CoStar Group Inc (CSGP) 2017 Q4 法說會逐字稿

Ladies and gentlemen, thank you for standing by. Welcome to the fourth Quarter 2017 earnings conference call. (Operator Instructions) As a reminder, this conference is being recorded.

I would now like to turn the conference over to your host, Richard Simonelli. Please go ahead.

Thank you, operator. Appreciate it. Welcome to CoStar Group's Fourth Quarter 2017 Conference Call.

Before I turn the call over to Andy Florance, CoStar's CEO and Founder; and Scott Wheeler, our CFO, I have some very important items for you to consider. Certain portions of our call today may contain 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 actual results to differ, include but are not limited to, those stated in our February 21, 2018, press release on fourth quarter and year-end results; and in CoStar's filings with the SEC, including our most recent annual report on Form 10-K and our subsequent quarterly reports on Forms 10-Q under the heading Risk Factors. All forward-looking statements are based on information available to CoStar at the time of this call. CoStar does not assume any obligation to update these statements whether as a result of new information, future events or otherwise.

Reconciliation to the most directly comparable GAAP measure to all of the non-GAAP financial measures discussed on this call, including but not limited to, non-GAAP net income, EBITDA, adjusted EBITDA and forward-looking non-GAAP guidance are shown in detail in our press release issued yesterday, along with definitions which have been updated for those terms. The press release is available on our website located at costargroup.com.

As a reminder, today's call is being broadcast live and in color on our website, where you can also find CoStar's Investor Relations page. Please refer to yesterday's press release on how to access the replay, which will be on about an hour after the call. Remember, one question, so make it a good one.

I'll now turn the call over to Andy. Andy?

Thank you, Richard. Appreciate that. That was an excellent preamble. Thank you for joining us today on our year-end 2017 earnings call.

4 years ago, on a similar earnings call in which we reported $476 million revenue run rate, we publicly set a goal of reaching $1 billion in annualized revenue by the end of the fourth quarter of 2018. I'm very pleased to announce that with the fourth quarter 2017 revenue of $254 million, we have met our $1 billion in revenue run rate goal an entire year ahead of target. And I'm even more pleased that we crossed the $1 billion milestone with the strongest sales momentum I have ever seen in this business.

In the fourth quarter of 2017, we generated $43 million in net new bookings, a sales increase of 47% year-over-year. For the year, we added $127 million in revenue and achieved our best sales year ever with $148 million in net new bookings, up 32% over 2016.

Leading the strong momentum, CoStar Suite net new bookings increased 100% in Q4 2017 versus the same period in 2016. Not that the same period in 2016 was weak, quite literally, sales of our flagship service were off the chart in the fourth quarter. We added more new companies than ever before, with 3,100 net new North American clients added at our CoStar subscriber base, which is nearly triple the number the same quarter a year ago. We added 5,863 new users to CoStar in the fourth quarter of 2017.

For all of CoStar, December was our best sales month ever by a wide margin, and January was our second best ever by a wide margin. As a company, we sold more in December 2017 and January 2018 than we did in the entire year of 2011.

In the fourth quarter 2017, Apartments.com had its best net new sales bookings quarter ever. Apartments.com sales were up 36% year-over-year in the fourth quarter. This was particularly strong performance since, historically, the fourth quarter for Apartments.com and all the ILS companies is normally a seasonally weak quarter. For the full year of 2017, CoStar revenue growth was 15% year-over-year compared to 2016.

4 years ago when we set that $1 billion revenue run rate goal, we also set a goal of reaching a 40% adjusted EBITDA margin for the fourth quarter of 2018. We've made -- we are making good progress towards our EBITDA margin goal and remain focused on achieving those very strong margins by the fourth quarter 2018, as we stated.

Net income for the full year of 2017 grew 44% year-over-year, and EBITDA increased 10% year-over-year. Our unprecedented sales successes in the fourth quarter 2017 drove unprecedented commission paths for our sales quarter -- force in the fourth quarter, that's a good thing. The ROI on these sales are fantastic and clear. This continued into 2018 with high commission expenses in January. This suppressed our EBITDA growth in the fourth quarter, but in the intermediate term, this exceptional revenue growth is expected to enhance margins.

We are in the middle of a seismic shift in U.S. commercial real estate information analytics and marketing landscape. In the middle of December 2017, our major competitor, Xceligent, suddenly and unexpectedly filed for Chapter 7 bankruptcy and liquidation. Xceligent had been in business since 1999 and was operating in over 40 major U.S. cities.

We believe that Xceligent had well over 5,000 commercial real estate customers. We believe that various investors over the years had poured over $200 million into Xceligent. When we acquired LoopNet, we agreed to divest LoopNet's partial interest in Xceligent as part of an FTC consent decree to replace any lost competition in the merger.

At the time of Xceligent's bankruptcy, we were engaged in a major lawsuit with them. A former employee of Xceligent had tipped us off that Xceligent was stealing and reselling our data on an industrial scale. We investigated and found that, that was true, that Xceligent had created thousands of fake passwords that have accessed our servers millions of times -- in fact, even 10 million times, to copy content from us.

We found tens of thousands of our copyright photos shot by our employees that were copied and resold on Xceligent's site. We had statements from many Xceligent employees that the theft was a widespread practice of Xceligent supported by senior executives directing their employees and contractors to steal CoStar data and photos and use them to build and sell Xceligent's products illegally. We believe we had a clear, great case.

If the lawsuit contributed to Xceligent's death, the real underlying cause of death was Xceligent's massive and sustained financial losses. We believe that Xceligent failed to make any profit for years. In fact, we believe, over the course of 2 decades, Xceligent never made a profit or even approached breakeven. We do not have exact numbers, but we believe that by the end of 2017, they were spending roughly $5 million a month -- or losing about $5 million against approximately $2 million or so of monthly revenue, so $2 of loss per $1 of revenue, roughly.

Immediately after Xceligent's bankruptcy, the recently fired CEO of Xceligent launched a brand-new company called Intrepid. That claimed to have roughly the same data coverage and software that Xceligent had the prior weeks when it closed its door. The bankruptcy trustee appointed an Xceligent Chapter 7 filing, shut the startup down relatively quickly.

With Xceligent's failure, somewhere around 5,000 companies suddenly found themselves without a commercial real estate information system. This left us with an unprecedented opportunity to act quickly to pick up tens of millions of dollars of orphaned business, an important moment.

We pulled out all the stops to respond at the maximum level of effort. I personally worked around the clock on one weekend December loading up every credit card I had in order to send holiday gift baskets to almost every former Xceligent customer. I literally was debating with folks at customer service centers in India at 4:00 a.m. telling them I was good for an increase in my credit line. In the weeks that followed, we hand-delivered thousands of gift baskets, getting our salespeople into the doors of thousands of these people immediately was a good investment.

On short notice in January, we then launched a 30-city roadshow to market where Xceligent had a presence. We hit the road with all of our senior executives, personally visiting 1,500 of Xceligent's former clients to let them know all about the valuable CoStar service we had to offer them. We went to markets big and small, all of them important, from Houston, Atlanta, Miami, Minneapolis, Little Rock, Northwest Arkansas, Oklahoma City, Kansas City, Columbus, you name it.

Our objective was to meet as many former Xceligent users as possible to educate them on the depth and breadth of our service offering. More importantly, we wanted to demonstrate the substantial investment we had made and we're making to cover local markets with real quality, and why having great data, technology costs what it does for a subscription.

Our service costs more than Xceligent's did and it's well worth it, and it's sustainable. All told, about 1,000 CoStar staff were involved in these meetings in person or via video conference, with hundreds of flights. I believe January 2018 was likely our highest travel expense month ever. In order to get 1,500 companies in the rooms on 1 week's notice, we give away Apple iWatches to highlight our very cool upcoming CoStar for Apple iWatch product.

Remember that the potential net present value of signing up each one of these clients typically can be $50,000 and much, much larger. So giving them an Apple iWatch to get them into the door for a sales pitch, where you have about a 30% to potentially 60% close rate, is a no-brainer.

The effort has already resulted in our signing over -- up to about 1,000 former Xceligent clients, 1,000 former Xceligent client firms. In addition, this already had a dramatic effect on further improving our data as these new clients are regularly contributing updates and sharing information with our research team much more proactively than they had been.

It's my belief that more of these clients -- the more these clients we sign up now, the more we will eventually sign up. I believe this Xceligent development represents a $50 million plus annual revenue opportunity for us. Many of the former Xceligent customers that have yet to sign up for CoStar were relying on LoopNet or LoopNet Premium Searcher as a low-cost alternative to CoStar.

So updating you on LoopNet. Yes, it was another driver behind our fourth quarter sales achievement, and the successful integration of CoStar and LoopNet database has made that possible. Prior to the fourth quarter of 2017, loopnet.com and costar.com were each supported by their own separate databases. They were linked and the information flowed back and forth between these databases, somewhat imperfectly. But by definition, it could be improved.

The separate databases drove higher costs for both CoStar and our clients while degrading data quality. CoStar's service is our highest-quality professional information tool used by more than 100,000 industry professionals. LoopNet is a marketing platform with unmatched audience of approximately 5 million monthly end-users shopping for commercial real estate each month -- or a month. Those services are well-regarded for their respective brand strengths.

There is brand confusion, however, significant brand confusion, because LoopNet also had 3 legacy information products that were, frankly, inadequate for professionals. They were Premium Searcher, Property Comps and Property Facts, low-cost information solutions that were not proactively researched like CoStar is, so the information they presented were notoriously incomplete and inaccurate.

Premium Searcher was the largest LoopNet information product and allowed users that paid a monthly fee to see both the advertised listings on LoopNet as well as the ones that brokers had listed for free. This means that a Premium Searcher could see 20% to 40% more listings on LoopNet than a free LoopNet user could see.

As long as LoopNet remained both a great marketing platform and an inadequate information solution, there was too much brand confusion, which prevented us from optimizing the marketing solution potential of LoopNet or the information solution potential of CoStar.

With the pre-integrated LoopNet and CoStar, half the commercial real estate world updated LoopNet and the other half updated CoStar making neither one the clear and simple go-to solution. That increased research costs and degraded quality.

Several years ago, a number of LoopNet unlimited listing plans were sold, which allowed brokerage firms to market huge numbers of listings on LoopNet for as little as a few dollars a listing, which is just inadequate. These plans were misused and cannibalized both LoopNet and CoStar revenue. As part of the integration, we've been discontinuing these unlimited listing plans in addition to Premium Searcher itself. And there's a little bit of a headwind created to our sales as we wind down those unlimited plans.

We notified the client base in the early fall of 2017 of our intentions to discontinue these products. Almost immediately, a large volume of the LoopNet Premium Searchers began migrating to CoStar. As these users began voluntarily migrating from LoopNet to CoStar, we lost the revenue they were paying LoopNet for information, but typically, gained 4x that revenue for a CoStar subscription.

Just a few weeks ago, we notified Premium Searchers that their service will be discontinued this month. There are a small number that will be discontinued throughout the year, but the overwhelming majority are discontinued as of this month.

As we discontinue Premium Searcher in February 2018, we expect to lose approximately $1.8 million in monthly revenue immediately, but expect to more than make up the lost revenue within the year as we convert many of these clients to much more powerful and profitable CoStar information solutions.

We expect to see a sales headwind in February 2018, but with offsetting gains through both 2018 and '19. This clearly suppresses margin expansion in the first 2 quarters of the year, but drives really valuable, potential intermediate and long-term EBITDA expansion.

There is never an easy time to intentionally eliminate $22 million of annual revenue. But the current timing seems the best I believe, and I believe will help us sell more CoStar information and more LoopNet marketing. With LoopNet Premium Searcher and Xceligent gone, I believe there is more market clarity as clients migrate to CoStar.

Through the end of January 2018, we converted over 5,400 LoopNet Premium Searchers or heavy searchers to annual CoStar subscription users at nearly $520 net new per month. Many of these LoopNet users upsold to CoStar were paying 0 before, and those paying were paying about $145 per month on average. So the blended average LoopNet price was $49 prior. That successful conversion drove the massive surge in CoStar sales in the fourth quarter 2017.

We're not done with that. We have really just begun this conversion process. There are approximately 80,000 more discontinued LoopNet Premium Searchers or heavy LoopNet searchers who now have dramatically less access to free content, and we're focused on upsigning them to CoStar this year. This is a great investment for these prospects to make in CoStar, and it's the most important revenue opportunity, I believe, we've ever had. It is our #1 priority.

We believe that we will continue having success upselling LoopNet accounts. The amount of information a broker gains by upgrading from LoopNet to CoStar is impressive and clear. In a market like Phoenix, in the past, a broker may have had -- previously using LoopNet, a broker may have access to about 75% plus of the available listings on Premium Searcher. Now that broker can only see 44% of those listings.

In addition, only CoStar for Phoenix provides details on tens of thousands of tenant sales comps, lease comps, market analytics, industry news, forecasts and a lot more. We feel it's also a compelling investment many will make. We also believe that they will renew at a very high rate, that's been our experience.

For such a major shift in the competitive landscape, we commissioned a third-party research firm to conduct extensive market research and focus groups during this past December, January and February. We wanted to understand better, real time, how commercial real estate firms were reacting to events so we could better tune our sales and marketing messages.

One preliminary result you might find interesting is the response of more than 1,000 CRE professionals surveyed had to the following question. A CRE information services you use, which one do you tune -- turn to first? 56% stated CoStar, 31% said LoopNet, 3% said Catylist, 1% said Real Capital Analytics, 1% Pierce-Eislen, 1% AIR, less than 1% Reis, less than 1% CompStak, less than 1% Axiometrics, less than 1% Proptylink, and 4% said other. So that's 40 -- 87 -- that is 87% CoStar Group and 3% for the next closest player.

There was one moment of humor in one of the focus groups when the moderator said, "With Xceligent, what's the greatest weakness?" And there was a long pause. And then one of the participants offered up, "They're bankrupt."

In addition to operating an information service at xceligent.com, Xceligent operated a marketing website that competed with LoopNet, called CommercialSearch. When CommercialSearch was up and running, LoopNet captured 23x more traffic. Now LoopNet captures more than 40x more traffic than the second closest CRE marketplace competitor.

These developments are possibly the best news for CoStar Group in a decade. Not sure what the better news was on the prior decade, but we'll just leave it there. As we move to capture all the potential value now open to us, we have made significant investments and discontinued significant revenue faster that planned earlier, thereby driving down margins in the first half of the year.

I believe trading a few hundred basis points of margin for several quarters in order to jump on a potentially once-in-a-lifetime significant long-term revenue and competitive gains is an obvious choice. This is a business, not a spreadsheet. Even so, we're not moving away from our 40% adjusted EBITDA margin goal for the fourth quarter of 2018. We are more confident than ever of our ability to reach the goal because of all these recent market developments.

One of the key factors that made all of these sales possible was our investment in our Richmond global research center. In a little over a year, we elevated the industry's best database and analytics offering to new heights of excellence. We are now proactively updating 92% of our over 1 million active listings every 30 days, in person or over the phone, while vastly improving our tenant data with our new nearly -- newly hired 250 tenant researchers.

With the addition of CoStar Listing Manager featured in October 2017, our clients and users are adding hundreds of thousands of updates directly themselves each month in real time. In the first 4 months, they've averaged over 700,000 updates to listings per month. That's a lot of work now being done directly by the folks listing the properties that used to be done by our researchers. Our investment in people and technology made this possible and our data products have never been better as a result.

With an integrated and easier-to-maintain database, the new ability for brokers to update their own listings and greater industry cooperation, we believe and expect that research costs will begin trending back down within the year.

As you know, we recently were pleasantly surprised by the timing and receiving approval from the FTC to move ahead and close the ForRent acquisition. We closed the acquisition yesterday morning, and there was a little bit of use of our capital there. This is the largest acquisition we have done to date, measured by revenue and the number of employees in the acquired firm.

I met with our new colleagues in Norfolk, Virginia yesterday to welcome them to CoStar and to thank them for joining forces to strengthen the #1 multifamily Marketplace network in the United States. And I see a number of our colleagues are on the earnings call today. Welcome, everybody, again.

ForRent has 16,000 advertised properties on its network of multifamily sites. In addition to the primary site ForRent.com, they also offer the targeted sites AFTER55.com, CorporateHousing.com, and ForRentUniversity.com. Similar to ApartmentFinder, we plan to run ForRent as a separate website with its own distinct and different website experience and user interface.

And we've -- as we have demonstrated, it is valuable to have multiple leverage consumer brands in the apartment rentals website space. With the acquisition of ForRent.com, our multifamily sales force increases by nearly 50%. I'm expecting an incredible year from this bigger team in 2018.

Just like ApartmentFinder, we now plan to do a specific branding campaign or TV campaign ForRent, as we did with Jeff Goldblum for Apartments.com. We're focusing all of our efforts on Apartments.com. We plan to focus on online marketing for ForRent. However, there is no doubt that the advertising and brand work for Apartments.com will benefit ForRent because it gives our salespeople better access to buyers because of the power of the unprecedented Apartments.com marketing reach.

We intend to move with all possible speed to connect and drive ForRent from the same content-rich back end that powers Apartments.com and CoStar. Also, we're moving as quickly as we can to give ForRent's customers additional exposure on the existing Apartments.com network. We believe that this will reduce customer churn at ForRent.com pretty quickly. The timing of the closing will reduce margin over the next 2 or 3 quarters, but we expect to expand margins significantly in the quarters beyond.

Apartments.com continues to grow and strengthen. In less than 3 years, we have transformed the industry. We're extremely proud of what we've done since signing the Marketplace side of the multifamily business in 2014 when we purchased Apartments.com. The 2015 acquisition of ApartmentFinder and WestsideRentals acquisition in '17 were awesome, and we're looking forward to ForRent's contribution beginning today. A lot of companies, it's a mouthful.

Multifamily marketplace revenue has grown to $280 million in 2017, up from $86 million in 2014. With revenue of $76 million in the fourth quarter 2016, we finished the year with an annualized revenue run rate of $304 million. I believe that is, by a wide margin, the largest revenue run rate of anyone in the space.

Adding ForRent's pro forma revenue to that, in 2018, our multifamily market revenue approaches $400 million, and we expect to exit 2018 with over $440 million of revenue run rate for our multifamily Marketplace business. It's a lot of progress from the $86 million. The business is solidly profitable, and we expect we will continue to expand margins as we add revenue.

In 2017, we had the most traffic of any apartment listing website. In the fourth quarter 2017, according to comScore, our network attracted over 44 million unique monthly visitors in the aggregate, for an increase of 44% year-over-year. While our closest competitor, RentPath, had 21 million unique visitors in aggregate, which represented a 7% decrease year-over-year.

In January of this year, with the new ForRent-enhanced Apartments.com network, we would have almost 3x the number of visits that RentPath had. For the 27th consecutive month, Apartments.com was the most visited apartment listing network. We had more than 468 million visits for the year, up 25% year-over-year. This is 2.5x the number of visitors to the RentPath apartment sites.

With the acquisition of ForRent, it was no longer necessary to renew our 2-year agreement with Move.com. We expect to be adding millions of additional unique visitors and visits from ForRent. We will save millions of dollars here by discontinuing that relationship.

Throughout 2017, our Apartments.com network was ranked #1 in traffic in 206 U.S. markets or 98% of all markets tracked, and #2 in just a small handful of markets. With the addition of ForRent, we're now ranked #1 in every U.S. market tracked. Our active rentals increased 31% in 2017, and we now offer 1.1 million apartment availabilities that have incredible value, bringing consumers and property managers together.

In 2017, we delivered tens of millions of leads to apartment property managers and owners, resulting in more than 5.1 million leases that were executed because of our sites. 5.1 million Americans moved into homes they found in Apartments.com in 2017. We are delivering meaningful traffic with real renters better than anyone else in the industry by far. Our lead-to-lease conversion beats all others by far.

Apartmentos.com had an excellent first year as we generally -- generated nearly 8 million visits to the site, resulting in 15 million property views. Our PR campaign generated 4 million impressions. Apartamentos.com is the only exclusively Spanish language rental listing site in the United States. With 20% of the U.S. renter market speaking Spanish, this is a tremendous market. Using Apartamentos.com, property managers can receive inquiries or rentals translated into English and Spanish.

We'll begin our new advertising campaign for Apartments.com on March 12, 2018 featuring Jeff Goldblum as Brad Bellflower. I genuinely think Jeff Goldblum is absolutely loving this role. In 2018, we expect to reach 95% of the U.S. households with over 5 billion impressions with another robust national campaign, which will run from March through September.

Our TV campaign is expected to conclude 8,000 commercials. We plan to use a combination of broadcast, cable and syndication television, so we'll be able -- we'll be in the top prime time shows, season premieres, major sports events, such as NBA Playoffs and college football. We will also be reaching the cord-cutting audience with on-demand video, such as Hulu, and streaming video and devices like Roku, PlayStation and Xbox. Even my kids will see Jeff Goldblum.

We're going everywhere renters are, including print and social media. Digital renter retargeting on social media generates millions of visits and hundreds of thousands of leads a year. You will also be hearing Brad Bellflower on local radio. We plan to run an estimated 18,000 spots across 10 major markets. We will also have a strong presence on streaming audio platforms, such as Pandora and Spotify, where we expect the number of impressions to exceed 100 million.

With the addition of ForRent sales teams, we now have nearly 350 people -- salespeople in our multifamily sales force, actual producing line folks. Looking ahead, with the best sales force in the industry, we expect to continue to penetrate the market opportunity faster and provide unprecedented client service in the process as a top priority.

Once again, CoStar Real Estate Manager continues to shine. It turned in another magnificent quarter of sales as companies continue to move to compliance with FASB ACS 842 (sic) [ASC 842], pretty exciting regulation, which requires them to include the value of practically all leases on their balance sheets.

In 2019, we estimate that over 3,500 U.S. issuers will be required to comply. And in 2020, another 1,500 private companies will be required to comply. It's basically a Y2K all over again for commercial real estate, and we believe we have the best accounting solution -- or management solution in the business.

We only have about 5% at this point of total clients and so we have a lot of room to continue to grow here. I expect that sales will remain strong from this group through the next couple of years at least. Real Estate Manager turned in its best year ever with a magnificent fourth quarter sales and a fantastic December. Looking at the chart, it pretty much just goes right through the roof in the last part of the year.

In 2017, net new sales were up 183% over 2016. A CoStar Real Estate Manager salesperson, [Jerry Pring], set the record for the highest subscription sales month ever in CoStar company history in December. In addition, Real Estate Manager turned in another solid quarter of notable client signings including nationwide, Citibank, KeyBanc and Ingersoll Rand, Priceline, BNY Mellon and, one of my favorites, Ryder Trucks. This business -- no really, they're a great firm. This business is profitable and we're investing in it to drive what we believe is significant long-term growth and a unique opportunity. A number of these folks who subscribe to CoStar Real Estate Manager also subscribe to CoStar.

Over the past few quarters, we have released powerful new analytic capabilities in the CoStar product. Our proprietary same-store rent series fully control for the changing composition of the properties and spaces on the market, and offer the most accurate view our rent trends to the properties' submarket, market or national level. We have also pioneered property level forecasts to take into account every building's recent performance, future leasing and current vacancies and rents.

Finally, all of our analytic reporting is now truly real time. Any time any of CoStar's 1,850 researchers updates information on a property or when one of our 100,000-some brokers who are putting in listings directly update something, that new information is immediately reflected in the forecasts, the reports and the data export.

We believe these features provide our clients with the best tools to analyze the state of commercial real estate market and gauge the outlook. To harness the full potential of this analytic tool kit, CoStar assembled the largest team of real estate analysts in the industry, led by a seasoned team of senior economists. Our data shows a broadly healthy marketplace. Commercial real estate vacancy levels are at cyclical lows and pricing has reached all-time highs.

In the office sector, vacancies are stabilized at 10.3% where they stood during the second half of 2017. It's a pretty healthy level. New construction remains subdued, with overall construction numbers running at half of last cycle's peak. Fortunately for owners of existing buildings, leases rolling over now had typically been signed 3 to 5 years ago, new lease rates are higher than those rolling over. Therefore, net operating income growth in the office sectors is prime for good news in '18. I think that's true with a number of the asset classes.

However, rent growth has slowed over the past year. A strong growth of '15 and '16 have given way to more typical gains, about 1.5% year-over-year for office, 2% for multifamily, 2.5% for retail. Industrial, on the other hand, continues to post extraordinary growth in excess of 5%.

Other signs also point to a maturing cycle. REITs have underperformed the broad market over the past year. However, supply for industrial multifamily and office will put pressure on tight fundamentals. The homeownership rate has risen for 4 conservative quarters, heralding the end of an unprecedented era of moving one way towards apartment demand. Commercial real estate transaction volume, which reached all-time highs in '16, was lower in '17. And CoStar's analysis of transaction data shows flattening prices. Cap rates also appear to have risen marginally in response to interest rate increases.

That said, the string of good economic news, 3 quarters of healthy GDP readings and strong job numbers, as well as the passage of the Trump tax cuts, should provide some fresh tailwinds to real estate fundamentals. However, the encouraging economic data has also put the fed on notice and the FOMC raised the policy rate 3x last year, with future hikes expected this year as you know. The prospect of higher interest rates could erode an essential value proposition of commercial real estate this cycle, which was the wide spread of derisk-free rates.

For the multifamily sector, on the other hand, high interest rates may support more demand as would-be homebuyers are priced out of the mortgage markets and continue to rent. If you have inflation as the primary discussion topic of 2018, that should be good news for the commercial real estate market which is viewed as an effective inflation hedge, best of all, for the hospitality sector, which reprices daily.

That said, when hot markets stabilize, some bad deals can get done, that's why we're building out new analytic offerings and strong news organization to give our customers, brokers, property managers and owners/developers the tools to navigate these more volatile waters. Taken together, we see a healthy but maturing commercial real estate multifamily sector.

If you have investments in commercial real estate, debt, equity or any CRE-related companies, we recommend you subscribe to CoStar as the best source to keep abreast of market developments as they happen. Also, let your friends, colleagues and even vague acquaintances know.

So 2017 was a fantastic year for CoStar, and we feel that we have more opportunity than ever. Recent developments, like investments to capture share following Xceligent's bankruptcy and the opportunity to close on the highly potential accretive ForRent acquisition have created margin pressure for the first half of the year. But we believe we'll dramatically expand EBITDA generation in our immediate term and remain on track for our 40% adjusted EBITDA margin goals in 2018.

And at this point, I'll stop talking and turn the call over to our CFO, Scott Wheeler. Thank you, Scott.

All right. Thanks, Andy. It certainly was strong momentum exiting 2017, which, of course, sets us up for a very strong financial year in 2018 and beyond.

As Andy mentioned, the revenue growth in the full year 2017 increased 15% over 2016; while our revenue growth rate in the fourth quarter of 2017 was 16% versus the prior year. Organically, our revenue growth rate in 2017 came in at 14% for the year and 15% in the fourth quarter, after normalizing for the 3 small acquisitions this year and the THOMAS DAILY acquisition in 2016.

Let's look at our revenue performance by services. CoStar Suite revenue growth increased to 15% in the fourth quarter of 2017 versus the fourth quarter of 2016. This came in above the top end of our 13% to 14% guidance range and accelerated from the strong 13% revenue growth rates in the first half of 2017.

The strong revenue growth is, in large part, a result of our investment in Richmond in our research capabilities and in our success in converting the LoopNet information users to CoStar. We expect CoStar revenue growth rates to improve further in 2018 as we accelerate the LoopNet conversions and we reach more of the former Xceligent customers that are in need of commercial real estate information.

Revenue growth rates for CoStar Suite are expected to be in the 18% to 20% range in 2018, which is a significant increase from the approximately 13% revenue growth rates over the past 2 years.

Revenue growth rates in the Information Services sector remained negative in the fourth quarter of 2017, as expected as we continue to wind down the LoopNet information products. As Andy discussed, we've decided to accelerate this conversion actively and effectively discontinuing the vast majority of our LoopNet Information Service offering by the end of February 2018.

LoopNet information revenue is expected to drop from $32 million in 2017 to just $4 million in 2018, a reduction of almost 90%. Conversely, revenue from strong sales in our Real Estate Manager business and our other Information Services products line is expected to increase from approximately $40 million to around $52 million, a growth rate of almost 30%. As a result, we expect the total revenue from Information Services to decline at a rate between 20% and 25% negative on a year-over-year basis throughout 2018.

Given our success to date up-selling these LoopNet information customers to CoStar, we expect to more than offset all of this revenue decline with sales of CoStar Suite in the coming quarters.

We had a very strong fourth quarter in multifamily, as revenue increased 26% year-over-year and 23% on an organic basis. As a result of continued strong sales, we expect organic revenue growth to continue at over 20% in 2018. With the addition of ForRent, we expect multifamily revenue growth of between 40% to 45% for the full year. Consistent with our initial estimates, we continue and expect the revenue contribution from ForRent to be tin he range of $75 million to $85 million on an annual basis post-integration.

Finally, our commercial property and land revenue grew 19% year-over-year in the fourth quarter of 2017. Organic revenue growth, adjusting for approximately $2 million in revenue from the LandWatch acquisition, was 12% in the fourth quarter. Strong revenue growth continued in our LoopNet tiered advertising products, growing over 60% in 2017 versus 2016, while our LoopNet Premium Lister revenue growth moderated somewhat in the fourth quarter of 2017, as a result of temporary disruptions from the CoStar LoopNet integration and our decision to discontinue certain non-subscription advertising products. We expect organic revenue growth in commercial property and land in the 12% to 14% range for 2018 with growth rate at the lower end of our range in the first half improving towards the upper end of our range by the end of the year, as our planned improvements are implemented and our sales force efforts accelerate.

Gross margins came in at 77% in the fourth quarter, broadly in line with last quarter, down 200 basis points from the fourth quarter of 2016, reflecting our increased investments in research. Vast majority of our cost of revenue is related to our research operations, which are now broader and more effective than at any time in the past. The related improvement in data and product quality is certainly producing the strong CoStar Suite sales growth we've been experiencing.

Operating expenses for the fourth quarter of $145 million were unfavorable to our forecast, primarily due to the higher commission expenses related to our outstanding fourth quarter sales performance. This was noted in the 8-K that we filed on January 17, 2018.

In addition to the higher commission cost, we increased our marketing and other sales-related cost, as Andy described, late in the fourth quarter, following the Xceligent bankruptcy in order to reach customers that were suddenly without information solutions. Although unplanned, we expect these investments in the fourth quarter and early in the first quarter of 2018 to generate significant returns in revenue growth this year and beyond.

Finally, our costs associated with the Xceligent litigation and bankruptcy activities were $4 million in the fourth quarter, bringing our total spend on this matter to approximately $13 million for the year, in line with our expectations.

Our fourth quarter adjusted EBITDA of $78 million was approximately $9 million, below the midpoint of our guidance range, due to the higher commissions, CoStar marketing and selling expenses just mentioned. The resulting adjusted EBITDA margin came in at 31%.

Our EBITDA of $237 million for the full year of 2017 represents a 10% increase versus the full year of 2016, while adjusted EBITDA of $280 million increased 9% versus 2016.

Net income for the full year of 2017 was $123 million, 44% ahead of the prior year and $9 million higher than we had expected. This $9 million of additional net income is primarily the result of lower income tax expenses, along with the reduction in interest expense related to the recent debt restructuring.

There are a number of positive developments impacting our tax numbers this year, so let me walk you through the individual components. First, we recorded a $7 million tax benefit related to the change in accounting rules during the second quarter of 2017 for share-based payment transaction.

Second, we recently completed the study of our research and development costs over the past 5 years, and we were able to recognize $8 million of research and development tax credits in the fourth quarter.

Finally, we revalued our deferred tax liabilities in the fourth quarter to reflect the lower corporate tax rates prescribed in the new U.S. tax law that was passed in December 2017, this resulting in a $7 million benefit. Altogether, these items reduced our effective tax rate to 26% for 2017, down from the rates that were in the high 30s previous.

Non-GAAP net income for the full year of 2017 was $154 million and includes our traditional adjustments for stock-based compensation, acquisition-related expenses as well as adjustments for the writeoff of prior debt issuance assets related to our Q4 debt restructuring.

Our cash investment balances were approximately $1.2 billion as of December 2017, and we currently have no outstanding debt and maintain an undrawn revolving credit line with $750 million of capacity.

Yesterday, we used $350 million of our cash to close the ForRent acquisition, and we'll continue to evaluate investments in strategic acquisitions, in line with our growth strategy.

Now let's take a look at some of our performance metrics for the quarter. As Andy mentioned, our sales force, which totals approximately 725 sales reps at the end of 2017, delivered $43 million in net bookings in the fourth quarter of 2017, increasing 47% versus Q4 2016, an all-time high for the company. This $43 million of net bookings includes a negative $9 million in net bookings in LoopNet Information Services. You recall, we stopped selling LoopNet information products when we integrated the CoStar and LoopNet databases in October.

In the first quarter of 2018, we will discontinue the vast majority of our LoopNet information products. As a result, we expect almost $20 million of negative net bookings from LoopNet info products in the first quarter, as we terminate substantially all of our agreements with the month-to-month LoopNet information subscribers. Over time, we expect our sales team will continue to sell these customers our flagship CoStar Suite products for a net positive revenue result, but expect lower total net new sales in the first quarter of 2018 when compared to the $43 million of net bookings we delivered in the fourth quarter of 2017.

The renewal rates on annual contracts were 91.3% in the fourth quarter of 2017. These were up 90 basis points from the 90.4% in the fourth quarter of 2016 and 30 basis points above the renewal rate achieved in Q3 of 2017. The sequential increase in the renewal rate is most notably a result of improvements in our multifamily business. The renewal rate for customers who have been subscribers for 5 years or longer was 97%.

Subscription revenue on annual contracts accounts for 80% of our revenue in the quarter, up from 77% this time last year. I expect this number may temporarily decline, as we add the ForRent revenue to our metrics in the first quarter of 2018, similar to the impact of prior acquisitions in this space.

I'll now discuss our outlook for the full year and the first quarter of 2018. We expect revenue in the range of $1.17 billion to $1.19 billion for the full year of 2018, which includes partial year ForRent revenue in the range of $65 million to $75 million. Excluding the ForRent acquisition, we expect revenue for the existing organic business to be in the range of $1.105 billion to $1.12 billion, which implies an annual growth rate of 15% to 16% over 2017.

When we complete the integration of ForRent, we continue to expect full year revenue in ForRent in the range of $75 million to $80 million with eventual adjusted EBITDA margins in the range of 45% to 55%. Currently, ForRent EBITDA margins are around 15%. We believe it will take 12 to 24 months to fully integrate the business. Accordingly, we expect the acquisition to be dilutive to our expected 2018 adjusted EBITDA margin by approximately 200 basis points.

We expect revenue for the first quarter of 2018 in the range of $269 million to $272 million, which includes an assumption of $7 million to $8 million in revenue from ForRent, representing approximately 1 month of revenue.

Gross margins for 2018 are expected to remain at approximately the same level as 2017, prior to any purchase accounting amortizations from the ForRent acquisition. The amortizations related to acquired intangible assets will impact our cost of revenue, and these will likely increase. We'll be able to provide more clarity on gross margins after our Q1 earnings release when we complete the purchase accounting for the acquisition.

We expect total marketing costs to increase approximately 5% in 2018, prior to the ForRent acquisition, which is the first increase in our marketing spend level since 2015. The focus of our spend will shift towards increasing brand reach in our multifamily business, as Andy mentioned. As in prior years, the advertising spend is more heavily weighted in the first half, with the second quarter expected to be our largest marketing quarter, with as much as 40% of our annual marketing budget expended in Q2. As a result, we expect the second quarter to be the low point for adjusted EBITDA margins for the year, as was the case in 2017.

In 2018, we expect to continue to invest in growth initiatives, while, at the same time, growing our adjusted EBITDA margins. Our investments are focused on taking advantage of these unique growth opportunities that Andy mentioned in both our CoStar and multifamily businesses. For CoStar, we continue to build our product capabilities by expanding our field research, improving our tenant products, building on our news content. We'll also add to our commercial real estate sales force to improve both the reach and market coverage of that team.

In addition, we're expanding our research capabilities in London. For multifamily, we're investing in new software capabilities to benefit renters in Apartments.com, and we're expanding our sales capabilities, primarily through inside sales. Additional investment initiatives are planned that will capitalize on these unique market opportunities. Overall, we expect approximately $25 million of spending in 2018 against these initiatives.

Additionally, we will continue to incur legal fees related to the wind-down of the Xceligent litigation matters in order to protect our stolen data through the bankruptcy process as well as to conclude the open litigations that were ongoing in both India and the Philippines against the subcontractors. We've assumed approximately $4 million for ongoing Xceligent-related fees in the first quarter of 2018 with these costs diminishing in the second quarter and beyond.

Considering these investment initiatives and before the impact of ForRent, we expect our adjusted EBITDA margin to increase by around 350 basis points over 2017's adjusted EBITDA margin. This equates to full year 2018 adjusted EBITDA margins of around 33%, at the midpoint of our range. Adjusted EBITDA margins for the year, including the ForRent acquisition, is approximately 31%, at the midpoint of our guidance range. We expect the adjusted EBITDA in a range of $365 million to $375 million for the full year of 2018, which includes approximately $5 million to $7 million of adjusted EBITDA associated with ForRent. Regardless of the near-term margin dilution associated with the ForRent acquisition, we remain confident that we'll achieve our goal of the 40% margin exiting 2018.

In the first quarter, we expect adjusted EBITDA in the range of $70 million to $74 million, which includes the negative impact of approximately $3 million from ForRent. The ForRent results are negative as a result of typical purchase accounting adjustments.

In addition to the dilution from ForRent, as in prior years, the first quarter expenses include seasonally higher costs related to payroll taxes, our annual sales conference and annual standard increases for personnel. Our marketing costs also increased in Q1, as we increased spending ahead of the apartments rental season.

Finally, we're discontinuing the LoopNet Information Services in February, which has a negative impact to both revenue and EBITDA, again, that we expect we will recoup this as we continue to upsell these clients to CoStar over the coming months.

We expect 2018 non-GAAP net income per diluted share in the range of $7.01 to $7.21 based on 36.5 million shares. For the first quarter, we expect non-GAAP net income per diluted share in the range of $1.32 to $1.40 based on 36.3 million shares. These ranges include a revised non-GAAP tax rate of 25%, which is well below our previous 38% non-GAAP tax rate. This is primarily the result of the tax law changes enacted under the Tax Cuts and Jobs Act passed in December 2017. This rate reduction has the effect of increasing our non-GAAP net income by approximately $44 million or $1.22 per diluted shares, at the midpoint of our guidance range.

In addition to the tax changes, CoStar will conform to the new accounting standard for revenue recognition in 2018, known as ASC 606, beginning in the first quarter. Impacts to revenue expenses are not material and they are included in the 2018 outlook.

Overall, I believe we are well positioned in 2018 to continue the acceleration of our revenue growth rates, while managing cost and investment to continue expanding our margins. I look forward to updating you on our progress as it goes throughout the year.

With that, I will now open the call for questions.

(Operator Instructions) Your first question comes from the line of George Tong from Goldman Sachs.

I'd like to dig into your margin outlook. You're continuing to target 40% EBITDA margins exiting 2018, but you're elevating your investment spending in the first several quarters of the year. Out of the categories of investment spending that you've outlined, how much of that investment do you expect to be temporary in nature, thus giving you confidence that you can actually hit your 40% target?

Yes. George, this is Scott. When you look at what we're investing in, I listed a number of things: expansion of our sales force; improvements in research; to continue expanding a little bit internationally. All those things are intended to go after these revenue growth targets that we have and also increase the revenue pace as we go forward. So what we'd rather do now as we go forward, unlike 2017, where we made significant investments and didn't grow our margins, as we go forward, we want to keep pacing in these smaller investments as we go and keep raising the margins, like we're doing now, the 350 basis points. So we balance the investing we need to fuel growth in the future with the margin improvements that we need to make to both hit our commitments and to give the returns that we committed to. So I don't think I would call any of them temporary. The litigation costs we had in Xceligent, obviously, were temporary and we'll see those moderate as we go through the year. But the rest of these things are worthwhile investments for the business.

But things like integration of ForRent are temporary or the roadshows are temporary or the unusually high commission costs are temporary or the $0.5 million of chocolate baskets for Xceligent customers are temporary. iWatches are temporary. There's a lot of temporary stuff in there that is spent in the first half.

And definitely in the first part of the quarter, we had a good, at least, $5 million of those types of spendings.

Man, were the kids impressed when I said I bought $0.5 million of chocolate over the weekend. I won't do that ever again.

Your next question comes from the line of Peter Christiansen from Citigroup.

Super helpful commentary here. I know that you haven't given us margins by segment in the past. And I was just wondering if you could just at least give us a sense of what the margin trajectory has been like, I guess, the last 12, 18 months in the multifamily side. And as you scale that, cover your fixed cost, how has that progressed generally?

I'll take a rough shot there. We aren't describing -- we're not reporting it at that level. However, I can give you a feel for it. I did a simple chart the other day of investments being big, red negative bar charts and then 4-wall profit being green. And basically, we went through -- in '15 and '16, we -- '15, we went through a very significant year of investment as we ramped up the nationwide brand advertising. Sales kicked up in '16. Those investments basically dropped. Net investment dropped in half, as revenues offset. And we were -- I was very pleased in '17 to see clear-cut strong 4-wall profitability in Apartments.com. And then after we get through all the noise and friction of closing a major acquisition, and even without it, then you move into even stronger margin growth. So I think the margins we're seeing now, the contribution we're seeing from Apartments.com is very successful and, certainly, dramatically more margin than we ever saw from the individual acquired company. So really good results, we think, people that we're very happy with.

Your next question comes from the line of Brett Huff from Stephens, Inc.

In the -- I want you to reiterate what you said on something, the incremental amounts that you guys are making on the incremental LoopNet user. I think you said the blended rate between the non-payers and the payers was something like $50 a user per month, and you're making net $500. So just reiterate -- I want to make sure I got that right. And then how does that -- is that the data that we've gotten so far on the 5,400? And do we expect that same kind of pricing lift to continue? Or will it taper? I mean, are we getting the most juicy leads now and it should get less juicy over time in terms of pricing? Can you kind of go through that?

Sure. Yes, so you're right. It's again just like the last time, we really pushed this a couple of years ago. The majority of the folks choosing to subscribe go from heavy searching in LoopNet to subscribing to CoStar to get better information product. There are folks paying absolutely nothing to LoopNet currently. So most of them are paying nothing. Those who are paying are paying $149. The blended rate is $49. We're picking them up at about $520, $540 net, as they go into CoStar. So fantastic, 10x sort of uplift. To answer your question, are we getting the low-hanging fruit first, absolutely not. I've sat there, I've been in 20 or 30 cities in the last month and sat in a lot of focus groups watching people talk. And this, it's pretty straightforward to us. But as people sort out the differences between the different products and LoopNet as a marketing platform, CoStar as an information platform, it takes a little bit of time. Still, the #1 competition to CoStar is these folks using the -- being heavy searchers in LoopNet or formerly using Premium Searcher. With the changes in the integrated back-end, when these people go into LoopNet product today, they're now seeing all the CoStar icons showing listings that they can't see using their LoopNet account, and it's very compelling. And it will take a little bit of time. The people aren't going to immediately commit to a 1-year, 2-year, 3-year CoStar contract. They've got to actually sort out. Sales people got to -- have to meet with them. Honestly, we have a bigger opportunity than we have number of salespeople. We mitigate that. We're shifting our resource around. But it's something that takes time. And there isn't something -- there's still very significant system flow of folks who, we think, will continuously come in and upgrade to CoStar. February will be -- we are eliminating some of the standard pricing we had in December, January. We're being more rigorous in our pricing controls. But big picture, we anticipate it will flow throughout the year and into '19, and it's a very compelling proposition -- value proposition to folks. But it just isn't something where to turn a switch and people will jump over. It takes time.

Your next question comes from the line of David Ridley-Lane from Bank of America.

So LoopNet had roughly 1,700 paying customers. You had another 80,000 pretty intensive users. Given the conversions to date, you converted just 5% of the base. I'm curious what drove the decision to accelerate the conversion, particularly because you had the Xceligent bankruptcy, which is obviously a large opportunity that you need to capitalize on as well.

Okay. So I think the number of subscribers is close to 32,000 because you have to look at folks who were getting Premium Lister with a subscriber element or unlimited listers. So the numbers is not -- I think it's higher than 17,000. It's closer to 30 -- 32,000. And then you had another 50,000, 60,000 that we classify as heavy searchers on LoopNet. The reason you would accelerate -- like, so you're -- originally, we had contemplated rolling out that Premium Searcher over a 12-, 18-month time frame just to continue the rolling time period. When you look at the folks who are using Xceligent as an information system, a very, very large number were supplementing the inadequacies of Xceligent's system by using LoopNet, so it's actually one and the same thing. And the changes that occurred at Xceligent, you're working against yourself when you are providing a Premium Searcher product to someone that you're trying to convert into CoStar. So you're probably -- you're providing a low-cost, low-value product to them to prevent them from making a decision to go over to CoStar. And you plan to eliminate it anyhow and you communicate it to them prior to the Xceligent bankruptcy that you're going to eliminate it anyhow. And also, I think we have to work to a slightly higher standard now on how we treat all our customers given our increased competitive position. So we didn't want to be in any situation where we were picking and choosing which markets or which people we eliminate Premium Searcher on and have any image that we were handpicking discontinued Xceligent clients to discontinue their Premium Searcher. So we went ahead and just did across-the-board and including, in some cases, making the right choice, the difficult choice to eliminate Premium Searcher with a person who was also subscribing to CoStar currently, so it's just good customer service. But they're related. They're directly related. It's one and the same decision-making process. And by doing this, we facilitate a faster, good result across the overall, though you take a -- it requires a little bit of extra character in the first half of the year. By the way, I listened to -- we have a pretty sophisticated phone system, monitoring system. It can listen to sales pitches and you just listen to -- I listen to 100 salespeople trying to sell various products to former Xceligent folks. And you can see the person saying, "Well, I'm trying to figure out what I'm going to do," while you're watching them do 100 searches that we can LoopNet Premium Search it, like we're just working against ourselves here. So that's really why they're related.

Your next question comes from the line of Sterling Auty from JP Morgan.

So I want to go deeper into the conversions and the price. So you're netting out, call it, $500 on these conversions. Given that huge untapped base that you still have yet to convert, based on the focus groups that you've done, what do you think is realistic in terms of how -- what percentage that you can bring over and at what price point? Because I think, at the beginning of the process, you've talked about maybe doing a $200 level of functionality, maybe a $400 level of functionality and then we'll call it CoStar. Had that kind of pricing approach or strategy morphed as you've done more work in the space?

Yes, it changed a little bit. So it's a -- we initially were focusing on, as you said, the $200, $400. We're not getting a lot of pickup in the $200. People are -- if they know what they're -- if they don't know what they're doing, some of them accidentally buy the $200. But this is what they do for a living. This is their trade. This is their Bloomberg terminal. And the $400 solution is incredibly powerful and gives them an amazing information solution with tons of revenue-generating opportunities. The $200 is adequate and meets everything that LoopNet Premium Searcher was doing and a lot more. So I think it's probably 19 of the $400 to 1 of the $200. And that's just what people are opting for. And I think it's actually shifting even further. So I think it will be like sub-5%, sub-3% of the lower cost solution at the customers' behest. So -- and again, in the focus groups, it's just amazing how they're all just trying to figure out what's going on. You've got Xceligent. You've got these board systems. You've got LoopNet Premium Searcher, Premium Lister. You've got CoStar. And you would think it was pretty easy for them to sort it out. There's a lot of confusion out there. And what we're doing is we're really simplifying that decision-making process for them. And it will take 1 year or so to sort of really simplify this and get them migrated into the right products. But we're highly confident that a very substantial portion of these folks will ultimately be using CoStar for information and LoopNet for marketing. One of the things I don't think we spent a lot of time on the call that did come out loud and clear in these most recent rounds of focus groups is one of the beautiful things here is LoopNet remains viewed by our customers as an essential utility to marketing their listing to end-users. CoStar is viewed as an essential utility for -- a very valuable utility for information solutions. So as we're doing this, the marketing side of LoopNet, to me, has never looked stronger. The information side in CoStar has never looked stronger. And there's a whole bunch of painful confusion in the middle there in those information products, and that's going to take 1 year to sort out.

Your next question comes from the line of Andrew Jeffrey from SunTrust.

This is Oscar Turner on for Andrew. I was wondering if you can provide more color into the line of sight to the 40% exit rate margin target by year-end. And how should we think about the likelihood that sustained strong sales momentum leads to higher-than-expected marketing expenses through the fourth quarter?

So when you look at the progression, we expect, as we commented on, that as our marketing ramps up into the second quarter, margins seasonally go down and then they pick up into the third and then go up higher in the fourth. We've seen that same pattern in the last couple of years and our spend is concentrated around the TV and the broadcast for the apartments marketing in the summer months. So we really don't see that kind of pressure coming into the fourth quarter, unless something unusual happened. Like this year, with the bankruptcy of Xceligent, we went around and spent quickly and mobilized, but at least…

That put in a few million dollars.

But there's just -- there's not enough time there to spend the whole lot. So we don't see a whole lot of…

I mean there's lots of competition, but there's no Xceligent there.

Yes, that's right. So we feel pretty confident with where we're pacing the marketing spend. And we don't think it should cause any issue, as we go towards the margin at the end of the year.

You next question comes from the line of Bill Warmington from Wells Fargo.

So a serious question for you. So on the ForRent acquisition, you mentioned you're going to be bumping the size of the sales force on the multifamily side by 50%, which is a real positive. But last time, with Apartment Finder when you had a big bump in the sales force, we ended up with a number of issues. What gives you confidence that those issues are not going to repeat this time?

Bill, what sort of -- just remind me again, when you talk about issues, what sort of things come to mind?

The -- when you started to put the sales forces together that there were a lot of salespeople on the Apartment Finder side that you were changing around the commission structure, the sales territories, and a number of them left and had to be -- either voluntarily or not, had to be replaced and that set us back...

That's a good question, Bill. So we -- we have grown the sales force over time at a prettier clip. We reorganize it from time to time, as we have an unbelievable amount of opportunity to reach new customers on both the CoStar side and the Apartment side. In the Apartment side, one of the big drivers there is that people with smaller and smaller communities are spending money advertising on Apartments.com network. So we get to all of our customers or 95% of our customers every quarter for service, but we only get to 10%, 15% of the prospects every quarter. And we want to get to all of them every quarter. And one of our goals is to try to get to everyone with 50 units or up during the course of the quarter. Having a larger sales force helps us do that. The -- without a doubt, there are -- in any acquisition, there's always cultural issues. Someone has been in a pattern doing something within a company for 15 years and there's a big change like a merger with another company. People make decisions about what they want to do, and that's healthy and that's good. Some significant number will find the opportunity with Apartments.com really exciting. They'll think being on the clear-cut winning team is exciting. There'll be, without a doubt, more earnings potential for them with the Apartments.com network. And some people may have been -- alternately, some folks will decide that the last 20 years at ForRent had been great and they're going to do something else or hang it up now. They're both fine. And we do anticipate and have planned for some number of people to decide to do something different. So that number will come down a little bit during the year. Eventually, we'll rebuild it at a slow pace. But net-net, you're going to end up with a great infusion of talented people that decide to make this their long-term hold -- home. And it won't be the whole number that you start with, but it will be a good number. And they'll also bring a lot of knowledge and culture and that's valuable to us, the rev-up on everybody. So we're sort of copacetic with change drives decisions in people's career. And either way, we're really glad to have the opportunity. So I don't know. I hope -- I don't to be too mellow about it, but that's what I think.

Your next question comes from the line of Patrick Walravens from JMP.

So a question for you, Andy, I think, which is how long do you think it will be until you're ready to make your next major acquisition?

31 days, 6 hours and 2 minutes. Why do you ask? We -- there's a lot out there, as you know, right? So we're -- this is a pretty big one and we have to be reminded, right after closing ForRent, we were -- we're spending a lot of time and effort continuing to look at all kinds of opportunities out there. There is a broad field. There's a lot going on. I do think we'll spend a little bit of time just focusing on ForRent and making sure we do that right. You have to respect -- when you close one of the largest acquisitions by revenue and headcount, you have to respect that and make sure you do it right and not get too confident. But no shortage of stuff and so part of the bigger problem is selecting among an embarrassment of riches of opportunities to pursue. We are selective. We're selective on valuation, so we won't rush. But again, if you were betting against CoStar continuing to do acquisitions, you wouldn't be a very good better.

Your next question comes from the line of Stephen Sheldon from William Blair.

Stephen, we're really excited to have you on the phone.

So yes, you talked about seeing research cost starting to trend down some this year. So a few questions on that. First, is that commentary on an absolute basis or as a percentage of revenue? And second, as we look out over the next few years, how much leverage would you expect to see in your research budget? I mean, do you need to add much more headcount to research at this point? Or just given the database integration and the improvements in data quality and automation from products, like Listing Manager, could you keep your research headcount relatively steady over the next few years?

I think it's a thing that changes, and research is a -- as a percentage of revenue, it's a pretty modest investment. And it is a huge competitive advantage in the market. It's the one thing that people continuously underestimate when they try to compete with us. As Xceligent failed, they joined a long list of folks who have invested well over $1 billion in competition with us, and haven't gone anywhere. The #1 reason is they just don't invest in doing the research. So it's been a long-term competitive advantage. But things are changing pretty rapidly. We have so many people in our network now that a lot -- the things are changing. A lot of people have a long-term relationship with us and want to enter the data electronically, and they do it well. We are conservative about it. We don't release this new ability to edit your listings in CoStar and then, overnight, make dramatic changes to our research process. We are studying it, discussing it. The focus groups were helpful in understanding that. And only when we know we're not going to harm product quality, we make decisions to actually shift resources. There's continuously new demands for different kinds of research. But my general sense is that, as the year moves on, we're going to find that some significant portion of our traditional workload becomes automated. And that will allow us to keep the same headcount over time and not really grow headcount, so we can staff new initiatives without adding people, but that is sort of in the gut feeling category done from an extremely conservative operational point of view.

And at this time, there are no further questions.

Well, thank you very much for joining us on the call. And we look forward to updating you on our progress towards our 40% adjusted EBITDA margin goal in the Q4 of 2018, as we promised you in April of 2014.

Ladies and gentlemen, that does conclude your conference for today. Thank you for your participation and for using AT&T Executive TeleConference. You may now disconnect.