司亞樂 (SWIR) 2004 Q4 法說會逐字稿

Ladies and gentlemen, thank you for standing by. Welcome to the Sierra Wireless Inc. fourth-quarter results conference call. During the presentation, all participants will be in a listen-only mode. Afterwards, we will conduct a question-and-answer session. (OPERATOR INSTRUCTIONS).

As a reminder, this conference is being recorded Wednesday, January 26, 2005. I would now like to turn the conference over to Mr. David Sutcliffe with Sierra Wireless. Please go ahead, sir.

Thank you, Operator. I am David Sutcliffe, the Chair and CEO of Sierra Wireless. And I have with me here at the Company's offices Jason Cohenour, the Chief Operating Officer, and David McLennan, the Chief Financial Officer. Our agenda for today, we will cover -- we will have Dave McLennan cover the forward-looking statements disclaimer -- Q4 and full-year 2004 financial performance. Jason will cover business highlights and a product road map update. We will return to Dave to look at Q1 2005 financial guidance. Finally, I will summarize. And then as usual, we will go to questions from the callers.

And with that, I will turn it over to Dave.

Thanks, David. I'll start with the disclaimer. Forward-looking statements involve risk and uncertainties, including but not limited to changes in technology and changes in the wireless data communications market. These forward-looking statements relate to among other things, statements about future market conditions, supply conditions, channel and then customer demand conditions, revenues, gross margins, operating expenses, profits and other expectations, intentions and plans that are not historical facts. Our expectations regarding future revenues and earnings depend in part upon our ability to develop, manufacture, supply and market new products, which we do not produce today and that meet -- define specifications.

In light of the many risk and uncertainties surrounding the wireless data communications market, we cannot assure you that the forward-looking statements discussed will be realized.

On to our financial results -- our results are reported in U.S. dollars and are in accordance with U.S. GAAP. For the fourth quarter of 2004, revenue was 58.8 million; operating expenses were 15.8 million. And net earnings were 7.3 million and diluted earnings per share were $0.28.

Our results relative to the Company guidance provided on October 27, 2004 are as follows. Fourth-quarter revenue was 58.8 million, lower than our guidance of approximately 63 million. Gross margin was 38.8 percent, below our guidance range of 39 to 40 percent. Operating expenses were 15.8 million better than our guidance range of 16 to 16.9 million. And net earnings of 7.3 million were diluted earnings per share of $0.28 --were lower than our guidance of approximately 7.7 million or diluted earnings per share of $0.29. As well, cash flow from operations was 5.8 million, consistent with our guidance of positive cash flow.

Our revenue in the fourth quarter relative to guidance was affected by a lower than expected, although, still robust demand for PC cards. This was partially offset by better-than-expected results in our embedded module business. Our gross margin percentage varied principally, as a result of product mix.

Looking at Q4 performance relative to Q3, Q4 revenue was relatively flat at 58.8 million, compared to 59.1 million in Q3. Gross margin percentage also remained flat at 38.8 percent, compared to 39 percent. Operating expenses were 15.8 million in Q4, compared to 15.4 million in Q3. And net earnings were 7.3 million or diluted earnings per share of $0.28 in Q4 compared to net earnings of 7.1 million or diluted earnings per share of $0.27 in Q3.

Looking at our balance sheet relative to September 30, 2004, our cash position increased by $4.3 million to 131.8 million, driven primarily by a strong cash flow from operations. Trade accounts receivable, DSO, decreased to 27 days from 38 days, well below our target of 60 days. And that inventory increased 11.1 million from 8.6 million. This increase largely reflects an increasing component in finished goods inventory.

Comparing fiscal year 2004 to fiscal year 2003, revenue increased by 108 percent to 211.2 million in 2004, compared to 101.7 million in 2003. Gross margin percentage decreased to 39.6 percent from 40.5 percent. Operating expenses were 56.3 million in 2004, compared to 39.7 million in 2003. And net earnings were 24.9 million or diluted EPS of $0.96 in 2004 compared to 2.3 million or diluted EPS of $0.12 in 2003.

I would now like to pass the call over to Jason to review business highlights.

Thanks, Dave. I'll begin with business development highlights in our CDMA product lines. First, we announced a distribution agreement with Premier Wireless Solutions, a distributor of wireless machine-to-machine fixed and mobile communication products, for our embedded modules. Under the agreement, Premier Wireless will distribute and offer integration support services for our EM3420 embedded module.

We also announced that our AirCard 580 for EV-DO is now available in Australia for use on the Telstar network. Commercial shipments of the AirCard 580 to Telstar also commenced during the quarter.

In technology agnostic developments, we announced the introduction of AirCard Manager, a Web-based software solution for the AirCard product line designed to support enterprise, IT departments that require an automated seamless approach to remote software installation and end-user configuration.

Moving to GSM developments, we announced that our new quad-band AirCard 775 for EDGE networks was available for purchase from AT&T Wireless; now of course, part of Cingular Wireless. The AirCard 775 is the only class 12 device on the market that provides users with data rates up to three times faster than on GPRS networks.

In a related development, we are currently engaged with a number of EDGE infrastructure providers who are preparing to launch class 12 upgrades for select wireless carriers. These infrastructure manufactures are using the AirCard 775, as their class 12 reference device for interoperability testing.

We also announced at the MP 775 rugged wireless modem is now available in the United States for use on the Cingular Wireless EDGE network. And like the AirCard 775, the MP 775 delivers quad-band functionality for worldwide connectivity in data transfer rates averaging 100 to 130 kbps. We also commenced commercial shipments of the MP 775 during the quarter to carrier channels and government customers.

Moving on to Voq, we announced that the Voq Professional Phone is now available from USA.net. USA.net is offering the Voq for $299, when bundled with a voice and data service plan from Cingular Wireless, now hosted Microsoft exchange, or Pop Three Email Service Plan from USA.net.

In other developments and subsequent to quarter end, we announced two new appointments to our senior management team. First, Dan Schieler has been promoted to Senior Vice President of Worldwide Sales, responsible for global sales, distribution and presales technical support functions. Dan joined us in August of 2003, when we completed our acquisition of Air Prime, where he was VP of Sales and Marketing. Since joining Sierra Wireless, Dan has served as our VP of Sales for North America and OEM, where he helped grow revenue more than 100 percent in the past year.

In addition, Trent Punnett has joined us as Vice President of Marketing. Trent will be responsible for our global marketing initiatives, including strategic marketing, product line management and corporate communications. Prior to joining Sierra Wireless, Trent was VP of Marketing and Product Line Management at Motorola Next Level Communications, a manufacturer of broadband communications equipment.

I'm going to move now to segmented financial performance, starting with revenue by products segment.

In PC cards, revenue was down slightly to 62 percent of Q4 sales that is versus 67 percent in Q3. OEM revenue was flat at 31 percent of sales versus 30 percent in Q3. Revenue from MP products was up significantly to 4 percent of revenue versus 1 percent in Q3. And Voq revenue was up modestly to 2 percent of revenue versus 1 percent of revenue in Q3. Revenue from other categories was flat at 1 percent.

During the quarter, we completed our shipments of embedded modules to palmOne for the Treo 600 program. As a result, we expect that OEM, as a percentage of sales, will trend lowering in the near term. For recap of the year, PC cards represented 59 percent, OEM was 33 percent, MP was 5 percent, Voq was 1 percent, and other was 2 percent.

Moving to revenues by distribution channel. Revenues through carriers represented 17 percent of our revenue in Q4 versus 23 percent in Q3. Revenue through resellers, including those who sell to carriers, represented 51 percent of our sales in Q4 versus 47 percent in Q3. Revenues through OEM's represented 31 percent in Q4 versus 30 percent in Q3. And revenue that was sold direct represented 1 percent in Q4. For the year, carriers represented 24 percent, resellers 41 percent, OEM's 34 percent, and direct and other 1 percent.

Moving to revenues by technology. Revenue from GSM products tripled in Q4 to 25 percent of revenue versus 8 percent in Q3. Revenue from CDMA products was down slightly to 74 percent in Q4 versus 91 percent in Q3. And revenue from other was flat at 1 percent. During the quarter, we experienced a significant technology mix shift, driven principally by the launch of our new EDGE products. This shift has significantly reduced our technology dependence on CDMA.

For the year, GSM represented 16 percent of our sales, CDMA 83 percent and other was 1 percent. Revenue from new products, that is new product that has been introduced in the past 24 months, accounted for 82 percent of our revenue.

Moving to revenues by geography. Revenue from the Americas softened slightly to 80 percent in Q4 versus 91 percent in Q3. Revenue from Europe doubled to 10 percent in Q4 versus 5 percent in Q3. Revenue from Asia doubled as well to 10 percent in Q4 versus 4 percent in Q3. So while revenues from the Americas softened slightly in Q4, Europe and Asia had their strongest quarter of the year. This geographic mix shift was driven mainly by improving CDMA strength in Asia and sales of EDGE PC cards in Europe. For the year, the Americas represented 89 percent of our sales, Europe 6 percent, and Asia-Pac 5 percent.

I'll now provide an update on some demand trends we're seeing in the marketplace. Starting with PC cards -- demand for our PC cards products continue to be strong in Q4, although fell short of our expectations. Our channel partners reported record sell (indiscernible) to our products to end customers during the fourth quarter. Notwithstanding the strong end-user demand trends we're seeing, some of our channels have told us that they have sufficient inventory of our EV-DO and EDGE PC cards to meet short-term demand. This, combined with the launch of competing products will negatively impact the short-term channel demand for our PC cards products.

Our revenue in North America softened in Q4, while sales from Europe and Asia-Pac grew 100 percent, Q3 to Q4. As mentioned earlier, our business in these regions benefited from the AirCard 775 launch and sales of both our 1-X (ph) and EV-DO yellow (ph) PC cards. Some EDGE operators are also now testing class 12 service, which delivers significant improvements in data transfer speeds. We stand to benefit from the commercial launch of such services, as the AirCard 775 is the only available class-12 EDGE PC card on the market.

On to embedded modules. We have completed shipments of CDMA 1-X embedded modules to palmOne for the Treo 600. We don't anticipate additional orders for palmOne for this product. And as a result, we expect embedded modules, as a percentage of our revenue mix, to trend down significantly in the near term. In January, we announced design wins with two of our existing laptop customers, Panasonic and Itronix for our new EM5625 EV-DO embedded module. We expect to commence commercial shipments of the EM5625 during the first half of this year.

On a related note, we are seeing growing interest from Tier 1 laptop OEM's in 3G embedded modules. We're investing in business development and product development to address this growing demand.

On Voq, sales of the Voq Professional Phone continued in Q4. And we are continuing to see the corporate enterprise market and are engaged in a number of trial deployments. Based on feedback from the trials, we are continuing iterative improvement on the product. We continue to receive favorable customer feedback on the product concept and strong media reviews.

Voq represents a new category entry for the Company. And we are being patient, as we work through typical new category adoption challenges. We remain strategically committed to the VOQ product line and believe in its differentiation and positioning. We continue it -- to invest in the category and believe that over time, Voq will be a significant, positive contributor to our financial results. We're not expecting or counting on an overnight success.

Today, also for the first time, we are providing some insight into our product road map initiatives. So moving to that topic -- during 2005, we expect to increase our research and development efforts to ensure that we're well positioned with new products that take advantage of market opportunities associated with the deployment of 3G networks. Specific product development initiatives under way include UMTS HSDPA PC cards. We have accelerated the development of HSDPA PC cards, which we expect to launch in the second half of 2005. Given the compelling advantages and industry momentum of HSDPA, we no longer intend to bring to market a first generation UMTS product.

We're also developing CDMA EV-DO Release A PC cards. Release A brings significant uplink and downlink speed advantages to EV-DO. We've established a strong leadership position in EV-DO PC cards over the past 18 months. And we intend to extend our leadership position by accelerating the introduction of Release A capability and speeds.

We're also developing embedded modules for laptop manufacturers and other OEM's. We intend to capitalize on our strong embedded modules' experience and track record. Then, meet the growing demand we see from Tier 1 laptop manufacturers for embedded 3G capability. We will meet this demand by developing products and services specifically for this category.

And on Voq Professional Phone, we're developing the Next Generation Voq platform to take advantage of 3G high-speed networks, such as HSDPA. Our Next Generation of Voq phones will bring significant AirLink, feature and ID enhancements to the product line. We expect the first of these products to be available during the first half of 2006.

I'll now pass the presentation back to Dave for Q1 guidance.

Thanks, Jason. We are providing guidance for the first quarter ending March 31, 2005, which reflects our current business indicators and expectations. Inherent in the guidance, our risk factors that are described in detail in our regulatory filings. Our actual results could differ materially from those presented below. All figures are estimates based on management's current beliefs and assumptions and are subject to change.

Following our considerable revenue and earnings growth in 2004, we expect a significant reduction in our business in early 2005. We expect revenue in the first quarter of 2005 to be approximately $40 million less than in Q4 2004. This is partially a result of lower sales from embedded module products that will not include further shipments to palmOne for the Treo 600. We all expect a Q1 demand for PC cards will be lower, as a result of channel inventory levels at some of our channel partners that is already sufficient to meet their short-term customer demand and the near-term impact of increased competition; and therefore, result in loss of market share in the EV-DO PC cards market.

During 2005, we also expect to increase our research and development efforts to ensure we are well positioned with new products that will take advantage of market opportunities associated with the deployment by carriers worldwide of 3G networks. For Q1 2005, we expect revenues of approximately $18 to $20 million. Gross margin in the range of 34 to 35 percent. Operating expenses in the range of $16.5 to $17 million and a net loss between $9.2 and $9.9 million. This translates into a loss of $0.35 to $0.38 per share. As well, cash flow from operations is expected to be negative.

With that, I'm going to pass the call back to David to wrap-up.

Thank you, Dave. First, I'd like to summarize that -- 2004 and Q4. 2004 was a year of extraordinary growth and profitability for Sierra Wireless. Our revenues for the year are 211.2 million, up 108 percent over 2003 revenues, while earnings in '04 were 24.9 million, up from 2.3 million the previous year. Revenues for the fourth quarter were up 70 percent, compared to the same quarter in 2003, while earnings were 7.3 million, up from 1.9 million the year before.

In 2004, we have enjoyed a strong demand for PC cards, as a result of being first to market with our EV-DO PC cards and as a result of launching our new EDGE PC cards. Our design win with palmOne to supply CDMA modules for the Treo 600 was our most successful OEM design win to date. And while shipments on that particular product have been completed, the project has further established Sierra Wireless as a world-class supplier in the OEM market.

Finally, on 2004, we leave the year with a very strong balance sheet, cash of over $131 million.

Turning to the first quarter of 2005 and beyond that -- as has been mentioned, we've more than doubled revenues in 2004. We expect 2005 sales to get off to a considerably slower start. Our goals for 2005 will be two-fold. The first -- to significantly improve our top and bottom-line results, as the year progresses. And second -- to develop the right products at the right point in time to enter the market and capture the opportunities and the early to market opportunities I'd like to emphasize, which we believe will follow-up from a deployment of 3G networks and new protocols worldwide.

On this call, for the first time, we've provided a product road map that includes HSDPA and EV-DO Release A PC cards, 3G embedded modules targeted at Tier-1 laptop manufacturers, and the Next Generation of the Voq Professional Phone, which will bring feature, size, and AirLink enhancements to the product.

With that, I'd like to wrap-up and have the operator queue up questions for us. Thank you.

(OPERATOR INSTRUCTIONS). Pat Chiefalo, Merrill Lynch.

My first question, David, can you break out of the 40 million decline, how much is coming from the various issues you point out -- i.e., from palmOne -- the decline -- the decline in PC cards, can you sort of help us understand the details of that decline?

Yes, in 2004 and in the fourth quarter, particularly, our OEM revenues were 31 percent of the $58.8 million of reported revenue. And that means OEM revenues were approximately $18 million. PalmOne was by far the largest component of this. And that revenue, as we have said, and as we indicated in our guidance in October, came to a conclusion in the fourth quarter when we do not have additional orders; and consequently, no revenue from palm in Q1. So that's the first component and would be a very significant component of the 40 million you mentioned.

The second component is lower demand for PC cards, both in the EV-DO and EDGE arenas. And I would emphasize that we don't believe this is lower in customer demand, given the sell through data we have received on the fourth quarter. It's a simply lower channel demand, as a result in most cases of channels indicating to us at that they already have sufficient inventory for near-term demand -- and to a lesser degree, the introduction of our first competitor in the EV-DO PC cards market.

Okay. Just secondly -- can you help us characterize or understand in greater detail how the shortfall sort of came about throughout the quarter? Was it -- did this sort of issue of a 40 million shortfall evolve throughout the course of a quarter or was it a matter of one or two calls from a couple of carriers and then you sort of immediately realized that -- what was going to happen in the next quarter? I just wanted to sort of understand what happened during the quarter and how it transpired.

Okay. That is a fair question. Before the quarter began, on October -- sorry, August 31st, we announced that we expected to -- that we did not have future design wins with palm in hand. And so we expected the embedded modules portion -- the palmOne portion of our embedded modules revenue to tail off. In our October conference call, we updated that information and indicated that we expected to make our final shipments to palmOne during the course of Q4 and to have -- then, we had no orders for Q1. So that information -- and that is a significant part of the number you're talking about -- was in fact in the market before the quarter started.

As the quarter progressed we did work on shipments -- on manufacturing and shipments to palmOne. And they went well into the quarter before we got them finished. In terms of the other areas -- the PC cards channel demand -- that's something that we get very late in the process, as our partners and we go through our forecast update processes. In our case, in readiness for providing guidance.

Okay. So they came towards the end of the quarter?

Well in fact, to be clear, most of that information -- that is Q1 orders and information that we would base our Q1 guidance on -- that normally comes after the end of the quarter.

Sure, okay. And my last question just looking forward -- could you sort of characterize the type of visibility you have looking even beyond Q1? I know that may be hard to give. But do you have any sense of sort of customer orders or the trend of those orders after Q1? I mean, is it a matter of sort of having orders in the pipeline and having to ship those in the back half of the year? Or at this point, are we waiting to sort of hear new order rates from customers? I'm just tried to get a sense of how the revenue trend could go beyond Q1.

Right. Well, the first thing that we get before we get an order from most of our larger partners -- we have a rolling monthly demand forecasts -- and that's our first leading indicator on demand in future quarters. And then, we work to translate those forecast demand numbers into actual orders. And at this point, we have relatively low visibility on Q1 channel demand -- and lower than has been the norm in recent quarters. And that is one of the reasons that we have determined it's appropriate to set guidance for Q1 quite low.

It is way too early to know what Q2 and Q3 will look like. But we certainly have forecasts that would support -- very good -- in customer sell through. And you have to believe -- if you got good underlying and customer sell through, you're going to eventually receive orders and support of that demand, as channel inventories work through.

Andrew Lee, TD Newcrest.

Looking first at the operating expense guidance -- where are you guys running the business model now in terms of breaking even? What I'm getting is at -- you have got an increase in your Op-Ex, your revenue has drift of cline (ph) so significantly. Is there a point where similar to back in 2001, you would consider looking at tailing back in some of these initiatives that you have in the pipeline right now? Or do you just foresee this as being a short-term blip?

Andrew, to the first part of your question, our long-term operating model has not changed. And so you can do some pretty basic math and see that our operating break-even point would be somewhere between $40 and $45 million a quarter of revenue run rate.

To the second part of your question -- is there some point where we would consider reducing Op-Ex or curtailing initiatives? We're not at that point today. We are coming off of a year, where we grew our topline 108 percent and moved earnings from 2.5 million to 25 million. We have got a quarter in front of us, where our leading indicators have caused us to set our guidance very low. But we are not prepared at this point in time to believe that that means that long-term demand is somehow weak.

Okay. And looking at the Voq, did I hear you right? In terms of the next version coming out -- is that the HSDPA in the first half of '06? Or can we anticipate something ahead of that 3G product launch?

Well, we indicated both in fact. That we're continuing iterative (ph) improvement of the existing Voq product. And that we are developing new Voq products for 3G networks, including, HSDPA, which we mentioned specifically.

Okay. And then David, to interpret iterative (ph), does that imply new -- could that potentially imply a new wireless protocol?:

It could. And I'm -- we're not ready to make a call on that one way or the other here. Or we would have already come out and said it.

Okay. And this is the last question -- when you talk about the channels having record sell through in the context if knowing that NovAtel was going to enter the market -- it just seems to me that either -- it's on your shipment and/or on the customer order -- and have this size of a hold down in sequential revenue. There could of been aggressive ordering -- or you guys did not anticipate the impact of the competition. Can you just flush out in a little more detail as to where you are putting the blame, so to speak, on the sequential drop in the PC cards business?

Right. Well, we -- I wouldn't know that I would call it blame, but I think I understand your question. The analysis we have done and the explanation we have already offered on the call is that the primary cause of the lower guidance is specifically in the PC cards area. And I'd remind you that is across both EV-DO and EDGE and multiple partners. It is not specific to only EV-DO or to a single carrier partner -- that the primary cause is channel inventory. And a secondary cause is the effect of -- the anticipated effect of a new competitive entrant in EV-DO, where until now, we've enjoyed very close to 100-percent market share.

And so any new competitive entrant who wins anything other than zero sales is going to take some market share. And what market share they will be able to take is yet to be determined. And I expect will be the focus of some very strong competition between the companies, as the quarter progresses.

Okay. It just seems a little bit of a disconnect when Verizon is rolling out what seems to be a pretty aggressive road map with 32 markets. A lot of new well, -- but dozens of new markets in January and in channel inventory. So that's it's going to get at -- whether it was forecasting on their end or your end, as to what has caused this channel build-up -- if indeed the end market demand continues to want these products.

Right, well to be clear, we are comfortable with the end market demand, as reported by our partners. And I would point out to you that Verizon, we have been working with them now on EV-DO for approximately 18 months. They've been rolling out markets in stages over that period of time. And the latest increase in their footprint in the new markets they are rolling out is something that -- well, I know we have been planning for -- for some time, and I'm sure they were.

And like most channels, you would expect that as the channel prepares to launch a new product or to expand its distribution footprint -- that you would expect that channel to build inventory before it turns on those new markets. And so, I don't think -- although it sounds like there is a disconnect -- in the way you phrase the question, I don't think there is. And we certainly would not attribute the sequential decline in our revenues to any forecast issue -- in terms of future demands from any of our partners.

Mike Walkley, Piper Jaffray.

Trying to -- just build on some of the questions around thoughts for 2005. I could see how HSDPA and EV-DO reve (ph) getting to market early can help the back half of the year. But just maybe thinking to the June quarter, David, maybe -- and Jason, maybe can share with us some of your thoughts on how that starts to go up significantly, sequentially. Would it be the EDGE class-12 deals you're working with? Or do you think EV-DO inventory starts to burn off and picks up again in June?

It's a few factors, Mike. This is Jason. I mean -- clearly, we don't see end-user demand subsiding. In fact, we have seen it grow every quarter in 2004. We've got no reason to believe that is going to start to fall. So our belief is that we will eat through channel inventories and at the end of the day, at some point, result in reorders. So we expect to rebound on PC cards sales, as we go through the year.

And in addition to that, we've already announced and will launch a new embedded module product in the first half. We expect that will enable us to grow revenue as well. And in the second half of the year, as you pointed out, we've got more new products coming to market. So I think we've got a pretty robust product pipeline going that will help us rebound revenue and earnings. And the underlying customer demand is good.

Okay. Thanks. And then just going to the EV-DO channel -- you had that original 29 million contract with Audiovox. Have you shipped against all that already? Or is there any more to be shipped on that contract?

We have fulfilled that order, Mike.

Okay, great. Thanks. And I just wanted one more clarification, and I'll pass it off. On the Voq Phone, you said, a brand new platform of them is coming out the first half of '06. But should we expect continued improvements to the one during the course of the year. Does that mean you'll have new product announcements on Voq before this new platform in '06? Am I understanding that correctly?

We are not going to forecast or give any guidance on specific product announcements on the current generation platform. But what we have said is that we're doing iterative improvements to the platform.

Okay. And then a new platform will be the first half of '06?

That's right. The NextGen platform -- first half of '06.

Paul Coster, J.P. Morgan.

A few quick questions on the same old subject please -- can you give us some sense of how the carriers generally think in terms of week of inventory and how that changed coming out of the fourth quarter?

Well, I will touch on that. And I don't think it is going to be easy to answer that in generalizing across all of the carriers. You know, we have carriers in different regions of the world and on different AirLinks. And they each have their own practices.

I think the idea that you probably can take away is that when carriers are doing initial product launches, they build inventory above their normal run rate guidelines. And then, as that service launches and the products start to sell and they establish what the actual demand will be, the demand normalizes after the launch. Then, they start calibrating inventory down from the larger level that they have at the time of the launch to -- perhaps, in our product category, something like 6 to 8 weeks of inventory at a normalized run rate.

It sounds to me like -- because there's a little bit of a surprise here -- that maybe they've overshot on their downside from your perspective.

Yes, I don't think that, as I have said, I think I said already -- we don't believe that carrier estimates of demand or sell through are a root cause of why we have channel inventory in place.

Okay. Let me ask you a slightly twist on this question then. At least with one of your customers, the competitor has come out with a card that sensibly has the advantage of the latest chipset, one step ahead of yours, for a brief period. Is it just a binary decision at the point for the carrier, as to which card they carry? Or can they continue to sell your card that perhaps is now sort of -- part generation behind?

Well, yes, I think that -- I'm going to have to read behind between the lines on your question a bit there, but the product that we sell in the EV-DO Arena runs on a currently -- an actively supported chipset platform from QUALCOMM. And that platform -- we and QUALCOMM have continued to improve. And we introduced a new performance enhancement release to that product just during the late part of the fourth quarter.

We are also very familiar with the other chipsets in QUALCOMM's product line. And you mentioned, competitor is coming to market with a somewhat newer chipset. In fact, the EV-DO embedded module that we have just announced -- and announced design wins with Panasonic, Itronix and Atrode (ph) -- that product uses the newer MSM 6500 chipset. And finally, we are already in very active development on EV-DO Release A, which leapfrogs all of these Release 0 chipsets of various flavors, and moves the capability of the service and the product up a whole another step.

I understand -- I guess the question was with respect to the current inventory in the channel --

I'm sorry. I do recall that, and I didn't answer it directly. I will now. The normal practice in our product categories is for carriers to approve two -- and occasionally more, but usually at least two products on a given AirLink generation, and to carry those products in their channels side-by-side and simultaneously. Now, the announcements -- Verizon, in fact, announced their intention to do just that last March, March of 2004, when they indicated they planned to add a NovAtel wireless product to their offering in later 2004. And so, I think that's news that has been in the market.

Okay. And finally -- do you have any 10-percent customers in '04 that you can describe to us?

Dave can speak to that.

In Q4, we had three customers greater than 10 percent, and those three customers totaled 61 percent of our Q4 revenues. Sorry, 62 percent.

Eric Zamkoff, IRG Research.

Three quick questions -- first of all, wouldn't it be a fair statement to say that considering the demand that you are seeing in Europe in the market overall, that it's not the PC market -- card market that is slowing? It's the fact that you are not a player right now in the UMTS side of the market that is the biggest contributor to the shortfall.

Then secondly, why are you skipping UMTS in general? Because I'm hearing from other card suppliers in Europe, especially Vodafone, that there has been a lot of interest there. And then finally, third, will your HSDPA card be backwards-compatible with EDGE, and what chip is it using? And then, finally, when do you expect to see HSDPA chip availability?

Okay. We'll work through those. I think that was more than three questions; I'll do them as quick --

Three and a half. Sorry.

Okay. First, you asked is our lack of near-term demand a function of the fact that we don't have a UMTS product? And you could look at it that way, as you suggested, but that would not explain the sequential difference between our Q4 revenue run rate and our Q1 guidance.

Secondly, you asked, why are we skipping UMTS, particularly in light of the subscriber numbers and success that carriers in Europe such as Vodafone are reporting. And we have never planned to skip UMTS. And earlier in the call, Jason indicated that we are accelerating our UMTS development and focusing it entirely on HSDPA -- which we now believe has significant industry momentum and critical mass, both in North America and in Europe -- and that's an initiative that you will be hearing more about us from.

You asked about which chipset we're using -- whether it will be backwards-compatible to EDGE and so on, and you'll be hearing more from us about the details of that product, its feature sets and the chipset it uses in coming months. But we are not prepared to spell that out in a lot more detail right here today, for competitive reasons.

And the final question you had was related to the timing of availability of HSDPA chipsets, and we have indicated that we expect to bring HSDPA products to market. And the timeframe we plan to do that in, I don't want to try and comment on individual underlying chipset vendors' availability schedules.

Let me make sure I'm reading you correctly. You don't characterize it as skipping UMTS. You're moving to HSDPA. The problem is carriers today are rolling out UMTS. Are you saying that you're not going to have any participation in plain UMTS, and therefore you're going right into HSDPA and that is a back half of the year possibility? If you're going to have a card in the back half, when do you think you're going to actually have sales related to HSDPA?

Okay. So the first thing that I should make clear is that when you distinguish plain UMTS and HSDPA, HSDPA products and networks and certainly the work that we're doing on our product will produce a product that is backwards compatible to plain UMTS networks and services. And so, we're not going to skip plain UMTS, but we are going straight to a product that is capable of not only plain UMTS but also HSDPA.

And our read of carrier and infrastructure vendor investments and deployment plans for this year and next year -- and this is of course, our opinion -- and it is that HSDPA is gaining a tremendous momentum and will become a very significant opportunity for us when it goes to market. And you might -- I can just add to that, plain UMTS notwithstanding the fact that it's 3G in a definitional sense, it actually has a relatively modest data speed. It's not even as capable as the EV-DO Release A products that we are already shipping today. And so we think that while plain UMTS is an entry point for those European operators, the thing that's really going to get things moving both in Europe and North America is a much higher speed capability, which HSDPA will deliver.

Gotcha. And finally, and I promise this is it. Recent QUALCOMM commentary hinted that the HSDPA chipset from them won't be ready until late 2005. Are you seeing the same kind of scenario?

I don't have any comment on that.

Gus Papageorgiou, Scotia Capital.

Yes, clearly, Q1 guidance here is a little disappointing for the market, I think, both for yourselves as well. I was wondering, given the current process of viewing end market sell through and supplying the channel, obviously, there's been a disconnect here if it's only temporary. But is there anything you can do, you think, to improve that process of viewing end market sell through? And then, how you supply the channel to kind of improve that process and avoid any kind of future negative surprises at this level?

Well, Gus, I appreciate your empathy there. And I will tell you that we have surprised to the upside a number of times using the same process. And now, we have disappointing guidance to offer at this stage in the quarter, using the process we've run our business on for quite some time. So the substance to your question -- are there ways we can improve our business and our processes, so we have less volatility -- especially less volatility to the downside? That's certainly a question that we will have more of a look at. But at this stage in time, I think we've got a proven process. It is producing a prediction today that none of us like. But that's the way it is right now. And we'll see if we can make that result better and we'll also see if we can make the process better.

Just a question on the HSDPA development -- have you been working with any of the major equipment vendors so far? Are you going to be engaging with them?

Yes, we have. And yes, we expect to.

Multiple, or just one?

Multiple.

John Grandy, Orion Securities.

Thank you, I have a couple of questions on the Q1 guidance. First, relates to the gross margin. I would have thought all other things being equal with the revenue mix involving less OEM product and more PC card product that the gross margin would rise. Instead, you're expecting it to decline. So can you tell us -- is there an issue there of working through some of your own inventories? Or are there other factors that go into that?

John, it's Dave speaking. I think one of the factors that is putting pressure on gross margin in Q1 is that the volumes are significantly lower. And by virtue of having some fixed operating costs in the gross margin calculation, as those costs get spread out over lower volumes, it does put pressure on gross margin. The biggest driver of that is volume related.

To what extent -- your manufacturers -- to what extent do they have volume guarantees from you and to what extent can you just turn off the tap when you need to?

Well first, the (indiscernible) allocations that Dave's referring to are internal to Sierra Wireless' cost structure, rather than external in our supply chain -- in large part. There things like amortization on test equipment, which is a capital investment for certain level of capacity. And if you have to amortize it across a smaller number of units, of course, it impacts your short-term GM. The good news is that it cuts both ways, and when your volume picks up, your GM recovers automatically.

So -- and then the second part, with our supply chain we do have supply agreements with key supply chain members. And in those commitments -- in those agreements, we may commitments on a purchase-order basis. And the farther out you go, the more flexible and more leeway we have in varying those commitments.

I understand. So then, when I look at the earnings forecast for Q1 -- I wonder if you could fill us in on what your expectations are for income taxes? Are you expecting to have a recovery? Or are you in a position where that is not possible?

In this loss situation, John, we will not be in the income tax recovery scenario. We will have to probably book some small amounts of large corporation tax -- but fairly minimal stuff.

I understand. And finally, can you talk a little bit about levels of price competition that you're seeing or expect to see in the PC cards with the entry of a competitor into that marketplace?

John, this is Jason. I think whenever you have got a situation where there's growing customer demand and the entry of new competitors, there's going to be pressure on price. And we are seeing that in some of our markets. That's one of the key reasons, candidly, for increasing investment on Next Generation platforms because ASP pressure tends to be less on newer platforms.

I understand. Any restructuring or one-time charges that you anticipate in Q1 or indeed in 2005?

We have no restructuring or one-time charges in the '04 financials we gave you, none in the Q1 guidance, and none anticipated at this time.

Hasam Imam, Thomas Weisel.

This is Rakesh (ph) on behalf of Hasan. I see -- I wanted to go to the Sprint channel. Are you guys seeing any declining trend in the Sprint channel for one (ph)? Or a --

Well, you know, I can't really comment on what our specific customers are seeing in terms of sell through. Sprint continues to be an important customer for us. We have made significant shipments to Sprint over the course of the year for 1-X PC cards. And as you know, later in the year, Sprint announced an intent to deploy EV-DO service. And we think that we are strongly positioned to benefit from their deployment of EV-DO.

Okay, and can you comment on any up-tick on the Verizon front?

In what respect?

I understand like, of course, going forward you'll see the demand may be being impacted from Verizon but in fourth-quarter -- was there any decline, compared to previous quarters for your cards shipments?

Well, again, I want to reinforce what Jason said a moment ago. We don't make comments on individual customers' business trends that is definitely not our business to do. If you want to get an update on volumes that Sprint or Verizon -- we definitely have to point you at their Investor Relations executives.

Our overall observation for Q4 and CDMA carriers of which Sprint and Verizon are both -- CDMA business represented over 75 percent of our total revenue in Q4. And in aggregate, our channel partners reported record sell through of our products during Q4. So I can't answer -- we won't answer questions about individual partners, but in aggregate that is the trend. And CDMA carriers made up a large percentage of that trend.

Chris Guidi, SG Cowen.

Multiple part question -- just want to get a better feel for the excess inventory. Any danger that it could impact pricing above and beyond what you're seeing in terms of normal price pressure or any risk that you may have to write it down or take it back? Then, second and third questions are on Cingular -- just want to now if you expect continued orders from the combined Cingular/AT&T going into 2005? And on increased competition at Verizon, NovAtel has been public -- we know that NovAtel is a second competitor there. But do you expect further competitors entering Verizon over the course of '05?

Okay. Well, on the first question, which was with access channel inventory -- do we expect that to create pricing pressures or returns -- where the channel owns the inventory, the pricing of our product out to the end customer is in their control, not ours. And so, we just -- I can't answer that. That would be a function of the carriers -- or sorry, the channel's decisions. But the second part of your question --

I mean regarding when you expect to get reorders from them --

Right.

Are they likely to be more aggressive in pushing you for pricing, since they have more product than they need currently?

Yes. I think the -- if you look at what we've reported, which is that end customer sell through appears to be strong and the channels have inventory that means they don't need to order from us right now. And that's why we are reporting a lower guidance for Q1.

So that's less a question of pricing than it is of getting the orders -- when we get the orders. We expect to get orders when channel inventories are down to levels that those channels are ready to place reorders. The pricing pressures in our business usually come from our competitors. And that's where I would suggest you focus on trying to understand the pricing dynamic.

You asked about -- do we expect to continue to sell to Cingular now that AT&T and Cingular are combined -- or AT&T Wireless, I should say. And we definitely expect to do that and with both the AirCard 775 and the MP 775 in the fourth quarter, we commenced shipments through Cingular's channels. And we see the merger of Cingular and AT&T Wireless services as an opportunity for us to go from one proven partner at AT&T Wireless to an expanded relationship with Cingular as a whole.

You then asked about Verizon and the increasing competition there. You mentioned NovAtel, and you asked if we expect additional competitors. Well, we are not sure what to expect there. We have heard rumors of additional competitors beyond NovAtel coming to market.

And I think that's really a function of what Verizon decides about whether they want to have two or three vendors or more in a particular category. They do have some practical maximum limit of the number of vendors in a category. Particularly, a category like this, which is a specialty category. Because it's very significant cost to the carriers to qualify and support each new product that it puts into its channel.

Glen Tracey, Pacific.

Just a follow-up on the Cingular one -- given that you are shipping to Cingular, do you know of any reason that would explain why Cingular doesn't appear to be advertising Sierra's EDGE card on their Web site, whereas they are advertising the Ericsson product?

Yes, I think in a nutshell Cingular has just gotten our EDGE AirCard through the combination with AT&T Wireless services. And I believe if you look at the AT&T Wireless services channel information, you would see our product.

Okay. With regard to the EV-DO Release A -- I am sorry but I think I missed the release timing that you might have mentioned for that product. Can you repeat that, please?

We did not provide a release timing on EV-DO Release A other than -- I will tell you in qualitative terms, we're planning to be first to market and enjoy a very strong first to market position on Release A, as we have already enjoyed for quite some time on EV-DO release zero.

It's fair to say that it is in 2005, though?

It is fair to say -- we expect to be first to market.

Okay. Voq volume -- product volume approximately doubled in the quarter, can you give us some sort of idea of what you're expecting in terms of volume going into Q1?

Glenn, this is Jason. Thanks for pointing out that Voq volume doubled. It was, of course, doubled off a very modest number. And just in terms of directional comments on Voq, we are being very patient with the product. We are continuing trial deployments with corporate enterprises, which are going well. And we are continuing to go through iterative improvements to the process along the way. In the fullness of time, we expect that is going to result in benefit, of course, to sales growth. But we're not going to give forward guidance on specific product categories.

Okay. With regard to the question that was asked previously about manufacturing agreements -- do you expect the lower channel demand for the PC cards and existing manufacturing agreements are going to require you to increase inventory during Q1? Are your agreements flexible enough that you can keep the inventory relatively constant in Q1?

I think we are going to have to see how that plays out. We are still early in the quarter. We do have a fair amount of flexibility at the 60, 90, 120 days level. But I'd like to see how the quarter plays out before we try and predict that closely. We do have inventories at a pretty -- what I would characterize as a pretty safe and healthy level coming off of 2004, $212 million of annualized revenue. Inventory right now is fairly modest at $11 million.

Okay. Last question I have -- and this is kind of back to your outlook for Q1 is -- I'm just trying to understand exactly what kind of information your channels give you. I understand that you're getting some sell through information. And they are telling you that they have enough inventory for the short-term, but I guess what I'm wondering is -- do they tell you in more detail what the actual inventory level that they still have is? And are you then able to use trends on sell through to predict for yourself when they are going to run out of inventory and when reorders are likely? And if that is the case, do you have reason to believe that this is something that is relatively localized to Q1? And there should be a reasonable change in Q2?

Okay. So I think there are two substantive points to that question. The one is -- what is the process and what data do we get visibility on? And the other is -- assuming we have got our own ability to estimate -- when do we think we would be through channel inventory issues?

On the first one, again, there are so many different partners we work with, not only cellular carriers but resellers and OEM's that you cannot make a simple answer to that question. Because the practices vary from one channel to another.

The typical channel situation is that we see rolling forecasts on a monthly basis. And periodically those rolling forecasts turn into new orders for us. We also get from almost all of our channels a monthly sell-through data. And so, we are able to calculate because we know sell in, and they report sell through to us. So we are able to use math to calculate their ending inventory. And because we know their ending inventory and their sell-through rate, we can do the kind of calculation you just asked me about. In other words, we can make own predictions based on the data we have got on when inventories would be down, when reorders would occur and so on.

To the second part of your question -- does that information that we have today give us enough insight to be confident that this channel inventory is going to be worked through in this first quarter? I'd like to be able to answer that. But the challenge is to try and answer that. We're effectively going to give guidance for another quarter. And we don't have the confidence level in the data for Q2 to give guidance for Q2 yet. And by the way, that's completely normal in our business. In the last 10 quarters, 9 of the 10 quarters, we have provided a single quarter's guidance only. And that's no different than the situation we are in today, as far as the guidance window.

I certainly understand your reluctance to give Q2 guidance. But I guess I'm not really looking for guidance on sort of absolute numbers, as much as an expectation of trend and whether your current view would be to say something to the effect -- this is something that we think is relatively localized in Q1. And we expect to be coming out of it to some degree in Q2. Or this is something that is more likely to work itself through in Q2 some time. And it's more likely something later in Q2 or Q3 that is going to be -- we're going to start to see a sort of come back out of it. So I'm trying to get a feel for a kind of timeline here.

Right. Well, generally, given that we have seen end customer -- the reported sell through go up every quarter in 2004. And our channels report a record sell through to us in Q4, we would expect that the channel inventory issue will not persist for very long.

Okay. I have beaten you on that one, as much as I can. Thanks.

Well, I gave you the directional answer you asked for.

Yes. I understand. Thank you.

Operator, at this point, we are fairly long in the tooth (ph) of the call. I wonder if we can take one more question.

Jeff Kvaal, Lehman Brothers.

Thanks very much. I have two questions for you folks. One, Dave, for you, and one, David, for you -- I think. First is -- cash burned in the first-quarter -- do you folks have an estimate for what that may come to? And then secondly on the market share, do you have some way of gauging what your -- the new reality of market share may be for you in the Verizon channel? Thanks.

With respect to the cash position going through Q1, Jeff. We've guided that we will be negative from a cash-flow perspective. At this point, I'm not willing to predict things like working capital requirements and things like that to give us that specific number.

I will just add to that -- the main driver -- we would expect the main driver and cash flow in Q1 to be operating, you know, cash consumed by operating losses. And we have guided on that.

Turning to the second half of your question -- what's our view on -- or visibility or metrics for market share changes now that we have a specific competitor in the Verizon channel? Boy, they just announced that they were commencing sales from the Verizon channel out to the market on January the 17th. And it is way too early to know what the market share is going to look like. Remember the competition in these kind of products is not only a battle for shelf space in the channel. It is even more importantly a battle for sell through to the end customer. And we have a year and a half of selling EV-DO products through to end customers, particularly enterprises. Our competition is new to the game in this particular products base and flavor. And they are going to have to work through a curve on that. How that is all going to play out, we don't know yet.

Well, to everyone else on the call, I hope we are not leaving too many callers with questions in the queue. But the call has gone on over time here. If you have any further questions, you can reach management by calling David McLennan, the Chief Financial Officer, of the Company at 604-231-1161. And thank you very much.

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