Overstock.com Inc (OSTK) 2006 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the second quarter Overstock.com earnings conference call. My name is Jennifer, and I will be your coordinator for today. [OPERATOR INSTRUCTIONS] I would now like to turn the call over to David Chidester, Chief Financial Officer. Please proceed, sir.

Thank you. Good morning and welcome to Overstock.com's second quarter 2006 conference call. Participating with me on the call today is Dr. Patrick Byrne, Chairman and CEO of Overstock.com; and Jason Lindsey, President and Chief Operating Officer.

Before I turn to my financial summary, please keep in mind that the following discussion and the responses to your questions reflect management's views as of today, July 28, 2006, only. As you listen to the call, I encourage you to have our press release in front of you, since our financial results, detailed commentary and the President's letter to shareholders are included and will correspond to much of the discussion that follows.

As we share information today to help you better understand our business, it is important to keep in mind that we will make statements in the course of this conference call that state our intentions, hopes, beliefs, expectations or predictions of the future. These constitute forward-looking statements for the purpose of the Safe Harbor provisions under the Private Securities Litigation Reform within the meaning of Section 27A of the Securities Act of 1933 and Section 21E of the Securities Exchange Act of 1934.

These forward-looking statements involve certain risks and uncertainties that could cause Overstock.com's actual results to differ materially from those projected in these forward-looking statements. Overstock.com disclaims any intention or obligation to revise any forward-looking statements. Additional information concerning important factors that could cause actual results to differ materially from those in the forward-looking statements is contained from time to time in documents that the Company files with the SEC, including, but not limited to, its most recent reports on Forms 10-K, 10-Q, 8-K and S-1.

With that, I'd like to discuss our second quarter results. Please note that all comparisons will be against our results from the same quarter last year unless otherwise noted. Total revenue was 160 million, an increase of just over 6%. Gross margins were 14.4%, down 70 basis points from last year, but up 40 basis point over Q1. Gross profit dollars remain flat at 23 million.

Sales and marketing costs were 7.6% of sales, down 200 basis points from 9.6 last year, and this is a reflection of a 16% decrease in overall marketing expense year-over-year. Technology and G&A expenses combined were 17% of sales in the quarter compared with 9% last year. And while we anticipate both technology and G&A expenses to flatten going forward, we'll continue to see difficult year-over-year comparisons for the next two quarters.

Total operating expenses were flat quarter-over-quarter at 40 million, which represents 41% growth year-over-year. This translated into a $16.7 million operating loss for the quarter or 10.4% of sales. Our net loss was 15.8 million, within our expected range of 14 million to 18 million. This equated to 9.9% of revenue and a $0.78 loss per share. The net loss was less than our operating loss due to a non-operating gain of 1.9 million we received from liquidating our 50 million of foreign currency bonds during the quarter.

In addition to selling the bonds, the Company raised 25 million by selling just over 1 million shares of common stocks in early May. Our cash balance at the end of the quarter was 46 million with an additional 30 million of availability on our inventory line. Operating cash flows during the quarter were an outflow of 6 million and free cash flows were an outflow of 12 million. For the trailing 12 months, cash flows from operations were an outflow of 40 million and free cash flow was an outflow of 72 million.

With that, I will turn the call over to Patrick.

Thank you very much, David. Financial results are right down the middle of where we thought they would be. Again, I'm sorry about this time last year is when we were coming off -- well, started coming off the rails. And it's going to take us into the third quarter to get everything back on.

The projects we've been working on to improve the business are progressing nicely. I'll be getting into that. But basically, we had 20 projects we expected to finish by the end of June. We finished 13 of them and we have another six that we should wrap-up -- actually, we'll check off today. And the infrastructure we have in place is almost ready for us to go back and hit the go-fast switch and start accelerating again. I think we're really sort of four to eight weeks away from saying, okay, we got through that. We got things back on the rail and it's time to hit the go-fast switch.

I am going to go to the -- you should have a webcast screen in front of you -- you control your own slides, and I'll just call out the page that I'm on. We've made this really simple this time, so there is no lag and so forth. Okay, page 2. There is a safe harbor statement, basically identical to what Dave Chidester just said.

Let's go to page 3. Management changes, some were described in the letter. The old-timers here will remember Jason Lindsey, and I think welcome Jason Lindsey. Jason has actually been -- I think he was President for about two months and then had to retire for some personal issues. But since then he has actually been involved in the business and stayed on at a one day a week capacity. He's come back as President and is taking particular responsibility or specific responsibility for the buying team. I've taken over the internal marketing personalization, email -- well we call it internal marketing as opposed to online marketing and offline marketing. So the website and everything related to customer analytics.

Page 3 -- page 4, I'm sorry. Operations update, I'll be talking about order flow management, customer care and inventory management. And I think just the bottom line is what you're going to see is a business in - I've made no bones about we came off the rail and we had a lot of fixing to do -- but everything has gotten stitched back together pretty nicely. We're getting out of rehab and ready to start walking, let's say, knee rehab, not other kinds of rehab -- knee rehab. And we had a knee scoped or something, we're ready to start walking again, if not running. You'll see what I mean.

On page 5, and Jason can talk to this more, we actually call this customer aggravation points internally, order flow management. And I've described this before in a couple of settings, but it's basically there's about, I think, 17 things that have to be done from the time you push enter my order, submit an order to the point that something shows up at your home. And 17 things have to go right. But if something -- some things may not go right, and it doesn't -- for example, if you don't get an order confirmation email, well, that's not good, but it's not as aggravating as not getting your package or not having tracking information that is aggravating, but not as much as not getting your package.

So we rate those with different numbers of points. And then, if something is, say, a 5 point problem and it's 20 days old, then that's 100 points of aggravation. And we started this when we had our big problems last fall and it was literally in the millions. It's come down to 20,000. In fact, it should be under 10,000. What this means is the problem is 99.5% cleaned up. In fact, the Japanese have this metaphor, if you -- when you're fixing a manufacturing problem you drain the lake until some rocks emerge, and then you stop and you blast the rocks, and then you drain the lake some more until some more rocks emerge.

What this has let us do is -- that's the approach we've taken to this. And we have just drained the lake so far, it is so -- I mean this really comes down to a few dozen orders. It could be some emails that didn't get sent or something, but it means that things have gotten so drum tight -- and for comparison, if we go back and reconstruct where we were when we thought things were going great, and we were going 100%, this was probably around 100,000 points. In another words, if you're sending out 1 million orders a month, or 1 million packages a month, then it's -- if you do something wrong to 1,000 customers, that can easily turn into a few hundred thousand aggravation points.

So we have this -- our process is much tighter than it was even before we made these system changeovers. And in fact, what happens is we keep on getting it down close to, under 10,000 and in the process discover, well, here is one other little crack that it's possible for something to go wrong, and then we go and fix that. But the things that we -- and our data is so granular now, the things that we can turn up though are getting to be tinier and tinier problems. And this has been Jason's effort along with taking over merchandising.

Slide 6, I talked about our NPS score last time. That stands for Net Promoter Score, and there is a book out by Fred Reichheld, The Ultimate Question. And I notice since I've mentioned this, I'm starting to see a lot of people talk about it. Reichheld is the guy -- he is a banking consultant who wrote The Loyalty Effect, which sort of set in motion the CRM movement.

And this is a relatively new book by him. It says that the ultimate question you want to ask about any company is would you recommend this company to a friend and have people score you 1 to 10 -- consider 9 to 10 people -- 9, 10 promoters of the company, 1 through 6 are people who're going to detract, slag your reputation, 7s and 8s are neutrals. And if you subtract the 9 and 10s from -- if you subtract the 1s through 6s from the 9s, 10s, you end up with your Net Promoter Score. What percentage of the people that deal with you are going to go out and say, hey, I had a great experience.

And ultimately, he makes an argument that this is the most basic fundamental question you can ask about how well a company is doing for its people. I go into this detail because I mentioned this last time and people have asked about it. For reference, he says in that book that the average Net Promoter Score at an American company is 8 to 10%. And that the superstars -- he has a list of about 20 superstars in his book who score sort of 40 to 70, I think -- maybe mid 70s.

So now go to slide 7. I put a scale on this, this time so you can see. We were up in the 70s. And we fell -- if you look at what happened, we did our implementation last August, November things really collapsed, and it says that we fell to only 38, which is still a very high NPS. But I think that that was a lot of goodwill from our customers. That may have been people, because if somebody has ordered from you 10 times over a few years, and then you do something -- you screw up once -- they're going to -- so in truth, I think our real NPS score from new customers was probably quite a bit lower, could have been zero or lower.

It has rebounded, as you can see. And we have a pretty complex situation. The only other people I remember from the Internet in that superstar list were Amazon and, I think, Barnes & Noble as e-tailers. And I'd point out it's a little bit harder to do what we do than it is to send books. In other words, if you get a book to somebody in a timely way they tend to be happy. And we see that in our own customer satisfaction when people are buying books and movies and videos and stuff, it's easier to satisfy them than if you send them a lamp. I mean the lamp can be broken and things.

So this -- we set as our goal to get over 60. We are over 60, which is according to Fred Reichheld's book a fantastic score. I think that there may be ways to get this over 70. And one of the great things about our -- I know I go on and on about this data warehouse, but we have such granular detail so we can see, well, what customers are scoring us lower and what do they have in common. Do they have in common that they ordered lamps or a certain type of lamp? Or ordered from certain partners who didn't ship properly? Or is it because -- we're breaking it down to the SKU level. And satisfaction and dissatisfaction by SKU. And then it's been fantastic. We've discovered, okay, there's -- here we're shipping out some furniture and that had some disproportionate dissatisfaction. And we discover well it's really the packaging. So we went to the -- we have a warehouse in Indianapolis and we changed the whole way we package it and returns in that dropped 50%.

So it's the granularity of data that we have is going to keep us busy for the next -- it means that we can drain the lake another 10 feet and find a lot more rocks to fix. So moving on that may be more detail than you wanted - I get excited about this stuff.

Go to page eight, brand recognition. This is among all consumers. We've gone; we tend to dip as you can see from the Christmas to the first quarter and that's happened again. But we've built a nice brand. It's not obviously as established as Amazon yet among all shoppers.

If you go to slide nine, you see among Web shoppers it's really quite outstanding. We've gone from 26% to 65% now. So we've built a good brand, and again that you do see that drop off from December to April, but that's normal looking back through the years.

Okay. Slide 10, GAAP inventory terms. We were just generally making improvements until we got into our troubles. But we're climbing out of it. Again, this is, to me, this picture represents a lot of things. We had a good business going, we stumbled, but we're picking ourselves up, dusting ourselves off and getting back up to speed.

I think, incidentally, I know a few years ago I opened a letter saying that the rhythm of the Dao is like the drawing of a bow. We have -- for example we in our margins, our gross margins, we built up the capacity to get better and then suddenly over the space of a year or 18 months they surged 5 points. I think that the same thing is happening with our gross margins and with our inventory turns. I think that given our new granular data, we're going to see -- starting within a couple quarters a dramatic increase in inventory turns and margin.

Marketing efficiency, we spent 19% less and have 8% more sales year-to-date. Here is something we call nectar, I think contribution margin is the official term although people use contribution margin differently. But basically gross profit minus marketing expense. And I'd like to see this get over 10 -- when this gets over 10, we will have a nice real business. And I do think that's possible.

IT Project update. There's really three levels. I won't go into as detailed an explanation as I've done before. But just think of our website as a stack of three systems. There is the whole shopping database, which is Oracle based, then there is the ERP system, which handles the orders and sends them to the warehouses. And then there's the electronic -- Enterprise Data Warehouse.

The shopping database, we upgraded from the Oracle9i to 10g. That's a big deal; that was the last of our big IT projects. Oracle 10g, for those who are tech investors, Oracle 10g at the risk of being cynical, well a cynic will say it delivers what 9i promised. But be that as it may, it's the real thing, it's great. It's so much more robust, faster recovery times, a cost based optimizer, which means that queries get run appropriately. It just, it's so much -- it's a great system.

ERP, our capacity was 10,000 orders an hour under the new system we installed last August. And that was problematic because we got, I think we capped at 8,000 orders in an hour, sometime in the last Q4. But that was again a factor of the hurried installation or the rush to installation.

We had an outside consulting firm come in from a couple hundred thousand dollars, look at this; there is about 5,000 settings in the database and they spent six months studying it and then they gave us this is how you should set these 5,000 setting. We did it and we have a -- we now have theoretically 150,000 order an hour capacity - I mean we tested to that. We stress tested to that. So that means and even if you say 100 to 150,000 about 15 times the demand that we went through in the last year's fourth quarter. So those two -- so that layer is set for a long time.

And then the Terra Data data warehouse, we have super sized it with a new system. I think I'll leave it to Terra Data - they're going to be announcing some time soon what they've done here. But they have super sized the system and that's important for a bunch of reasons like we're now personalizing e-mails, but we had trouble personalizing more than about 500,000 e-mails -- we send nearly 10 million. We have 22 million addresses, but we send 10 million about three times a week. And we could personalize about 500,000. We're actually through some brands in the IT group, we, I think, figured out a way to get to about 2 million. But with the super size system we'll be able to handle a lot more and we're very glad we upgraded.

So put together, and that new system actually is in-house. It's in, we're [sailing] over I think - we're running in parallel then switching over to the new system in a couple of weeks. So put together we have now finished our transition from a company that was running on a bunch of homegrown systems that we wired together that worked and carried us for a long time, but we're just keeping problematic -- keeping alive was getting increasingly problematic to now three industrial, all three tiers are big industrial strength rock solid systems. This is the first year we've actually been ready from a capacity standpoint. We've been ready for Christmas and we're ready in July.

Slide 14. Propeller, there's a slight altering, of course. I do expect that planted on August 10, you should -- the night of August 10th, we're going to roll out Propeller. Now these things can slide. I've taken over these issues and I would be happy to answer questions about them.

And by the way Holly decided, Holly was running this for six months, but she decided for her own personal reasons. It's been a long five years for her. And she decided to step down. So - but I've been looking forward to taking this over.

Other projects, the warehouse mezzanine is now finished. It cost us -- let's say it cost us X to pick a package and say that one picker can do P picks in an hour under our -- in our normal warehouse. In the new mezzanine they can do about 140% of P and the effect on that, on X, the cost, is going to be to eliminate about, because picking is just one aspect of X, one component of X.

It should be enough to knock about a third or maybe 30% out of X. And so that translates into a -- I'm so excited about the mezzanine because it's not going to give us a full percent margin, but maybe half a percent margin. And we're already running out of it very nicely. We're getting about 41% of our picks out of a very condensed high volume picks base.

We have revamped the returns process. As I mentioned just a few minutes ago, we actually do a lot of statistical analysis of returns, look for any outliers, and then go and fix those outliers. If it's a partner product or if it's out of our warehouse, we figure out why it has too many returns and we dig into that issue and we fix it for that SKU and for – if it's something like a lamp that has -- that's getting broken in shipment, we not only say, let's fix the packaging around that lamp, let's review how we fix all of our lamp packaging. And if we have a partner who have too many returns, the same kind of complaint, we dig in and we actually take them off the site until we're satisfied that they have fixed that problem. So it's really worked out well in reducing, I think, our returns cost will be coming down significantly too.

In outlook, I reiterate what I've said before. This is really the first three quarters at the most of this year. Third quarter is going to look pretty much like the first two. And then we do anticipate hitting the go fast switch, meaning getting back up to speed in the fourth quarter. We may even hit the go fast switch a couple weeks before the end of this quarter, but it is not enough that it will make a difference to the financial outlook for this quarter.

Share ownership -- I'm going to talk about our new DRS system. I hope you notice, I'm less visible in the crusade than I have been in the past. But it's not the case that all shareholders want me to be less visible. And I feel torn. I have a conflicting set of duties. I'd love never to worry about Wall Street again. But I hear from shareholders who buy tens of thousands and hundreds of thousands of shares and who never get them and who tell me I'm just going to stop buying, because I don't know what I am paying for. It's not like it's my job to see that they get what they pay for, I think, but I think that's the government's job, but I think I have a fiduciary duty to do something.

So let me explain what we're doing here. If you can see - well, if you print these out, you will be able to see them better. But what this is is this is a simple bird's eye view of how share ownership works. And there is Overstock on the left. And when we issued shares into the market they went into a transfer agent. And I know for most of the people on the phone if you're Wall Street you understand this already, but bear with me, there may be some shareholders who don't.

So we issue our shares and do the transfer agent, that's Computershare. They're injecting them into the system, they go into the DTCC, the banks, the broker dealers are arranged in hub and spoke system around the DTCC. They all have accounts at this Depository Trust & Clearing Corporation. And as their clients buy and sell, they just really get moved around among accounts within the DTCC.

If you own stock through an account at, say, Morgan Stanley, let's call that the bank at the 12 o'clock position in the circle there. You're a beneficial shareholder. It's owned for your benefit but the -- it's held in street name and that's the street name of the bank that you have your account at. And so Morgan Stanley actually has those shares at its account within the DTCC.

At the -- in the upper, in the cloud on the left is book name, it's booked in your name. The cloud on the right is street name. And then the cloud on the left is called registered shareholders and on the right they're beneficial shareholders. And then if you don't object to us knowing who you are, you are a non-objecting beneficial owner which is called a NOBO and if you object you're an OBO. So we get NOBO/OBO lists and we can see who some owners are and such.

Go to slide 17. Share Certification. And this is not a recommendation, it's just an explanation. I run into shareholders all the time who talk about, gee, I tried to get my shares and the certificates and take them away from those shorts. And I tell them I've never really told people to get their shares in certificates. It's and I'm not trying to stick it to the shorts, shorting is legal and it's fine.

If -- I am trying to stick it to people who are counterfeiting, and that's a problem. But people tell me that they can't get their certificates. If you try to get it -- if you're a beneficial shareholder and that's the cloud in the upper right, and you try to get your certificate, what you're saying is you're saying to your bank, "I want my certificate." Your bank, the guy there at the 12 o'clock position is sending the message to the DTCC, let's say you have 1,000 shares.

The DTCC says they have maybe 1 million shares; that bank has 1 million share in its accounts. They take out 1,000, send that message back to Computershare electronically, and then Computershare prints up those 1,000 shares in certificate and delivers them to you and that's your -- you become a registered shareholder then.

Now, I hear over and over, this is essentially impossible for bigger owners - it's kind of funny, smaller retail owners are having more success than big owners. Now, I'm not suggesting anybody does this, but if you want to do it I think you should be able to do it. And my shareholders are not able to do it. And I think I am supposed to have some duty to them.

So if you go to slide 18, you'll see the solution. Our board has passed -- approved our involvement with the Direct Registration System. How this works is something like -- is a lot like the paper share. You tell your bank, I want my shares DRS'ed, that message is supposed to be sent electronically to Computershare, which then registers your ownership electronically.

And everything -- so you're issued in a sense some e-shares, but it's taken out of that whole system and it's taken out of all the slop that is going on within the settlement system. It does mean that you can't loan them. And I know people are getting 40 or 50% interest to loan their shares and more power to you.

I hear people -- I can't believe this is true, but I hear people are paying, or I saw something that said people are paying 54% to borrow shares, but I hear that to borrow real shares, people are paying $0.06 a day which is about 120% interest. I don't know if that is true. But you -- more power to you if you want to loan your shares. But if you do this you won't be earning your 50% interest anymore. You will on the other hand be assured of your ownership.

Now, interesting thing about this is because it's all electronic it should be able to be done in 24 to 48 hours. There are no excuses. When you try to get paper, I know you -- I've heard 100 different excuses from shareholders saying that my broker told me I can't -- it's in the DTCC type-1 or type-2 or it's in this or it's been lost in the mail or the sun is rising in the East or anything. But they just don't get their paper.

DRS takes all the slop out of that. And they should be able to -- in 24 to 48 hours, you should be able to get your shares registered, if they bought them. In fact if you go to slide 9, I'm sorry, 19, there is a statement from the DTCC website that a quick and cost-efficient transfer, if you're interested, just contact your broker.

Now, I suspect that what's going to happen is the same thing that is happening to people with paper. And that is, you're going to call and say that you want to DRS your shares and your broker is going to make -- well, some people's brokers who have not been honest, there is most brokers I know are honest. But there are some who apparently aren't honest.

And they're -- if when they try to make the same excuses that they make with paper, you'll know in this case, it's a lie because they should be able to DRS your shares in 24 to 48 hours. If they can't, what that means is that when you sent -- suppose you bought your stock at -- keep the math simple -- 38 and now it's 18. Let's say that you bought 1,000 shares at 38 and you sent them $38,000.

And now it's at 18. If you say DRS my shares, and they come back with all the excuses that is what they come back with the paper, what that really means is when you sent your $38,000, they took it, they bought themselves a nice dinner, they bought their wife a present or something, but they didn't execute the trade because if they executed the trade, they can DRS your shares in a day.

And what I suspect some people are going to do is then say, "Wait a second, I thought I just lost $20,000 because it went from 38 to 18." But if I sent you $38,000, and you never executed the trade, just give me my $38,000 back. I'm not advising people to do that, of course, but -- my system has hung up a bit.

If you go to slide 20. I am advising that if you -- I'm not even advising you to DRS your shares. But if you DRS your shares, try to, and your brokers come back with some hooey about -- the same hooey they give about people trying to get paper, or even -- then what that means is they took your money and they never executed the trade.

And if you're having any trouble, I suggest you send an email to this guy Wes Christian who is standing by to help anybody who has that problem. I'm not telling you how to deal with it. But if so, I've heard from, I don't know how many countless dozens of shareholders who say "I've bought and bought and nothing settled and nothing settled and I don't know."

Well, if you paid your broker money, you should be able to call him and say, "I want my shares DRS'ed tomorrow." And if he can't do it next week it means somebody is fibbing. Okay, well, so I hope that was -- I know that won't be of interest to most people but I know that there are a lot of shareholders out there who have had problems and I hope that clears it up for you.

Slide 21. Finished. Let us go -- I believe Jason is on the line, let us go to questions. Operator? Jennifer?

[OPERATOR INSTRUCTIONS] Your first question is from Aaron Kessler with Piper Jaffray. Please proceed.

Hi, Patrick. A couple of questions for you. First when you do hit that kind of go-switch later this year to reaccelerate revenue growth, how do you maintain the lower sales and marketing costs, or is that possible? Or would you expect an uptick in that cost?

Also, once we start to see some of the technology costs leveling off they continue to increase here, is that more of an '07 or '08 event? And also once we start to see an improvement in the conversion rates, would you expect to start to see that in Q3 with the roll out of Propeller? Thanks.

Thank you, Aaron. Stand by just a second. Funny you said Tech, Dave, why don't you do Tech first David. Because you gave some great data yesterday. Why don't you talk about when that levels off?

Yes. I think, Aaron, that is a good question. We did a lot of investment in 2005 and we're just really starting to see it flat and then I think, for the rest of 2006, you may see it tick up another 10% from where it is now. And then in 2007, we see it really starting to level off. Depreciation expense is going to start leveling off next year. We only anticipate it to maybe be a couple million more than it was this year. So really, 2007 is when you're really going to start seeing the benefits of that.

Yes. It's -- we -- really we think the second half of this year has a very shallow growth and then in next year you, I mean our goal is to keep things flat or very close to flat. And in fact, I think there may be conceivably a reduction. And then in '08, you start seeing $35 million of depreciation start rolling out. And it's -- start coming off our books.

Really starting in the second quarter, and by the fourth quarter of '08 it will all have dropped off. But I wasn't very surprised yesterday in reviewing it with David, how close to flat we are going from -- getting from here forward. On the other issues, Sales and Marketing, no, you he shouldn't see that tick up as a percentage. When I say hit the go fast switch it's much about internal technology and conversion and analytics and is driven that way and not by an increase in the marketing number as a percentage of revenue.

So and then, third question -- conversion. I really think that that -- the fourth quarter is when you'll see an affect. We are going to turn it on August 10, but we anticipate six weeks or so really -- we anticipate years of tuning it in and perfecting it. But I wouldn't expect to see any real improvement this quarter. I'm just not counting on it but I do think that in the fourth quarter we will have a lot of improvement from the fourth quarter forward.

Great. Thank you.

Thank you, Aaron.

Your next question is from Scott Devitt with Stifel Nicolaus. Please proceed.

Hi, guys

Hi, Scott.

Pat, well, let me just start with are you willing to give an Op Inc. number for 3Q like you did for the second quarter?

I think about the same. I think about the same.

Okay. And then --

Maybe a little bit better. Maybe a little bit better.

And then you're --

Maybe a little worse. I think it's going to be right in that same range.

Is that Jason?

That's Jason.

And then secondly, your inventory expectations as you head into the fourth quarter. Do you expect to build from current levels or not?

Yes. But not as much as past numbers would have predicted. But Jason, why don't you hit that?

Yes. I mean we're a retailer and we sell stuff that we hold. So we'll definitely forever, probably as long as we keep growing, see inventory peak sometime in the third quarter or early fourth quarter as we anticipate Christmas coming.

However, I think our peak this year will probably be at least 10% or 15% lower than our peak last year even though we anticipate sales to be up nicely year-over-year. So inventory definitely will go up from here. But I don't think it will be any higher than it was last year, and in fact I hope it will be lower than it was last year.

What we have really learned -- Jason why don't you expand as much as you're willing to on what we've learned about inventory. I'm not sure how much you want to give away, but -

Yes. I've really been focusing on three things. First is the management score. So making sure the customer experience is better. And that we don't stumble from the customer's point of view. And I think although stormy and team actually in customer service are still doing a lot, and they're implementing a new customer service system to help track customers' issues. And so you'll see a lot of improvements there and what we're doing in returns will help customers. As far as just delivering them their packages, there's not a whole lot more, I think the company needs to do to tighten their systems to make it better.

So that was kind of the first thing I really was focused on when I came back, and I'm not saying I'm done, but it's mostly done. And so now most of my attention is going toward the financial statements and how can we make the financial statements better. And the best way I think we can do that is by controlling our inventory. And the real thing that sticks out as you look at it is, I guess, the title of what you would call what we're working on is really SKU reduction. We have -- there's a big section in the middle that are very good, very profitable SKUs, and they happen to be the stuff we sell the most of so --

And then there's a whole bunch of other things on the outside that we've done really poorly on over time. And so this new Terra Data system and the data warehouse is allowing us to look at the data on a real, granule level and figure out where our winners are and where our losers are and we've put in a lot of policies and procedures recently in the buying team to make sure that as they place new orders for inventory, going forward, especially now as we're starting to look to buy for Christmas that we're really drawing to the winners of what have been the big winners for us that have had good margins and good velocity. And we're really trying to eliminate and not repurchase items, and SKUs in subcategories where we've stumbled in that have really put some real drag on our margin.

Another thing we've done as a corollary to this -- because we won't be buying in certain categories where we've done poorly but we don't want the selection for the customer to decrease. We spent a lot of time on the partner program and making sure and implementing in the partner program a score card system so that the partners can see how well they're doing and we're judging them based on how well they perform for our customers.

What we provide our customers with the partner program is we need to make sure that our partners deliver our customers the very best possible service and price. And unless we score them and monitor them and watch them over time, we can't help our customers by ensuring our partners are giving our customers the best price and product and service. So we've implemented a score card system, all of our partners just got that for the first time a couple of weeks ago, they'll get it every two weeks, and it will help us make sure --

It shows them I wouldn't call it judging them so much as showing them where they need to -- how well they're satisfying the customers and what they're doing that's creating dissatisfaction.

Yes. I think that's fair. I think it will -- it's really a tool for them to figure out exactly, just like Patrick said, where they can help satisfy the customers better. So we're going to be doing a lot over -- we've done a lot and we'll continue to do a lot over the next couple of quarters to make sure that the inventory we have is the right inventory, not only for the customer but also for Overstock and its financials. So because of all that I think you'll see inventory not be as high as it normally would.

Okay. And finally, let's see, fixed operating costs this year, at least in terms of my model, somewhere in the range of 12.5%, a couple percentage points of that is in non-cash depreciation. And on slide 12, your nectar, which I call contribution profit -- 6 and 0.8% and Patrick said -- you hope to get it toward 10%. Or at least that's when you think things start moving in the right direction.

And I'm a little bit unable to get, one, to I guess that contribution profit expansion in percentage terms. Kind of similarly to the prior question related to how you actually reaccelerate revenue growth without spending more sales and marketing dollars. And then also, in terms of even if you were able to get it to 10, how you can scale the fixed cost structure so that you can actually turn profitable. Because as you're growing at a slower rate it's a lot more difficult to scale the fixed costs. That's all I have. Thanks.

Okay. Well, Jason, do you want to say when do you think that will be over 10?

I haven't spent a lot of time looking at the actual -- that number -- as I have looked at the projections over the next couple of years. I don't know offhand when that hits 10.

Okay. I think it's over 10 in the first quarter. I think it's over 10 in the first quarter. So was that you saying wow, Jason or was that --.

That was me saying wow.

Jason is working, I think that is going to be over 10 in the first quarter. And if it's over 10 in the first quarter --

I'll buy you lunch.

Well, I think that will be over 10% in the first quarter, and if it's over 10 in the first quarter it can do better than that for the year. But I think that we're about to see it. And so and then going to the other part of your question, that we shouldn't -- spending is 7% of revenue on marketing, I think is a good number for now.

We should be able to get growth back not by spending higher amounts than that, but by doing better. I guess the common denominator of everything Jason is saying about the inventory, and what I'm looking at in the customer, is you probably already know the [Prato] Principle, the Economist who found this thing is statistically you find 80% of -- the 20/80 rule people call it, except in our case, as we look into things, it's not all -- the trick has been we didn't have the granularity of data to find the 20%. We knew that 20% of what? Of the products of the inventory was doing 80% of the benefit, but it was just beyond us to find -- and our customers, 20% of the customers doing 80% of the benefit. We now have the systems that can find the 20. And then you can say well let's eliminate all the costs associated with the other 80% that's only giving you 20% of the benefit.

And in fact the truth is, as we get into the data now, we see that it wasn't 20% giving us 80%, it may have been 10% giving us 80%. And we can just - we have I mean it's really opening our eyes about our business to be able to get the resolution we're getting now.

I think what all that means Scott is the way that that gets over 10% in the near term is there has got to be a quick increase in margin expansion so and margin expansion can come definitely from selling more of the 20% that does the 80% and less of the 80% that doesn't do so much gross margin for us.

So if our product mix doesn't really change that much, because remember 80% of the sales is coming from this small 20%. But we are able to eliminate the 80% of the stuff that does not sell that much but has a drag on margin, it could actually increase our margins dramatically even though the product mix itself wouldn't change that much. So I think, Patrick, what you're saying is it can get over 10 in several ways, but one of the big key ways obviously is your margins increase.

Correct.

Thank you.

So Scott, hold me to that. Remind me in the first quarter of my commitment to get that over 10% by the first quarter.

I will write it down right now. Thank you.

And asked me if I enjoyed my steak --

Okay. Next question, operator?

Your next question is from Bill Lennan with Wedbush Morgan.

Hi, Bill.

Hi, everyone, how are you?

Good. How are you?

Doing well. Let's talk about gross margin first. 40 basis points improvement sequentially. So could you give us an idea of how much of that was mix, say, [BMVG] falling out or organic improvement? And then in the prepared remarks, you talk about a drag on gross margin for the rest of the year, Patrick, I wonder if you could kind of put some -- frame that for us and help us understand how to be modeling gross margin?

And then generally, also I want to hear a discussion if you could on Q4. I think you said earlier, you talked about a shallow growth for the second half of the year. And I know you've already addressed Q3 specifically, but I wonder if you can just give us some thoughts on Q4 as to growth levels and where you think probability will come in, at least a ballpark figure? Thanks.

Okay. Gross margins -- Jason, people can back into what the gross margin is in our direct versus partner, can they not?

I think, yes. I guess just generally two things really happen I think in gross margin. You see a 40 basis point improvement but the truth is operationally there was much, much more improvement than that. But at the same time, we're getting rid of some of the inventory from last year that's moving slower and we're doing a lot of tests on some velocity of some of our inventory. And so, yes, net/net, it went up 40 basis points but what you really see is operationally we got a lot better at some stuff and then are actually the cost of some of -- the actual merchandising margin or the cost of the product actually went up during the quarter. So those two things together netted to only a 40 basis point improvement.

Can I tack a footnote on there?

Please.

Okay. I'm just making some notes. I would say, since people can back into it anyway that our partner margins are running on a GAAP basis, Jason, I don't want to give something away that's not in there.

No, it's in there.

So it's what, 18, 19%?

It's just over 17%.

Okay. And on our core -- on a GAAP basis is what, 10.5?

Yes.

So that's why you are seeing in the 14.4. A few additional comments. That 10.5 is really a combination of some stuff that's over partner -- in fact, some stuff that's over 20. And secondly, a bunch of crud that we're clearing up. How we got that crud is -- I shouldn't say crud, but it's financially crud. And that's why the margin is where that is. Last year, when we were still growing 100%, we got too hungry. And I think we lost our discipline buying.

We went out and bought too aggressively and didn't get deals that should have -- that met our normal standards. And we were just trying not to choke off growth, which has happened to us on one or two occasions. There was a Christmas we had that was a great Christmas, and we suddenly hit the skids the next quarter because we had nothing left in the warehouse.

And we probably -- nine months ago going into Christmas, got -- just lost some of our discipline. So there's cleaning that up. But what's different this time is we actually -- we have a great view not just into the buyer level or the store level or the department level, but the category, sub-cat and right down to the SKU level of where the real returns are. And it's just something we could never have done under our old system.

I mean the Terra Data gives us stuff out of push button that would have been weeks and weeks if it could work with Excel sheets on the way we were doing things before. So the idea is really just to sell through that stuff that's the -- and not replace it and just keep the stuff that -- that's over 20%, keep replacing that. I think clearly core margins should be higher than partner margins.

So that's why I think that once we sell through this stuff, you will see a real release. The bowstring will be released, and a real surge in gross margin. However, I think it's going to take us through the rest of this year to work our way out of those inventory positions. As once we work out, I think that we will see a dramatic increase.

As far as the operation costs, there's probably I think another 150 basis points that we can -- maybe 200, but at least 150 basis points that we can now improve on our current set of systems, on our new customer service app is going live Wednesday. Our new warehouse mezzanine -- if I do the math, we should gain 150 basis points in margins from that. But again, that's going to be -- I don't know how much that's going to be offset, but that's going to be offset by moving out of $20 or $30 million of inventory that we can comfortably move out of by the end of the year.

And then -- so that's why I think if you -- you'll probably see in the fourth quarter, I think the margins will stay about where they are this quarter and tick up in the fourth quarter, but then we'll be -- we'll be free. We'll be able to operate both with lower margin - we'll have turned some things into cash -- I mean with a lower inventory, high return and we will have released some cash. And as far as the fourth quarter, I do expect profitability. But in the -- I don't know, Jason, do you want to say?

No. I just -- it really just depends too much on the last of this. We really want to have any slow-moving inventory out and we just don't know yet, I don't think how much - - how easily and how quick that stuff's going to be selling. You've seen -- there is a drag on margins in the first and second quarter from selling this stuff.

And so far it has sold okay. We're getting rid of it. It doesn't look like we have -- obviously, we're more than adequately reserved I think for all of it. But just as far as having sold the last of it, so that there is no more drag on margins, we definitely want to have that happen by the end of the year and getting rid of the very last of it. I just don't know at what price it's going to clear. And so --

But I'm looking for profitability. And if we do get some lift from all the testing and analytics, maybe we'll have some meaningful profitability, which would be -- to me anything over $10 million I would say was great and meaningful and so forth. But I do want to see at least profitability in the millions.

Okay.

Thank you, Bill.

Thanks.

Your next question comes from Frank Gristina with Avondale Partners.

Thank you. Patrick, you mentioned returns in your comments. And I was just curious if we could kind of do the same analysis if returns were X in '06 for the December and March quarter, what are they down in, say, the June quarter, if you've reduced those aggravation points sort of to the 20,000 range in March? Have you seen an impact in returns as a percent of sales? And then put an edge on that, maybe, by expressing how much gross margin impact I guess a large return, was having on your gross margin?

Okay. Hi, Frank. I'll take part of that. The return project, which we call Project Boomerang has been wrapped up and is a project that said let's take items that are having a statistical disproportionate share of returns, take them off the site and fix them if they're coming out of our warehouse or call our partner and say, hey, this product is getting returned too much for this reason and let's address it and we don't put it back up on the site until it gets resolved.

And the discretion is taken away from the buyer and is actually given to this Project Boomerang team, which now sits with our customer care group. And so they're the ones who actually -- it has to be fixed to their satisfaction before it gets put back on the site. That really came live in June.

So I don't know what effect it would have had on the second quarter, but I do know that the projects we've done that with -- the SKUs we've done that with -- we now have a couple of months of experience -- are experiencing a 50% reduction in their returns. So it's a business process, didn't cost us any money. But it's -- now overall -- so first of all, because it came live around June 1st, I don't know how to quantify its effect in the second quarter. Dave Chidester, do you want to chime in on that or how you would quantify this effect going forward?

Yes, well, I think we've been making efforts all year. That project's just gotten implemented. But we have seen overall a slight decrease, probably 50 to 100 basis point decrease in our return rate this year. It's hard to extrapolate exactly what it's going to be going forward from a quarter's data. But we have seen it come down a little bit. And I think clearly, if we take SKUs that are -- have higher return rates and fix them that's going to have a positive effect on our overall return rate.

And it's -- I should have just thought of it that way. I do know that our return rate, our quantities of returns that we have to process, they were down significantly from where it would have been the same -- if we had run at the same rate last year. I did see that data a few weeks ago. And even that is not too meaningful yet because it's really only been two months of the year that this new system has been in place and working.

I would say I don't know if we released what our -- we look at the net cost of returns, which is a function of how much comes back and how much we can repackage and put up on the site again and sell versus how much we have to liquidate, et cetera, et cetera. And that comes out to, let's say, a total cost of C%.

We have about a 6% to 8% return rate, the actual net cost of returns. And you also -- I left out you factor in there, the cost of handling returns and then less how much we can recoup from customers who have destroyed things and such. And it's less than 6% to 8%. It's a single-digit number. I think that you can expect us to reduce that number by about a quarter, about 25%. So that's -- again that goes straight into margin improvement.

Great. And one last question. I noticed that - so your sales growth was 6%-plus, but your direct sales grew 14%. And I imagine this is a headwind on your gross margins until you can get some scale. But is this is a deliberate step or is it just seasonality between partner and direct, because I would assume that going forward, if all these things are implemented, you have much better margin control on the products that you ship out directly. So was this a deliberate focus or was it perhaps part of the very good prices on the inventory that you're flushing out?

Jason, why don't you...

The latter. Yes, you hit it right on, the latter. It wasn't intentional at all. It's just a function of customers are smart, and they find out where the best prices are. And if we think we have too much of something and we lower the price, then sales in that particular item increase. So you're exactly right.

Okay. So going forward, you're still not ready to try to merchandise your website to push the products that you're carrying; it's going to be kind of that same 60-40 mix?

Yes. We are set, so we can do that. And because we run into a situation in Christmas where if we have partners or one of our own warehouses that falls behind, we want to be able to actually ratchet down sales to that partner or our own warehouse. So we do have the systems that let us do that. And - but we do that until somebody catches up, if somebody falls behind in shipping.

And so by doing things like placing it in a less -- a product gets lower position in a subcategory, it will drop sales. We are wired to do that. But that's not what's driven this. I think, what were we, 49% in direct in last quarter? That's just based on organic pricing that Jason controls and not an overt decision.

Okay. Thank you.

I guess, time for one more question.

Certainly, sir. Your final question will be from Shawn Milne with Friedman Billings Ramsey.

Hello, Shawn.

Good morning, Patrick. A couple of questions. You talked a little bit about your tech spending leveling out -- maybe this is for David -- if you were to look out into '07 if you get the fixed nut put together between G&A and tech spending, what do you think for a number in '07? Maybe 110 to 120? Or can you do a little better than that?

And secondly, Patrick, I mean that's a pretty ambitious target. I think you actually just talked about -- I mean if you're going to get to 10% contribution margin, it sounds like you're betting on obviously that it's going to come from the gross margin side. I'm looking at my model Patrick and it looks like your best direct gross margin ever for the company was about 13, 14%. So it sounds like you're -- again, it sounds like there's a lot of -- you're betting on a lot of room for improvement on inventory turns. Thanks.

Okay. On the -- Jason, what were you just saying?

I think he's saying he wants a piece of my bet.

I think that you'll see -- just mark me down for 10% contribution margin in the first quarter.

And your question about tech and G&A range of 110 to 120, I think that's reasonable. I mean we're tracking to be somewhere in -- right in the middle of that this year. And I think our goal, that's a tough goal to hit, but that's our goal, is to be in that same range next year. So 120 is probably a good target for us next year.

Okay.

I mean we talk of the possibility of keeping things flat next year. Except, David, we have to have some increase in depreciation next year, right?

It will be 2 or 3 million more that's what we're projecting now.

2 million or more?

2 to 3 million, yes.

Yes, so there will be that increase, but other than that I think we're going to be able to keep G&A pretty flat and tech developers and the non-depreciation pretty flat. Then again, in '08, a whole bunch just starts washing out. And in fact, we were I think aiming for zero. That may be a little ambitious but I'm -- we're going to see, if we can come up with a plan that gets us to zero growth in G&A and Tech spending outside of depreciation.

Patrick, while I've got you, do you have any thoughts on what's your perception of eBay Express? And are you just keeping a close eye on it, thanks?

My perception is very good. It's -- I think it's a good program. I'm not keeping as close an eye as I should on it but I think it's -- I mean I've tried to be – you know how I feel like anything I say gets exaggerated and twisted these days by certain folks. But I am just trying to give my honest feedback. Like people ask me about Amazon Prime and I said well I didn't see the sales proposition there. I think that eBay Express is a very good program. I also would suggest keeping your eye on and this is out in the public now the whole Google. You know about what's going on there? I think that's - I've read one or two articles don't want to get in trouble by saying too much, but I think that that's going to be a pretty good competitive program.

Okay. Thank you.

Thank you. So nobody believes my 10% contribution margin? We'll see.

Well, you do the math, and our margins have to expand quiet a bit to get there. And I know we think it's possible and I know it can happen having it happen as soon as the first quarter as much as it has to expand to get to 10%. It'd be fantastic if we get there.

We have pulled the bowstring very, very tight and we are releasing the bowstring in about eight weeks. And I think that you'll be there in the first quarter. Are there any other people waiting to ask a question, operator?

No further questions at this time, sir.

Great. I'll wrap up. It just occurred to me to comment on something I hate commenting on this. And I'm going to comment on it I've just enough to say that I think in the future we're going to have -- there is some rumor going around on the message boards about a -- and the people are contacting me about a buyout and somebody has posted a buyout offer on the net. I'm mentioning this only to say I believe this is a hoax.

I have strong doubts about this. I, of course, have to pay attent -- and not treat anything as a hoax, but nobody should contact me asking me about this anymore. And I believe it's a hoax. And in the future, I'm now announcing, I'm not going to comment on anything like that that comes out, any rumors or things like that. Jason, did you have anything else you want to say about that or anything?

Not about that, no. It feels good to be back. I think we're both back, both really engaged in the business. You're working on internal marketing and trying to get conversion rates up, and I'm working on inventory and making sure we have the right stuff and really trying to drive our margins up. And hopefully we're both successful and you'll see some real results in the coming quarters.

Yes, I think we're both -- it feels I think that both of us like a bit of a step back in time because in the early days of this company Jason and I did just about every job in the place. Separately, but we worked and we switched executive hats all the time and we ran different things that we could find. And that's kind of what's happening here where Jason has stepped back in and taken so much data is coming out of these new systems that we need somebody like Jason to actually be able make use of all of that.

And I'm so excited about -- I know that I've been -- the one thing that the miscreants say, the [blaggarts] say that has some truth in it is -- despite there has been a distraction. It's not like I woke up one day and said I want to go get in a fight with the most powerful people in the world, but I felt I had to.

There is a new suit by Fairfax Financial and if you take the time to read that 89 page suit, it reads like a John Grisham novel. And you should read it. If you want to have an idea about what's going on with us, you can just read that. I mean there's so much that hasn't gotten public yet about us. But you can read that suit, and it's a good picture of the kind of stuff we've been fighting. And maybe it will make my actions a little bit more explicable to people. But I'm very excited to be back with an executive hat on running something within the company. And it's an area, to be honest, that has been gnawing at me for several years. And I'm excited to be engaged with it. And it's great to have my partner back.

Thank you.

Okay. Look forward to talking to people in three months, and, again, expect basically more of the same this quarter. And I'm sorry if this is going to turn out to be a one year mulligan where I screwed up and I'd let us come off the rails and it's going to take us a year to get everything back on the rails. But I think we'll be just fine. Bye-bye.

Thank you for your participation in today's conference. This concludes the call and you may now disconnect. Have a good day.