and carrier testing.

And is it fair to assume that you think the design cycle or the life cycle for your-- for any given laptop will be-- and your module will be similar to their redesign cycle, every 12 months or so?

Yes, I think the point, regardless of how long it is, the point is exactly what you just said. Our design win cycle, in practical terms, should be similar to the product life cycle of the product we're designed into, however long that is.

OK, fair enough. Any 10% customers in the quarter or how many?

We had two 10% customers.

They totaled about 29% of revenues.

Yes and that's actually an excellent question, because our customer diversity is improving each quarter. We're down to two 10% customers and those two totaled only 29%, which is well down from a peak of three more than 10% customers representing well over 60% of our business late last year.

So while it's difficult to have our gross revenues down on a year-over-year basis, we are serving a customer base that's much more diverse now, both on products, on geographies and on channels.

Is that also helping out your PC card margins, given the smaller volumes to more carriers?

Well, I think we're-- we haven't commented on that, but there's a reasonable inference that products tend to have volume pricing curves associated with them.

OK. We're running long on time here, so, operator, maybe we could take one more question if there is one.

Dev Bhangui, Haywood Securities.

I just had a quick question in terms of, I guess, the $2 million in terms of inventory decrease, did that come, essentially, from the CDMA side?

The $2 million of inventory decrease came from reducing inventories on core business products, that is, non-Voq products and we view that as a very healthy sign and we have not provided a breakdown, we're not-- and we're not going to provide a breakdown on exactly which product model those inventory decreases came from.

OK. Moving just in terms of, I guess, the gross margins and also the guidance in terms of Q3, Dave, you said earlier that there was a one-time supply chain hit in terms of 2 percentage points on the gross margin bringing it down to 31%. However the guidance for Q3 is also 31%. Is it a sign, as you explained, David, earlier, in terms of the volume negotiations with your [inaudible] guys or is that going to be the norm going forward, or is it going to improve when we start seeing the next generation products come back into the mix?

It's Dave, Dev. It's-- The Q3 number is affected by mix in the quarter. So that's why you're seeing margins in that area as we take on a higher-- an expected higher percentage of embedded module business in our mix in Q3.

OK. And I guess a couple of last questions. Your HSDPA shipments, which are going to commence in Q4 2005, do you expect the first shipments to go towards Europe or towards North America?

Well, we're targeting getting the product introduced in both markets, but we're not ready yet to declare which market will go first. I think we'll probably be in a better position to comment on that in October when we give our Q4 guidance and we're closer to the actual product introduction window.

OK. Any kind of guidance or color in terms of when you think-- just in terms of time-- timeframe is going to be the more-specific announcements on the laptop OEMs during the course of this year? And also, the EVDO Release A product that you're going to get some traction on, which you mentioned, in '06, is that coming from Asia or is that going to come from North America?

OK. On timing of laptop OEM product introductions and announcements around that, that's going to be driven by the laptop OEMs. We have said on today's call that we expect to commence shipments of EVDO mini PCI embedded modules by the end of this year.

And on interest in EVDO Release A, geographically we would expect that interest to come from North America and from Asia and we're already servicing EVDO Release 0 customers in both North America and Asia.

OK. One last question. Dave, you have said earlier in terms of the EM5625, is that-- is that shipment in terms of embedded modules is it to the rugged line of the products or is that to any kind of a consumer product?

In large part we would expect -- and the announcements we've already made on design wins on that product to date confirm this -- that that product's largely going to go to rugged industrial and specialty products. I will point out, it's the kind of product that could well be designed into wireless broadband OEM products such as the discussion we had a few minutes ago.

OK. I don't think I have any more. Thank you very much for your-- for your responses.

All right. Well, thank you, Dev, and thanks to all the callers. If we've cut anyone short here at the one hour mark, I apologize. Management is available at the company's offices at 604-231-1100 to take any followup calls, as usual, and thank you all for your time this afternoon.

Operator, we're all done.

Ladies and gentlemen, that does conclude the conference call for today. We thank you for your participation and ask that you please disconnect your lines.