CoStar Group Inc (CSGP) 2003 Q2 法說會逐字稿

Welcome to CoStar Group's second quarter 2003 earnings conference call. Today, we have with us Andrew Florance, President and CEO, Frank Carchedi, CFO, and Mark Klionsky, Senior Vice President of Marketing and Corporate Communication. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question and answer period. If you would like to ask a question during this time, press *, then the number one on your telephone keypad. Thank you. I will now turn the call over to Mr. Klionsky.

Good morning. I'm Mark Klionsky, and I'd like to welcome you to CoStar Group's second quarter 2003 conference call. Before I turn the call over to Andrew Florance, President and CEO of CoStar, let me state that certain portions of this discussion include forward-looking statements which involve many risks and uncertainties that can cause actual results to differ materially from such statements. Important factors that can cause the actual results to differ include, but are not limited, to those stated in CoStar's second quarter press release and in CoStar's filings with the SEC, including its form 10Q for the quarter ended March 31, 2003 under the heading risk factors. All forward-looking statements are based on information available to CoStar on the date of this call, and CoStar assumes no obligation to update these statements. Please note that for a complete copy of our second quarter press release, please refer to the press release on our Website, filed with the SEC or distributed by PR Newswire as business wire transmitted in incomplete press release. In addition, please visit our Website at www.CoStar.com\corporate\investor for a Webcast of this conference call and for the reconciliation of all non-GAAP financial measures discussed on this call to GAAP basis results. Andy?

Thank you, Mark. Welcome to the second quarter 2003 conference call. CoStar Group is currently in registration, so remarks on this conference call will need to be limited in scope. We are again very pleased to report continued strong earnings growth this quarter. Our EBITDA, which is our earnings before interest, taxes, depreciation and amortization, increased by 108 percent from $1.4m in the second quarter of 2002 to $2.9m in the second quarter of 2003. In addition, our cash balance has increased by $5.3m during the second quarter of 2003 over the first quarter of 2003. Revenues for the second quarter of 2003 were $23.2m compared to $19.5m in the second quarter of 2002, an increase of 18.6 percent year over year. We expect to reach GAAP basis net income in the third quarter of this year and expect to continue to show consistent earnings growth throughout 2003. Frank Carchedi, our CFO, will address the second quarter financials in more detail later in this call.

The company has continued to show strong earnings growth in a weak commercial real estate market. While we are encouraged by a turn from negative to positive net absorption of office inventory, office leasing activity and rental rates continue to decline across the aggregate United States. These declines have a direct negative impact on the financial health of our core client base. Our products have performed well in this environment, though, because they often represent a significant cost savings for our customers. In addition, we feel that our products provide a competitive advantage to our clients at a relatively low cost point in an increasingly competitive market. Our December 2002 release, the new CoStar Property 8.0 has been very well received in the market. CoStar Property 8.0 includes new features and capabilities that further enhance our value proposition, strengthening our appeal to new customer segments. During the first six months of 2003, we signed over 500 new customers, firms that previously did not subscribe to any of our services in a particular market. While the company believes that our earnings growth rate would accelerate in a recovering commercial real estate market, we believe that we remain well positioned to grow through the current weak market conditions.

As I have mentioned in previous conference calls, one of our organization's top priorities is improvement in the quality of customer service we provide to our clients. We believe that one of the key metrics of customer perceived value for our service is product usage. Proactive customer service manifests itself in higher usage rates of our products. This year, our entire organization, and our sales force in particular, is focused on driving higher usage of all of our products and our new CoStar Property 8.0 in particular. We believe the company will experience a number of benefits if we can successfully migrate the majority of our users to our new version of CoStar Property. First and foremost, we believe Property 8.0 is an outstanding product, and we want our customers to judge and recommend the value of our service based upon the superior capabilities of this new product.

Secondly, when we migrate the majority of our users over to the new product, we will discontinue the older service and thereby expect to save the company as much as $1m in operating expenses annually. We also believe that discontinuing the older service will eliminate some level of product theft. We hope that a significant portion of those denied illegal access will become legitimate subscribers and contribute to further revenue growth.

Finally, because the new CoStar Property 8.0 integrates directly into our customer relationship management system, it affords us much greater visibility into how our clients are using our product and how we can better serve them. We believe high usage rates are a good indicator of our future performance because customers who use our products regularly are much more likely to renew their subscription, buy additional products, recommend CoStar to other potential customers, and purchase our products as they leave one firm to join another. They are also more likely to feel that the charges for our services represent a very good value.

Our sales force has responded to this usage initiative with a dramatic increase in productivity. Our account executives conducted approximately 590 training sessions during the first six months of last year. During the first six months of 2003, the same group conducted over 5,100 face-to-face customer training sessions, often at the customer's site. That's an 800 percent increase year over year in face-to-face training sessions. Despite the additional burden on their time, the sales force maintained strong net new sales, resulting in a 2.7 percent sequential quarterly increase in revenues from Q1 to Q2. We believe this significant accomplishment represents a positive shift in our sales organization's culture toward a more intensely customer-oriented focus. We believe this productive new culture should solidify our customer relationships and win new customers, which, in turn, could drive long-term earnings growth and shareholder value.

We can directly see the impact of our hands-on training of our customers in the dramatic growth in usage of CoStar Property 8.0. In mid-December, only about 3 percent of our customer sites had access to the new CoStar Property system. As of this morning, just seven months into the rollout, 99 percent of our customer sites have access to the new system and 71 percent of our revenue base is actively using the Web product. Usage of CoStar property 8.0 has soared from approximately 1.4 million weekly page views at the end of the first quarter to approximately 2.5 million weekly page views at the end of the second quarter to approximately 2.8 million page views last week.

Let me compare these usage levels to our established successful comps product, which for the last four years has been our leading Web-based product. CoStar comps generates a little over 1m page views per week, and by comparison, CoStar Property 8.0 in less than a year is already generating almost three times that weekly usage level. The number of customers actually using the site has doubled over the course of the second quarter. We continue to receive positive customer response to the new product. These usage statistics confirm our best hopes for the smooth and successful deployment and adoption of this major new product. Jones Lang LaSalle will be the first of our major national customers to completely convert to the new CoStar Property 8.0, shutting off the older desktop software. We've already begun removing the desktop application from a number of their offices and expect their professionals to be fully off the desktop product by the end of August.

As I've mentioned, with the desktop version of CoStar Property, it was difficult for us to analyze who was using the product or how often they used it. With the new Web-based system, we can precisely track each individual client's usage of the product. Previously, we could only tell that a site was using the product and not which individual was or was not using the products. With our new visibility, our sales force is introducing the product to many, many additional new users at our customers' sites. Our team is specifically focused on converting rainmakers at our customers' sites to users. We believe the number of users currently logging on to the new Web-based version of CoStar Property may have already surpassed the number of people once using the desktop product. Further, we believe the potential audience for the CoStar Property 8.0 is much greater than the older version, and we are working on ways to monetize that potential.

Since the initial release of CoStar Property 8.0, we have released seven incremental upgrades, adding valuable new features each time. In fact, with the accessibility and flexibility of this new platform, we have been able to release more upgrades for this system in the first six months of this year than we were able to make to the older platform over the course of the preceding four years. In May, we released the much anticipated CoStar analytics module in CoStar Property 8.0. Perhaps the most advanced technology we have developed to date, CoStar analytics provides individual users with all the tools needed to analyze current and historical data on a local and national level with precision and enables customers to analyze several new data elements not previously available, including up to 15 years of historical data in some markets. The analytics module allows for a high degree of customization. Customers can analyze an entire market or a select group of buildings or even just one building. The customer defines the market by selecting which buildings or submarkets to include or exclude for analysis and defines the time period and criteria upon which to run the analysis. With coming releases, we expect that customers will even be able to use this module to forecast future market conditions, analyze their properties and portfolios, and benchmark their performance against peer groups. There are endless possibilities to the analytics that can be run, and we believe our customers will find this enhancement to be one of the most powerful aspects of property 8.0. We believe that no other competitive service comes close to offering such a robust, flexible tool for analysis and forecasting commercial real estate information.

With these enhancements and more on the way, we believe Property 8.0 will open up new avenues for growth for CoStar among rates, commercial banks, and institutional investors that invest in real estate equity, mortgages and securities. The analytics feature gives these customers incredible respect when analyzing their properties and portfolios. For example, a portfolio manager investing in CMBS, or commercial mortgage backed securities, can use CoStar Property 8.0 to look at the performance of a property compared to the general office market or to a property peer group. Every analyst can quickly compare the performance of [Renata's] trophy office buildings to Trizec's trophy properties or a custom portfolio based on any criteria of his or her choosing. Analysis that used to take days for these investment analysts will now take just minutes.

We recently signed a significant new license agreement with Cohen and Steers, one of the leading investment managers in the real estate securities field, and the first investment advisor to folks exclusively on REITs.

I talked a lot about CoStar Property 8.0. I also have some promising developments to report on CoStar Tenant. CoStar Tenant appeals to a broad cross section of our customer base. Building owners and their leasing agents use our tenant base to find tenants to occupy their space. Tenant rep brokers use our tenant database to keep their pipeline full, prospecting for space users that might be in need of representation. And a broad group of our vendor clients, such as moving companies, architects and design firms, furniture suppliers, telecom vendors, subscribe to our tenant data to prospect for sales leads. With approximately $15m in annual subscription revenues, CoStar Tenant is our third largest revenue generator.

In 2001, CoStar Tenant was one of our fastest growing products, and we have historically experienced high renewal rates for it. However, in 2002, we noticed that the renewal rates on CoStar Tenant were not quite as high as the renewal rates we enjoyed on our Property and Comps products. One of the reasons for this was that vendor clients use lease expirations on our Tenant product to find sales leads on companies that may be moving. Historically, we experienced our highest cancellation rates from those customers. On the last month of their annual subscription to CoStar Tenant, these clients could query for 100 to 150 valuable lease expirations for the coming year and then terminate their contract. About a year ago, we began working on a positive solution to this challenge.

In August 2002, we initiated a new research methodology to increase the volume and quality of move leads we produce, and we began testing an enhanced product offering to our vendor customers in New York, Washington and Chicago. This new methodology is generating hundreds of high quality move leads each month that are completely different from and in addition to the move leads we have historically generated by lease expirations. With this new dimension to our product, a vendor that cancels a subscription after downloading next year's lease expirations would miss out on almost 75 of the best move leads each and every month. It's been almost twelve months since we began testing this enhanced offering, and we have seen dramatic improvement in our renewal rates among vendors in two of the three test markets. Some vendor clients have told us the leads we're providing are the best leads they've ever received and now consider the product indispensable. Based on the highly successful test, beginning this month we are expanding the enhanced offering to eight additional markets: Atlanta, Dallas, north New Jersey, San Diego and San Francisco. We expect most, if not all, of these markets to perform in line with our successful market tests and result in substantial improvement in the overall renewal rate for CoStar Tenant. Anecdotally, we've already heard from vendors in these cities that had planned to cancel, that they have reconsidered and now plan to renew their subscriptions.

The profitability of our UK operation Focus has continued to grow. The Focus has made progress in a number of initiatives key to its long-term growth. One of the Focus' primary goals is to consolidate the number of very similar products that produces and distributes property information down from five to one. By consolidating these somewhat redundant products, we expect that the operation will reduce its costs and be able to focus on producing the single most competitive distribution system possible and enable the future integration of U.S. and UK information systems. On June 8, the Focus completed the integration of the unique features of its Focus 98 product into its flagship product New Focus, thereby eliminating the Focus 98 platform.

CoStar has been aggressively pursuing several new product development initiatives which we believe we can bring to market in a series of product releases and upgrades over the next four quarters. However, we feel that at this time it is not in the company's best interest to disclose the details of these product development initiatives for competitive reasons.

At this point, I'm going to turn the call over to our Chief Financial Officer, Frank Carchedi.

Thank you, Andy. As reported in our press release yesterday, revenues increased 18.6 percent from $19.5m in the second quarter of 2002 to $23.2m for the second quarter of 2003 due to the addition of a strong UK operation in the first quarter of 2003 and consistent continued organic growth of the core business. Gross margin has improved from $12.6m in the second quarter of 2002 to $15.5m in the second quarter of 2003, and margin percentages have improved from 64.5 percent to 66.7 percent over those same periods. Pro forma net income, which excludes purchased amortization, improved significantly from $256,000, or 2 cents per share, in the second quarter of 2002 to $1.4m, or 9 cents per share, in the second quarter of 2003. Meanwhile, our GAAP basis result continued to improve from a net loss of $1.3m, or 8 cents per share, for the second quarter of 2002 to a GAAP basis net loss of $367,000, or 2 cents per share, in the second quarter of 2003.

The year over year progress we have made is substantial, but I'm going to focus on a discussion of the second quarter 2003 results when compared to the first quarter of 2003. The sequential results of Q2 and Q1 of 2003 are important in understanding the company's progress and why we believe we are positioned for continued revenue and earnings growth. The total revenue grew sequentially by 2.7 percent overall from Q1 to Q2, increasing from $22.6m to $23.2m. The growth for the quarter was principally the result of further penetration of the potential customer base across our national platform, as well as the successful cross selling of products into our existing customer base. Organic growth of approximately 2.4 percent was achieved in U.S. subscription based information products. The UK operation continued to contribute approximately 7 percent of the revenue for Q2, consistent with Q1.

Including the London-based focus product, subscription revenues for the company account for approximately 94 percent of all revenues during Q2. Gross margin increased from $15m in Q1 to $15.5m in Q2, as margin percentages increased from 66.3 percent to 66.7 percent. Cost of sales increased slightly from $7.6m in Q1 to $7.7m for Q2. With regard to operating expenses, overall operating expenses, excluding purchase amortization, remain constant from Q1 to Q2 at $14.8m. On a more detailed level, selling and marketing expenses decreased from $6.6m in the first quarter of 2003 to $6.4m in the second quarter of 2003. The decrease was principally due to a decline in annual internal sales training costs that's typically incurred during the first quarter of each year. Software development remained constant at $1.7m from the first quarter of 2003 to the second quarter of 2003. General and administrative expenses increased slightly from $6.5m in the first quarter to $6.7m in the second quarter of 2003. This increase was due principally to an increase in consulting services and professional fees during the quarter. Additionally, purchase amortization and operating expenses remained constant at $1.1m from the first quarter of 2003 to the second quarter of 2003.

In summary, our pro forma net income per share was 9 cents for the quarter, an increase of 3 cents per share from the first quarter of 2003. This represents the seventh consecutive quarter of pro forma net income for the company. The continued improvement in pro forma net income results from sequential revenue growth, growth in margin percentages, which over the last year increased approximately two percentage points, and relatively stable operating expenses.

Our earnings before interest, taxes, depreciation and amortization improved from $2.4m in Q1 to $2.9m in Q2 of 2003. Our GAAP basis results improved from a net loss of $843,000, or 5 cents per share, for Q1 to a GAAP basis net loss of $367,000, or 2 cents per share, for Q2 2003. Reconciliation to GAAP basis results of all non-GAAP financial measures discussed on this call, including EBITDA and pro forma net income, is shown in detail in our press release issued yesterday, which is available on our Website.

Moving to the balance sheet, we closed the second quarter of 2003 with approximately $31.6m in cash and short-term investments, an increase of $5.3m over the previous quarter. Growth in cash is principally the result of strengthening operating cash flows and $2.2m in proceeds from the exercise of stock options for approximately 122,000 shares. Receivables continue to be well under 30 days of sales outstanding. We continue to have no long-term debt. We continue to believe that we have adequate resources to operate under our current business plan and that we are in a very strong financial position coming out of the quarter. Additionally, on June 15, 2003, the company filed an amended registration statement in the proposed public offering of 1,250,000 shares of common stock.

Now we'll discuss the outlook for the third quarter. For the third quarter of 2003, as we indicated in our press release, we expect quarterly sequential revenue growth of approximately 2.5 to 3 percent and pro forma net income to grow from 9 cents per share to approximately 12 cents per share. Gross margin percentage is expected to increase by another .5 to 1 percent in Q3 and is expected to continue to trend upward for 2003, based on revenue growth and a relatively fixed cost structure for research. Operating expenses, including selling and marketing, software development and G&A are expected to increase by approximately 1 percent overall in Q3 due to salary and cost escalation. Purchase amortization of approximately $700,000 embedded in cost of sales and $1.1m in operating expenses for the second quarter is expected to remain consistent for the third quarter.

Due to our progress on earnings, we may have growing taxable income. Although our net operating loss carry forwards are substantial, we may be subject to limitations on their use, alternative minimum tax, and state and local tax jurisdictions that may not recognize the loss carry forwards. As a result, we are expecting to incur approximately $200,000 to $300,000 of income tax over the last six months of 2003.

In the third quarter of 2003, we expect to reach an important financial milestone, as we expect to report GAAP basis net income of approximately 1 cent per share.

In conclusion, we believe we have and will continue to demonstrate the strength of our subscription based business model and our ability to execute our plan. We believe we are well positioned to continue to achieve revenue and earnings growth during 2003, much as we have in this difficult business environment throughout the past two years. We look forward to reporting our progress to you.

And with that, I'll open the call for questions and answers.

At this time, I would like to remind everyone if you would like to ask a question, press *, then the number one on your telephone keypad. We'll pause for just a moment to compile the Q and A roster. Your first question comes from Charles Carter of CSFB.

Hi. A couple of questions. The first question is you all had previously guided to $95m of revenue for the full year. I'm just wondering if you're comfortable with that level, or at least in that range, and then what your thoughts are in '04? Lastly, was there any meaningful foreign exchange benefit from UK in the quarter? Thanks.

Charles, as you mentioned, our original goal for 2003 was $95m in revenue. I think it's too early to conclude that we won't reach that, but obviously we will need higher growth rates to do so - higher than the guidance we've given for Q3. If we continue to perform at the current and Q3 guidance growth rates, we'd likely be closer to $94m for the year. We haven't given guidance for '04 yet. You understand how our business model works and how we achieve sequential revenue growth. You've seen the track record of sequential revenue growth, and we're optimistic that we can continue to have revenue growth moving into '04. It's your decision what kind of growth rates you expect to use in your model for '04.

Lastly, on the UK, any foreign exchange benefits of any significance?

There's no foreign exchange benefit that's running through the P&L. I don't want to get too much in detail to describe it. We've got about an $18m asset over there. The value of that asset has increased or fluctuated as the rates generally have gone up here in the last couple of quarters. There is no meaningful P&L effect from foreign exchange because the net earnings from the UK, the net dollars coming through the P&L, are not significant to our overall P&L.

Okay. Thanks very much.

Your next question comes from Dalton Chandler of Needham & Company.

Hi, guys.

Hi, Dalton. Dalton, I'm beginning to believe that you actually give a corrupted name on each one of these calls. Comic relief.

I do tend to change it often. I just wanted to come back to the 500 new customers. I think the time frame you said was in the first half of '03. Is that correct?

Yes.

Can you give us any sense for the customer mix? Was it pretty much the core customer base, or were you starting to attract some non-core customers in that group? Also, what kind of product were they purchasing, if there was some differential there?

Well, it's a cross section of a similar sort client base that we've been seeing. Some of them are non-core customers. An example would be a Cohen and Steers. We continue to bring on banks, which historically have not been the core of our customer base. We've had some of those new customer segments coming in. Then again, also, being mindful of the fact that we still believe we are very likely penetrated for the core business, like the brokerage firms and owners, we're still seeing a lot of business come in from brokerage firms. For instance, Grubb & Ellis came in and purchased our Property product and Comps in Phoenix, and that's a new add in our very core customer base.

OK. I think in previous quarters you've given us an indication of what percentage of revenue was new business and from existing customers. Do you have that number for this quarter?

It's been higher than normal because, again, our salespeople are spending so much time with our existing customer base.

Dalton, if you look at the gross new business that comes in for a quarter like Q2, approximately half of all that new revenue coming in would be from completely new customers - brand new customer penetration, and roughly half of the gross revenue coming in would come by a combination of cross selling and price increases within the existing customer base.

Probably in the in excess of 55 percent coming from existing account growth; almost 60 percent coming from existing accounts.

Okay.

That remains higher than our normal-- that remains higher than our historical numbers, where historically it's been totally balanced. Now it's slightly heavier to the existing account side.

Okay. Is the analytic product driving any of that, or is it driving some of the new customers as well?

I'm sure it is. I'm very confident it is a compelling reason for people to buy the product. We've gotten - I can only answer anecdotally, though. I've gotten e-mails from customers, heard from some new customers that they think it's an outstanding feature set. I think the quote was that, "You guys really hit the nail on the head with this."

Okay. Finally, just a balance sheet question. I calculate you've got the DSOs down to 23 days. I mean, how much more can that improve, and what sort of trend can we expect to see?

That's a good question, Dalton. Let me give some background on that. During the last couple of years, we have been implementing a lot of infrastructure changes throughout the company, including in the billing and accounting area. We have gotten a lot more efficient and effective in billing customers and collecting. I think there are a lot of other reasons for good collections: good products, good customer service. But, I think the billing function has been vastly improved. I think-you know, I have said to people in the past that I believe that DSOs could go under 30 days, and I think we're seeing that to some extent. The reason for that is the way we bill, which is that we will bill customers at the beginning of the month for that month, as many companies do, particularly if you're on a subscription based annual contract. So, the effect of that is that DSOs can readily drop below 30 days. And we've seen some of that. How much more they can improve, I don't know. But, our target has always been to keep them under 30 days, so we're very pleased with this result.

Okay. All right. Thanks a lot, guys.

Your next question comes from James Wilson of JMP Securities.

Good morning, guys. I'd love to get your thoughts-- obviously, some of this is dependent, of course, on how much money you make next year, but on how to think about taxes or paying taxes in '04.

Jim, I don't expect these taxes to be significant in the big picture of our earnings. I don't have a precise number for '04. I haven't given '04 guidance. You're right; it depends to some extent on how much we're earning, although some of these are minimum taxes and that kind of thing and may not be dependent actually on the total taxable income. I think the best answer I can give - I gave a pretty specific number for the next six months, which is $200,000 to $300,000 in total for the six months, for Q3 and Q4. My view of next year is that these will not be dramatic P&L items that will materially affect us. The principal reason I'm throwing that out there now is because, as you know, when you start having income tax expense at any level, it will result in a change in the structure of P&L presentation. I didn't want everybody to be surprised when all of a sudden they saw a new item on the P&L crop up. I am not concerned that this will be a meaningful P&L item. These are typically state and local taxes. There's a corporate alternative minimum tax when you have NOL usage. And as you know, if you've had any experience with alternative minimum tax, it's very hard to predict, and it's not intuitive. You can't basically say it will be X percent of whatever you make. It's a complex calculation, and it's going to round a little bit.

Okay. And, your best thought for the moment -- [inaudible] more material in '05, or is that still hard to tell?

It's hard to tell. I think as we get into the '05 zone, I think we are going to be getting closer to regular taxable income. I certainly hope that we are moving in that phase. I can't completely answer that question, but I think '05 and certainly by '06, I'd be looking at moving out of the position and well into the fully taxable income.

Okay. Andy, could you give a little color on the opportunities and prospects and things developing with commercial real estate lenders: banks, brokers, etcetera?

Certainly. Again, where we are with our product offerings is historically they've been designed and developed because of our origin towards serving analytical needs on a local city basis, so you could analyze vacancy rates, rental rates and that sort of thing on a New York City basis or an LA basis, it's difficult to work with portfolios that span across multiple markets. With our new product offering, we feel we've dramatically increased the analytical capabilities of the product. We've also made it as easy to analyze multiple markets as it is to analyze a single market. It's opened up - That dramatically decreases the amount of effort a bank or a read analyst or reading to put in to try and use our products to analyze their particular property portfolio to make it a lot more attractive to them. We're making good - We're getting some good reception there. We're going to push on that fairly aggressively. We released version 1.0 of the analytic module under the Property 8.0 family. We expect to release an upgrade to that in the next month or so, and that will be followed by another upgrade. You're going to start seeing a whole series of upgrades which will constantly improve the utility of that product towards that new what we perceive as very valuable market segment. We're going to keep chipping away at it. We're fairly encouraged with what we see there.

Okay. Very good. Thanks.

Your next question comes from Chris DeRose of Account Management LLC.

Hi, Andy. Hi, Frank. How are you guys doing?

Hi, Chris. Good.

A couple of quick questions just on the sale reps. How many reps do you have currently; turnover; and number of reps currently with over twelve months experience; where you expect to be year end or perhaps a year from now? And another follow-up.

Okay. You want age and weight?

Yes, please, and hair color.

Okay. We are currently at-- the trained, deployed group is currently at 68, with 90 total - 68 deployed. So, it's down slightly, but largely level. The experience level is continuing to incline. I don't have a specific number on over twelve, but it is higher than it's ever been, which is obviously beneficial. We have extended our training period somewhat, so you are not seeing - That's partially responsible for why you go from 70 some down to 68, because our training period now lasts longer. The new recruits are required to go out and do pure customer service for several months before they can enter into the general sales force. The turnover - We have in excess of 24, just anecdotally off the top of my head that are more than twelve months. If you include sales managers in that, that number would jump to - Oh, there aren't 24 over 12; there are 40 some that are more than 12. With sales managers, that number would be 60 some. That's higher than we've had ever for experience level. The turnover in the account executives for termination is lower than it was in 2002, and the voluntary terminations are about - the departures are about in line with 2002; they're not higher.

Okay. Your training dollars, I remember a year ago, 90 percent was going to new reps. This year, it's more of a 50/50 split in terms of your efforts to really get the existing guys up on especially the new products. Is that going to be changing a bit more now based on your comments towards the new reps as we go along? Have you got the existing guys trained up?

Well, we will continue to spend a significant amount of time and effort developing our existing sales force, even the folks with twelve months. There's a - I can learn a lot more about a lot of our different products. Our products are touching a lot of different market segments and have a lot of different features and functionality. There's an awful lot to learn out there. We're going to continue to invest pretty heavily in our existing sales force in their training, and we think that will yield a positive result in sales and also a higher retention rate for them. On the newer account executives, we're trying to give them - they're definitely getting, through exposure to our customer base in the customer support role, a greater familiarity with our products at a more rapid pace. We're going to probably stay at that 50/50 between focusing the new people and the existing people.

Okay. One last question. In terms of your research staff, the size, retention rates and growth plans for that? Also, when are you going to launch a hedge fund based on your analytics product?

The research department at this point is a stable head count. I think that would be probably 550 or a little over 500 and stable level. That is fairly stable at this point. Retention and turnover is in line with our historical levels over the last year or so and better than our turnover levels two years back or three years back. We are working on an initiative that we think will have a very positive impact in retention in our research department and will also enable us either to do more research with the same number of people or do the same amount of research with less people. That is, the integration of all of our research functions by geography rather than by product and geography. Today, largely, when a researcher - Take the building we're headquartered in here in Bethesda, Maryland. There will be one researcher from the Comps product tracking this ability, and there will be another from Tenant product tracking this ability, and there will be another from the leasing product tracking its ability, each with their different nuances. We have instituted a fully integrated research operation covering the area of San Diego, where researchers receive a little bit more training, and they cover all the aspects of research for an individual building, making them more efficient and more knowledgeable without the specific properties. It also eliminates the kinds of errors and inefficiencies that occur as you communicate how tenant movements might impact leasing movements and might impact asset sales movements. That's going to be a process that will take us at least another twelve months, but that will probably yield a positive result in retention because it makes the job more interesting and creates more promotion opportunities. It will - We're confident it will create a higher level of efficiency in our research operation. I'm seeing some things in our San Diego test that look very encouraging. For instance, this new integrated research team in San Diego is producing 50 percent more move leads for San Diego than the non-integrated team produces in Los Angeles, even though San Diego is a fraction of the size of Los Angeles. I'm seeing some anecdotal, positive signs, and I'll be out there next week talking to those folks and seeing how that's going.

Great. Thank you. What was that total research head count? I missed that.

Just over 500 in the United States.

Great. Thank you. Nice quarter.

With the UK, it's probably in the 550.

Great. Thanks again. Great quarter.

Thank you.

Your next question comes from Ken Abram of Wellington.

Hi, you guys. You know, I'm still scratching my head trying to figure out why you're doing an offering, particularly when I look at the cash flow that you have generated the last four quarters, and that seems to be doing even better, and the fact that if something terrific came up, you could always use your equity. I'm just wondering why we need another $35m on the balance sheet.

Ken, there are a number of opportunities we've been exploring. We think there are some interesting things out there potentially over the next year on the acquisition front. The companies that are providing information services to commercial real estate, where aggregating out two contents together, you can either create significant cost efficiencies or reach new client segments that you couldn't reach as stand alone companies. In approaching any of those acquisitions, it's always very nice to have some flexibility to use both your equity and some cash. Given the company's track record on doing a number of acquisitions that have we think been successful, they can move a little bit faster or more effectively when you have some flexibility into your equity cash mix. When our balances -- with the Focus acquisition, we used a major piece of cash that was on the balance sheet. We've begun to rebuild that position. As you point out, the cash position is building, but we don't feel comfortable doing any acquisition that would take our cash balance below that $20m mark, and we think the company would be better positioned if it had flexibility.

Okay. And, so I take it that those acquisitions out there - The thing that I worry about is doing anything like that and putting the cash on the balance sheet, which not this year, but ultimately is very dilutive, in the sense that if something fell through at the last minute and you weren't able to do something, then the dilution is the dilution. I'm just wondering - I can respect you not taking the cash balances down, but you could also use your stock as currency in an acquisition. Are just the potential targets out there unwilling to take your stock?

No. They're willing to take - I think there are targets out there who want to take the stock. It's just in the ten plus acquisitions we've done, it's always easier if you can do some sort of blend. To get them Focusing on the purchase price of their operation rather than on their investment in CoStar stock, you can focus on the issue at hand which is acquiring them as opposed to them acquiring our stock. If you do some sort of blend there and you get them to take significant portions of - they will take portions of our stock. There's not any one acquisition we're talking about here that would fall through and leave cash on the balance sheet. The company has a long track record of evaluating a whole series of potential acquisitions, and in each one of these, you're dealing with a variety of kinds of sellers. Sometimes, it's the mom and pop who started this company 20 years ago, and this is their retirement nest egg. Sometimes, it's an institutional investor. Again, you don't want to limit the company where we don't have any option but to use 100 percent equity. Again, I would like to also say, we're not looking at any one acquisition that's driving this particular offering. We think there are a number of interesting things out there that we want to follow up on. The company has done ten acquisitions or more in the last ten years. We think we're quite capable of finding some things that you'll find are interesting.

Okay. Thanks.

Thank you.

Your next question comes from Michael Grossman of SX Investment Management.

Hi, Frank. Hi, Andy.

How are you doing?

Good. Question regarding the number of new users that you're tracking on the Web-based system, which was an impressive number. I was wondering how you're taking advantage of that growth on the pricing side, and along those lines of pricing, how the analytics product is getting priced into existing packages.

Okay. The goal of our - We are seeing some immediate price increases as a result of adding additional users at some of the existing sites. There are a number of clients who have contracted who have paid for X users; they need to pay us Y per user. What we're trying to do is we're trying to -- In a fairly compressed time period, we are trying to dramatically broaden the base of usership inside these customer sites and not trying to create the friction of trying to get a piece of paper on each one of these additional users inside the site and rather capture that value at renewal time after we've reached some of our usage goals. We can go into a site where their contract might have initially said they had two users and it was a two-year contract. Their company has grown, and then they grow to have ten active users. When their contract comes up in January of 2004, hypothetically, we then go back and sit down and have a discussion based upon the fact that we can show them that these eight new users are logging into our product on average twice a day or every other day and capture the value at that point. We are very Focused on getting across the finish line with migration into Property 8.0 and terminating the old system because we get we believe a million dollar expense cut, and we also get to capture some revenue growth we believe from creating some conversion of illegal users to legal users. You had a second part of that question.

It had to do with the pricing on the analytics piece.

Oh. The pricing on the analytic piece we are - There are two ways that's being priced. To our existing customers, to a brokerage firm that's been using our product in Denver for quite some time, this is just an upgrade to the product they've already been receiving from us. It is something that drives, we believe, higher renewal rates and better pricing leverage with that segment. What it's really doing is it's opening up new market segments - for example, Cohen and Steers that we couldn't reach before at all. The price point for those companies, because of the breadth of their use, is significantly higher than what you'd see in a local brokerage firm. We price that by a finite number of users, and if they add additional users, there's an incremental fee.

Okay. And my final question is just on what your expectations are for free cash flow, if any, for the year.

I think that - Let me go back to sort of an overall summary for 2003 to answer that. We talked about the revenue earlier. Originally, our goal for 2003 was 40 cents per share of pro forma net income. Pro forma net income, I think, is a very good indicator of free cash flow. The reason for that is that it includes the depreciation charge which has been reasonably comparable to what we spend on cap ex. Looking at the pro forma net income number in that regard, our original goal was 40 cents, which as a side note when adjusted for purchase amortization should be consistent with Q2, would put us at a GAAP net loss of 5 cents per share. At this point, we expect to meet the 40 cents and possibly exceed both the pro forma net income target and the GAAP target by a few cents per share. Hopefully, that helps you put 2003 in perspective a little bit.

Okay. Great. Thanks.

Thank you.

Your next question comes from Gary Shero [ph] from JP Morgan Funding.

Hey, guys.

Hey, Gary.

Frank, how big is your NOL today?

It's approximately $80m in total.

When you said there are limitations, does that mean the timing of the use or that part of it's not going to be able to be used?

There are a number of limitations on it, Gary. Without going into the tax details, one of the principal limitations in effect is the corporate alternative minimum tax, which essentially means that you cannot ultimately use your NOL to completely take out your taxable income. There will be a corporate alternative minimum tax. In addition, there are potentially limitations under Section 382 for changes in ownership. I'm sure you're sort of familiar with that. We're studying that now, and as you know, the company has historically had many changes in ownership. Those changes will cause limitations in that number. We still believe we will have a substantial net operating loss curve for it, and we'll have substantial coverage of taxable income with NOLs. Hopefully, that answers your question.

Yeah. And, I don't know if you said the amount of revenue from non-core revenue this quarter? Is that pretty much flat with last quarter?

It is pretty flat. Off the top of my head - Give me a second, and I can talk about non-core revenue. The best way to look at it is non-core is approximately 6 percent of the revenue for the quarter. The 94 percent that I talked about is core subscription revenue, including the subscription based products that are core in the UK, the very similar products to what we have. Those non-core - As we discussed before, those non-core products are things like ad hoc searches in the database, one-on searches that are not under contract, advertising areas, which is an off the shelf software product, and that kind of stuff.

Right. Finally, can you just give a little detail on maybe growth by your top ten customers versus the rest of your customers or anything interesting in terms of growth in particular markets? Just anything that would look a little different in those areas?

I think probably the best thing that I can tell you is that growth in the top ten customers by and large was not different, which means that they continue to grow. Especially as we come into first and second quarter and we add the revenue streams in the United Kingdom, probably eight of the top ten customers were customers of Focus. That took up our overall revenues from those customers. Anecdotally, I am seeing additional buys from some of our top ten customers. So, we're continuing a long-term track record of the biggest companies in commercial real estate continuing to buy more of our products and services in different geographies and covering different product types. There's no news there, which is good news.

Okay. Great. I appreciate it. Thank you.

Thank you, Gary.

At this time, there are no further questions. Gentlemen, are there any closing remarks?

I'd just like to thank everyone who's still on the call after the discussion of next year's taxes. I'd like to thank you for joining us for this conference call. We look forward to talking to you for the third quarter conference call. Thank you.

This concludes today's conference call. You may now disconnect.