Brian Shore
Analyst · Needham & Company
Okay, thanks, Matt. This is Brian again. And a transcript of Matt -- Mike's -- Matt's remarks are going to be posted on our website, in case you want to check it out.
So I have a few things I'd like to cover this morning, before we go to questions. First of all, let's talk about the numbers for Q3.
I think the bottom line doesn't require much discussion. It's really the top line because the bottom line is driven by the top line and, in this case, with no -- nothing unusual about the bottom line, except the revenues, which are way off.
So let's talk about the revenues in the third quarter. Let's back up a little bit to the Second Quarter Call because, based upon the comments we made in the Second Quarter Call, the revenues really should not be a surprise, and let's go through that.
Okay, so remember, when we discussed that in the first quarter and also including -- the first quarter, including as well as the first month of the second quarter, which is June, the revenues were quite strong, but we feel that was really the inventory build that was artificial, based mostly in Asia. And we commented that, in July, the revenues fell off further and then, in August, even further than that, but also we explained that, in June, the June revenues were at the level of the first quarter. So we knew what the June revenues were. We knew what the total was for the second quarter. We knew that July was the middle month. August was a down month, so it would have been pretty easy to figure out within a small range what the August revenues were.
Then we said in the Second Quarter Call that the first 4 weeks of September, at least we have 4 weeks in our books, were tracking August. We also said that, we don't know. We always could just caveat about the electronics industry, probably in the last 20 years and probably 100 times now, how unpredictable it is and how it can turn on a dime and how a lot of smart people get fooled by the changes in the patterns in the electronics industry. But we were talking about going through that inventory correction, and we're at that low level in August and September. We also said that we don't know, but this is the unpredictability part about it, the caveat part about it. We said that we, of course, ask and ask and ask what people in the industry think in terms of customers and OEMs, and they were talking about maybe a recovery by the end of the calendar year. The recovery meant that the inventory would have been normalized. We'd be back to normal levels. Now at that time, I think we said, which I still believe to be correct, that normal is somewhere between the levels of the first quarter and the levels of the third quarter. The first quarter was artificially inflated. The third quarter is artificially depressed, because of the inventory workoff or work-down.
So we pretty much knew what the revenues were in September, and we also indicated we didn't see any recovery until the end of the year. At least, that's based among what people were telling us, with the caveat that we could have been wrong. So I think it would've been pretty straightforward to get a pretty good feel for the revenue levels in Q3, based upon our Q2 call. And yet, the bottom line is really driven very much -- in Q3, the bottom line is really driven very much by the revenues, which were quite a bit off.
So let's take -- so I'm just going through my notes. I just want to make sure that I don't miss anything important. I want to -- oh, here is something else you should know, that it's often something you ask. If you look at August, September, October, November, it's uncanny how flat it is. It's very -- there's very, very flat month-over-month-over-month. Those 4 months are very similar. I've never seen anything like that. It's uncanny. So if you take August and you take September, you extrapolate October and November, you have the third quarter.
Now remember I said that people were telling us -- in the second quarter, I said that people were telling -- during our Second Quarter Call, rather, I said that people were telling us that maybe there will be a recovery by the end of the year. Maybe the inventory will be worked out. And we don't know that yet. What I can tell you is the facts. And I always am very careful to explain. We don't forecast. We just give you the facts that we know. So often, we tell you what we have in the books for the current quarter.
So as far as the fourth quarter is concerned, we got to look at the month of December, which unfortunately is the difficult month to draw conclusions from, but what I can tell you is the last week of November and the first 2 weeks of December were up, as compared to that pattern for the last -- the prior 4 months, which is so flat. Those 3 weeks, last week of November, first 2 weeks of December, which is a 5-week month, were up. The last 3 weeks of December go down. Now why is that?
A lot of people would say, "Well, we had the holiday weeks." Christmas and new year is in the pattern, would be consistent with that, that during those weeks, things fall off. So it's hard to read what's going on, I think because we're trying to figure out what's going on in the context of these holiday weeks. What I can tell you is that last week of November, first 2 weeks of December were up; then the last 3 weeks, weeks 3, 4 and 5 of December, our fiscal weeks of December, were down, but that I don't think it's surprising. And it could be explained by the holiday situation. We just don't know that.
So at this point, we have very little visibility because the confusion of the holiday situation. And we have no books so far in the 4 weeks -- sorry, the 5 weeks of December, our 5 fiscal weeks of December. So as usual, we're not making any predictions or forecasting, but we thought you'd be interested in that information, which is the most current information we have in terms of our revenues. As I indicated in the last few quarters, we don't really talk about bookings anymore because we have such lumpy books within aerospace that they could be misleading, so not helpful often to talk about bookings. And we don't want to give information, even though it would be factually correct, which could cause us to reach the wrong conclusions.
All right, let me just see. Okay, so I guess, that covers that. And then there's a couple other things I want to go over with you.
And last quarter, one of our analysts was asking about electronics and how our market broke down. And it's a question that's come up from time to time, so we decided to provide some information. This is our market. This is not the electronics market. This is our revenue, how did -- our revenues in electronics, nothing to do with aerospace; how they break down.
The biggest segment for us is, let's call it, service providers: Internet, telecom service providers, sometimes called infrastructure. That would be somewhere around 50% to 60%. That's the biggest segment for us. Now normally, people include base stations in that segment, but we separate it because, for us, base stations is RF only, and that might be 10% to 15%.
Enterprise. That's something that's talked about quite a bit, but for us, it's a small portion of our revenues. It's only 10% to 15%. Semiconductor, small, maybe 5% or more. Aerospace and defense, that's 10% to 15%. That's going to be mostly in the U.S. And there's a small portion of other things like medical, instrumentation, industrial, which is we all lump in one category, not significant. But we thought it would be helpful for you to understand those dynamics, because it's a question that comes up from time to time.
The big driver for us is service providers, Internet service providers, and that's where you're going to see the highest-end materials, the highest-tech product. And we're talking often about very large backplanes, which go into service providers. We used to call it Internet service providers, but now it's just service providers, I guess. This is also sometimes referred to as infrastructure, talking about core hub routers, high-speed switches and base stations. But we separate base stations because, for us, it's an RF story.
Enterprise, servers and networking equipment, which is a smaller portion for us and although a lot of high-end product, not quite as high end as the service provider market, where we're talking about probably how [ph] sensitive is Q-sensitive back panels and backplanes, for instance; "highest data transfer rate" servers and switches. This is the highest-end product with the highest-performance requirements.
So sometimes, people ask about comparables, and it's confusing because they talk about a company that's really an RF company. And RF is actually an area where things have grown for us a little bit, but unfortunately it's a small piece of our pie, as I said, maybe 10% to 15%. So it's not going to drive our top line as much as the Internet service provider or the infrastructure market will. So when -- there isn't really a good comparable for Park in electronics, not -- no public one, anyway. And even the private companies are not good comparables to use. Park is very unusual in terms of its market focus. If you look at the -- our competitors, they're going to have -- they have large exposure to markets that we just really don't spend much time in and then for good reason. That's not an action [ph]. And it is part of our strategy. We try to focus only on the high-end area, where we could distinguish ourselves and protect ourselves a little bit.
Okay, hopefully, that was helpful because we haven't done that before. And I should say, I want to say, that these are estimates. It's important that I explain that, maybe even a little bit guesstimates. And why is that? Because often when we sell to a circuit board company, we don't know every sale where it's going, who the end market is, who the OEM is or what application or program it's on. So this is based upon a lot of work with the OEMs, where we can't tie it to POs. So there are some estimation and guesstimation involved here, but we never less -- as long as you accept the information without understanding, felt that it would be useful information in trying to understand Park and why Park might be different than other companies.
Okay, let me give you a little bit of an update on General Electric, which is a top 5 customer of ours. We commented, in the last couple of quarters, we're on the 747-8 program. And revenues for the 747 programs should be a function of how many 747s Boeing sells, and that's through GE Aviation, I should say. These are engine programs for the 747-8. And right now, it doesn't seem like 747s are a real hot seller but we cross our fingers that maybe Boeing will get a big order for 747s made for cargo. We don't know.
A320neo -- the lead engine for the A320neo, we discussed that, I think, in our last call. That's just ramping up now. That's a very big program. That's a new program, and we're going to be starting to supplying to that program this year. And that is a new program and quite a large program.
Let me share with you a little news that we haven't discussed before. It's actually been in the news recently, a couple things. I don't know if you noticed, but the ARJ21 regional airline -- an airliner made by the Chinese aviation company called Comac, recently received its type certification in China. And the airplane uses a CF34-10A engine, I think, yes, that's right, which is a GE engine. We are on that program.
There's other news that you might be interested in, which is that GE Aviation is flight-testing what they call their Passport engine, and that's used for -- on Bombardier aircraft, 7000, 8000 global aircraft. And actually, they put it on a test with a 747 test bed. That's being flight-tested now. We are on that program.
There's another airplane that this Comac company is working on and hopefully will certify soon. That's a big deal that the airplane we did -- that we just certified is a regional jet. The next one is a COMAC919, and that's a competitor, I believe, to the single-aisler aircraft made by Boeing and Airbus. That's a LEAP engine that will go on that program, similar to the LEAP engine that's used on the A320neo. We are on that program.
There are other programs which we're not up at -- we don't think it would be appropriate to discuss right now, but since those are 2 or 3 programs that were in the news, just in the last week or 2. You might look it up or Google them. I thought you'd be interested to know we are on those 2 programs: the -- sorry, the Passport program; as well as the ARJ, what they call the -- this is the Chinese regional jet ARJ21 with the CF34-10A GE engines.
Last item I want to cover is 2 new products are expected to be commercialized this month: 1 in electronics, 1 in aerospace. And these are significant products, both of them, both, I would say, quite high-end, new-technology-type products. So watch out for announcements on our new products. Like I said, they're scheduled to be commercialized, that means announced, this month. And hopefully, we'll make that target. I'm not promising, but that's our schedule now.
I think I've mentioned this to you before, but Park has a little bit of a different philosophy about commercializing products than maybe some other companies. We hold back products much more than others probably would. When we commercialize a product, we feel very confident in the product, as we tested and tested and tested it and it's ready to go. That means, the day it's commercialized, our customer can call our customer service and order the product, and we'll say, "Okay, fine, we'll have it for you next week." It's not like, "Well, we don't really have that product yet." And that's what commercializing means to us.
Okay, so those are some updates, and that's all I have for now. Operator, can we go to some questions, please?