Ladies and gentlemen, thank you for standing by, and welcome to the CoStar Group third quarter earnings conference call. At this time, all participants are in a listen-only mode. Later we will conduct a question-and-answer session with instructions being given at that time. If you should require any assistance during today's call, please press star and then zero. As a reminder, today's conference is being recorded. I would now like to turn the conference over to your host, Richard Simonelli. Please go ahead, sir.
Thank you, operator. Good morning, everyone. Welcome to the CoStar Group's third quarter 2015 conference call. Thank you for joining us. Before I turn the call over to Andy, I have some items for you to consider. Certain portions of this discussion 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 October 28, 2015, press release on our third-quarter results, and in our filings with the SEC. All forward-looking statements are based on information available to CoStar on the date of this call. CoStar assumes no obligation to update these statements, whether as a result of new information, future events, or otherwise. Given last call, we anticipate that no executives will actually cry on this call.
As a reminder, today's call is being broadcast live and in color over the internet on www.costargroup.com, where you can also find our CoStar investor relations page. A replay will be available. You'll be able to access that at 1-800-475-6701 within the U.S. or Canada, and 320-365-3844 outside the U.S. The access code is 370580. It'll be available about an hour after the call today and will be online for about a month. I'd like to now turn the call over to Andy Florance. Andy?
Thank you, Rich. Good morning, and thank you for joining us today for our third-quarter earnings call. We are reporting strong results and growing forward momentum in the business. We had exceptional sales performance in the third quarter. Our revenue in the quarter increased 24% year-over-year to $189 million in the third quarter of 2015, compared to $153 million in the third quarter of 2014. Net new sales on annual subscription contracts were $31 million for the third quarter of 2015, a 102% increase over the third quarter 2014. A doubling. Annual subscription contracts in the third quarter of 2015 increased nearly 22% sequentially from the second quarter of 2015. Through the three quarters of 2015, we have added over $85 million in net bookings. We had more bookings in the second and third quarter of 2015 than the entire four quarters of 2014.
While those numbers are very impressive, they understate the sales team's productivity. During the past two quarters, the finder sales team signed an additional $30 million-plus of contracts, converting clients from print to pure digital apartment advertising business. That was not net new revenue, so it would not be in the $85 million I mentioned. When those 110-strong finder sales force comes back into focusing on pure net new, having accomplished their conversion task, I expect to see a tailwind behind our booking numbers, our already strong booking numbers. With strong sales in both core CoStar services and in the apartments products, we are very pleased with this growing sales momentum. Excluding the impact of Apartment Finder, our business grew organically at a very strong 4.2% sequentially in the third quarter over the second quarter. That is up from the Q2 over Q1 sequential quarterly growth rate of 3.2%.
In those numbers, you can see our recent booking success beginning to translate into recognized revenue acceleration. Our annual subscription business continues to enjoy a high trailing 12-month renewal rate of 90%. Our core CoStar information renewal rates remain very steady at an extraordinary 94% plus. While we continue to move overall post-acquisition renewal rates up significantly for both LoopNet and apartments, they have historically been lower renewal rates, and when you blend those with the CoStar high renewal rates, it brings the overall rate down slightly to 90%, but it's all moving in the right direction. The core CoStar information business is solid, with CoStar Subscription Services achieving a sales booking increase in the third quarter of 2015 of 22% year-over-year. We continue to add to the quality of our service offering with the development of new products such as CoStar Market Analytics.
We designed CoStar Market Analytics to meet the needs of property management companies, apartment owners, and lenders who need access to comprehensive daily rent and occupancy information. Owners and management companies need this sort of information so that they can set and optimize their rents and occupancy levels. Small movements in rents and occupancy levels can be leveraged into very large movements in value. With cap rates at historical lows, this leverage can be magnified even more. For example, with a fairly typical portfolio of 1,500 apartment units averaging $2,000-a-month rents, can push its rental rates just $100 a month and hold occupancy, it can create $35 million in value. Conversely, if it pushes rents a little too far beyond the market, it can quickly lose 5% of occupancy a month and destroy $35 million in value.
Even apartment owners using yield optimizers need comp data to manage and audit the yield maximizer's pricing. We do not believe any other information provided in the market delivers adequate, comprehensive, or certainly not timely rent data. Our competitors typically have research departments, a small fraction of the size of CoStar's, and can only deliver 2 to 6-month-old data on one-quarter as many properties as we do. That sort of weak data is useful as a starting point, it's not good enough to use in underwriting loans and definitely not for setting pricing. Many property management companies spend a lot of money on in-house research, shopping their competitors in order to get the current rents they need to do the job right. An on-site leasing consultant might spend a few unpleasant hours one day a week secretly shopping their competitors.
For each call they make, they're likely getting 1.5 calls shopping them right back. We believe that the community might be spending $3,000-$10,000 annually per property shopping competitors and being shopped. This could aggregate up to $1 billion plus in annual industry spend of property companies shopping each other and not very efficiently or effectively. Worse yet, while these leasing consultants are busy secret shopping one another, they're not answering the phone when the real renters call them. We have made or monitored approximately 10 million calls to leasing communities this year, and 76% of the phone calls to the leasing offices during business hours go unanswered. Given that these communities are spending $billions annually trying to drive calls in their leasing offices, we believe the potential loss of wasted advertising leads is in the $billions.
We have designed more than half a dozen processes to collect rents and property information. We have a huge advantage in that we have a large digital flow of rent data into apartments.com and Apartment Finder. Many of the investments we are making into people collecting comprehensive pricing data is the same content we are using to make Apartments.com so successful. We're having multiple revenue streams off the same data expense and creating a competitive efficiency. For the first time ever, multifamily owners, investors, lenders, and property managers have the real-time rent and sales comp and a lot of the information they need to optimize their income. The CoStar Market Analytics service is very closely related to apartments.com and Apartment Finder, and that they're all key pieces in the rent maximization puzzle.
We plan to sell the products together and gain competitive advantage over those selling just a partial solution. We'll pause for there to let President Obama go by. Early adopters are giving us very positive feedback. We intend to put an emphasis on this service. Last week, we took steps to expand the pool of the sales team selling CoStar Market Analytics by 100-plus sales representatives. We have sold nearly $12 million of net new sales of CoStar Market Analytics since February of this year, and we have a robust pipeline of deals pending. Of these CoStar Market Analytics deals, many of the contracts included packages with our Apartments network, adding an additional $10 million in net new sales to these contracts, making the total related sales $22 million in subscriptions in the first eight months.
Companies like Weidner Apartment Homes, HSL Properties, National property manager Bozzuto, Hunt Mortgage, and many others are now using CoStar Market Analytics. Bozzuto is a good example of a major advertiser deciding to expand its relationship with us by increasing their marketing spend in the same contract they used to add CoStar Market Analytics. It was only in February of this year that we launched the new apartments.com site, which leveraged our technology and experience to create a better way for renters to search for an apartment online. In March, we began the first-ever significant business consumer apartment marketing campaign in the industry that not only reached tens of millions of potential renters, but demonstrated to the multifamily property managers and owners that we were serious about building and maintaining the premier go-to site on the Internet for the multifamily industry.
Since the beginning of the campaign in March through the end of September, we generated over 6.8 billion impressions, with the vast majority coming in the key age demographic of 18 to 49 years old. Most of you have seen the television campaign featuring Jeff Goldblum. It was responsible for 2 billion impressions. We also generated over 2.3 billion impressions digitally, 2.2 billion impressions for out-of-home and radio, and other social media added another 142 million impressions. Our website traffic and consumer engagement are exceptionally strong. In September of 2015, apartments.com experienced a 74% year-over-year increase in unique visitors, according to Comscore. For the sixth month in a row, apartments.com has the most trafficked apartment listing site according to each of the major marketing authorities, Comscore, Experian, Hitwise, Alexa, and Compete. That's pretty much the triple crown plus.
In September of 2015, apartments.com was the number one apartment listing site in consumer engagement with the most page views and minutes per visit, according to Comscore, Hitwise, Alexa, and Compete. Some of the services don't actually have all those metrics. Throughout the campaign, we have conducted brand awareness surveys to measure the success of the campaign. One of the strongest indications of brand strength is measured by unaided awareness, which is based entirely on brand awareness and recall. We asked renters, or a third party asked renters, "If you were looking for an apartment on the Internet, what site would you go to?" Back in February of this year, before we launched the marketing campaign, apartments.com was at 16% unaided awareness. Way back then, Craigslist was by far the leader with approximately 38% unaided awareness. In other words, nearly four in 10 mentioned Craigslist without being prompted.
At the end of September, we took over the number one position with 40% unaided awareness for apartments.com. It's apartments.com that four in 10 mention as the place to go online to rent an apartment. Craigslist was essentially changed, and every other competitor was less than 21% unaided awareness. The perception of apartments.com as a leader surpassed key competitors and ranks in the top three based on this same study, while its strengths are being characterized by renters as smart, an ally, honest, and trustworthy. Our competitor, rent.com, responded to the apartments.com campaign by spending a significant amount of money on TV with their Legit-a-Master campaign, featuring the comedian J.B. Smoove gyring his bits and pieces on an elliptical. Yeah, that actually just happened on Earnings Call. In February, before their campaign began, rent.com's unaided awareness was approximately 20%. 20% at the beginning of the campaign.
At the end of September, after the campaign, it was still stuck right around 20%. That's half of apartments.com's 40% unaided awareness. More importantly, with the most current Alexa numbers, apartments.com has nearly 60 million monthly page views, or about three times rent.com's 17 million. I have to say that I am grateful to our ad agency, RPA, for apparently effectively investing our money. A survey of 25,000 U.S. renters published this month by J Turner Research, covering the period from June-July 2015, demonstrates how far we've moved the needle in the six months since we launched the new apartments.com website and marketing campaign. When asked which website did they use to find their last apartment, renters surveyed responded the number one place was apartments.com at 26%, followed by another great site, Apartment Finder, at 23%.
23% of the renters said they did a generic internet search. Renters saying they couldn't recall was 19%. Rent.com came in fifth place with 17% of renters saying they used that site. Apartment came in sixth place with 16%. Craigslist came in at 15%. For Rent came in 10th with 12%, which there was no ribbons awarded. When the response "I don't recall" is beating all of our competition, we're obviously pulling away from that competition a bit. Quite simply, the headline on the apartments.com marketing campaign is, It Worked. The impact on sales has been astounding. We achieved $65 million in bookings company-wide in the second and third quarters alone. When you consider the high incremental contribution margin from these sales, along with the expected high renewal rates that become an annuity over many years, we believe that return on investment is very promising.
When you look at the story at a more granular market level, the story is also very impressive. Here in Washington, when we acquired apartments.com in 2014, we had $5.1 million apartment advertising revenue in the Washington metro area. 18 months later, that revenue has grown by 59% to $8.1 million. If you were to add our apartment information revenue to that CMA, the growth would be well over 60% in 18 months. We believe we are and will continue taking a lot of business from a broader range of competitors. I believe that this year we have established apartments.com as an absolute leader. In 2015, we've invested heavily in marketing we felt necessary to properly introduce and brand a great new apartments.com. We also accelerated our search engine marketing spend on Apartment Finder to facilitate a successful migration from digital and print revenue to pure digital revenue.
We have fielded a lot of questions from investors wanting to have some guidance on our spending plans for marketing in 2016. We can now say that we expect to maintain an aggressive level of investment that we believe is required to maintain our newly established leadership position and high customer satisfaction levels. The headline is that in 2016, we plan to reduce our marketing spend below our 2015 levels. We expect to lower our total marketing spend by $20 million below our 2015 levels. Reduce by $20 million below 2015. It's more expensive to introduce a new brand than it is to maintain one. We have successfully introduced that new brand. As we move to the second year of this campaign, we believe we do not need the same level of spend to achieve brand awareness and continue the momentum.
We believe that the apartments business will move to EBITDA positive in 2016. We continue to be committed to increasing the quality and breadth of content on our apartment sites. This is expected to increase engagement and help attract even more traffic to our sites. Our market research tells us that apartment reviews are very important to renters searching on the internet. We built a marketing campaign and contest around winning Rent for Life by writing a review of your apartment on our site. We received over 155,000 reviews during the contest period. Nearly 1,700 per day. I believe that's close to 10 times the number of reviews that came into apartments.com in the prior three years total. The winner of the Rent for Life prize was Martin Huttick from Woodbridge, New Jersey, Woodbridge Township, New Jersey. I believe he was very happy.
Market research tells us that consumers love our immersive 3D virtual tours for apartments with using the Matterport technology. We now have over 37,000 of these 3D videos on the Apartments.com site. More impressive, these videos are extremely popular with renters, as we've had over 7.2 million views of these immersive apartment walkthroughs. It is important to both us and our clients to know well in advance when new apartment buildings are being built. These new buildings are a stiff competition to our clients, require a change in their leasing tactics. These new buildings are also good revenue opportunities for us, as the new apartment buildings looking to lease up have the largest advertising budgets. In order to catch as much construction activity as possible, we launched an aerial surveillance plane a few months ago that flies over cities to identify new construction.
It's a high-wing plane equipped with a state-of-the-art GeoEye ultra-high-resolution camera system with augmented reality data overlays. Of course, everyone needs one of those. We are lucky to have a few very experienced military pilots handling the flights and a team of aerial researchers operating equipment. It's been an amazing resource in identifying construction sites. We've covered 28 metro areas in the U.S., and in that time, we've added nearly 120 million sq ft of new construction. I believe that we will ultimately identify close to 1 billion sq ft of previously unknown construction by the time we cover the entire U.S. That could be millions in potential advertising revenue. We hope to fly over the country twice a year, though initially, we probably won't be able to go that quickly.
Protecting this and all of our proprietary content is important. We monitor competitors to make sure that they're not illegally stealing our content to gain unfair advantage. Unfortunately, not surprising, we often do detect theft and recently have. We filed lawsuits before and anticipate again filing one or more lawsuits alleging theft of a large volume of our proprietary data in the next month. We believe the facts are clear, and the evidence is very damning. We believe we will prevail in the court of law on this issue and expect to receive substantial redress, including injunctive relief. Let's turn from that segue to Apartment Finder. We've made tremendous progress at Apartment Finder in the five months since we closed the acquisition on June 1st. Our goal is to offer our advertisers exposure for their communities on a wide array of heavily trafficked rental sites.
Adding Apartment Finder to the family meets that goal. You will recall that we purchased the assets of Apartment Finder at approximately seven times EBITDA. Apartment Finder is an established and well-respected brand in the industry with a great team of professionals behind that brand. One of the challenges the brand faced was its substantial legacy print apartment directory business. As long as Apartment Finder had a print directory in a highly competitive Internet marketplace, it would have a proverbial boat anchor around its neck. Most of its advertising contracts provided its advertisers with exposure on the Apartment Finder website and in the print books distributed across the country. A significant risk factor for us when we did the deal was whether or not we could convert most of those clients from print to pure digital contracts. I believe that risk is now diminished almost entirely.
I am very pleased with what Finder sales leader Marcia Bollinger and her team have accomplished in just a few months. At the time of the acquisition, we had approximately $65 million of that core advertising revenue to protect through the print-to-digital conversion process. In those contracts, we had to maintain the revenue levels throughout the conversion process. After just two quarters, we've converted this print-based company into a pure digital one. We have shut down the print business and retained 95% of the contracts, representing $64 million in core advertising business. After two quarters, we are no longer printing books. I believe they printed books for probably 25 years, something like that. Maybe longer, actually. At the time of the acquisition of Apartment Finder, we anticipated that we could achieve cost synergies of approximately $20 million by discontinuing those print products and other various non-core services.
Through September, we have already achieved $15 million of annual cost synergies through elimination of the print business and cuts of approximately 100 associate staff. We believe we'll see another $5 million in various cost savings and reach the expected $20 million synergies in the short term. As we mentioned in past calls, in addition to eliminating print at Apartment Finder, we eliminated various other non-core or unprofitable revenue streams, such as social media consulting, street rack distribution services, banner ads, and promotional printing services, including cozies. In total, we eliminated $13 million of non-core and profitable revenue and anticipate elimination of $20 million in associated expense costs mentioned above. This sort of cost reduction process is always hard on those involved, but I believe that Finder is much stronger now and is ready for rejuvenated long-term growth.
Since June of this year, our combined product design teams have reimagined all of the Finder desktop and mobile search products to make them industry leaders and highly competitive. The small army of our combined software developers have worked really diligently over the past two quarters to build these new products and bring them to market as quickly as possible. In addition to building the new client-facing services, they had to rebuild just about every software system at Apartment Finder. They had to connect Finder to CoStar's apartment back-end research, billing, fulfillment, sales support, and accounting systems. I am pleased to report that at this time, it appears that we are within three weeks of completing all of those projects.
Once we complete all the software, we plan to port all the Finder content and client data into the new systems, test it very thoroughly, and then expect to go live with a new website mid-December. I'm amazed and very impressed by what our software and product teams keep accomplishing. I was CoStar's first original software developer, no longer the best, and have some appreciation for the scale of the task this team is accomplishing here, and it's very amazing. Going forward, all of this will integrate the back ends of apartments.com, CoStar, and Apartment Finder, thereby leveraging the same research system, support, and sales platform to power our entire apartments network, consisting of apartments.com, Apartment Home Living, and Apartment Finder, as well as valuable CoStar information analytics.
This significantly leverages our investment sales, support, product accounting, technical, and research teams, since their efforts will be monetized across yet another major marketing platform. Once complete, all of our Apartment Finder and original Apartments.com sales staff will sell the very same compelling network of Apartments.com sites. I believe that only in combination can these two sales forces give us a sales team with the coverage, scale, and experience we need to reach all the many potential clients we have for our industry-leading products. The Apartment Finder acquisition has been a great success to date and brings a great new team to CoStar Group, and I believe it will continue to help us transform our apartment listing services. The LoopNet marketplace remains vibrant as we reached 10 million registered LoopNet members. When we signed the deal to acquire LoopNet, it had $78 million in revenue.
Since then, we've grown it and reduced costs, and it now has $78 million in EBITDA instead of $78 million in revenue. As my great-grandmother always used to say, "Buy them for revenue and convert it all to EBITDA." She didn't really. We should finish this year with about $150 million in LoopNet revenue, we're approaching doubling the top line since the acquisition. We got some great businesses in LoopNet beyond just the commercial real estate component, such as our land business and our businesses for sale on the internet, both of them showing great promise. At the time we closed the deal, all 40,000 LoopNet advertising clients were on month-to-month contracts. Today, 33,000 are on annual contracts, 7,000 on quarterly contracts, and 18,600 are on month-to-month contracts.
This migration to longer-term contracts has significantly lowered the churn and cost of sales in this business and given us more visibility in the revenue stream. We have plans to make substantial upgrades to the LoopNet website in the middle of the first quarter of 2016. We expect to incorporate some of the successful features we have deployed on Apartments.com, such as an Apartments-like map view, larger main listings, images, and a photo carousel. We want to make it even easier for our premium listers to manage their listings, upgrade their plans, and allow them to easily buy tiered advertising. We have seen some solid early successes from our differentiated advertising offering. Similar to what we offer at Apartments.com, this enables brokers and owners in office, industrial, and retail to pay more for LoopNet premium lister in order to move up in a relevant search result with a larger ad.
This provides clients with a need to sell or lease property an opportunity to do it more quickly using the LoopNet marketplace. Earlier this year, we began a tiered advertising structure on LoopNet offering Diamond and Platinum ads, it's given them the ability to pay more to sort higher with a large ad. We've been achieving, and for the rest of the year, next year, we continue to seek higher prices per user in an effort to reduce internal competition and cannibalization between the LoopNet information products and the CoStar information products. Over the next year or so, the two will be merging together. Higher prices are having a positive impact on the LoopNet bookings.
In the third quarter, LoopNet bookings were up 74% year-over-year, and as we've continued to increase pricing at LoopNet, we've achieved our first and second highest quarters of net new bookings in the second and third quarters of 2015, respectively. We had an excellent quarter recording our highest-ever U.K. revenue in the third quarter of 2015, with GBP 4.2 million in the third quarter. We achieved sequential quarterly growth of 4.2%. Earnings were strong in the U.K. EBITDA margin for the third quarter was at 20.1%, the highest it's ever been. All these numbers are a testament to our strong leadership in the U.K. and the potential global appeal of the CoStar service offering. In Canada, we now have over 2 million in annual contracts with over 100 clients and 725 users.
Keep in mind, we just opened the Toronto office in March of 2014, the uptake has been extremely strong, and I believe the Toronto market has a revenue potential in the tens of millions. We've just opened up services in Calgary and Vancouver towards the end of the second quarter and have already landed 14 good firms. We plan to open up in Edmonton and Ottawa in late Q4 this year, followed by Montreal in the second half or later part of 2016. We believe we currently collect more inventory listings than any of the competitors in the local market brokers in Toronto, Calgary, and Vancouver. I believe we have outpaced our local competitor in revenues at this point, though they've been in the market for 18 years. The commercial real estate market continues to display great strength.
Bottom line, we're definitely in a very strong commercial real estate market. On the leasing side, absorption's up, vacancies are down, and rents are growing. The year-to-date net absorption of office, retail, logistics, and apartments was 7% higher than the same period a year earlier. The composite vacancy for the four main investment property types stands at 6.9%, 30 basis points lower than a year earlier, and year-over-year rent growth averaged 4.3%, well above inflation. Sales volume to date is 20% higher than a year earlier, which is within 3% of the peak in 2007. This near-record flow of capital to real estate has driven the CoStar Commercial Repeat-Sale Indices up 9% over the past year. That's just capital appreciation. Real estate markets in most geographies in the United States are strengthening.
For example, within the office market, 65% of the submarkets recorded a quarter-over-quarter improvement in occupancy, a market cycle high. 52% of the 54 metros now have occupancy rates above the 2006, 2007 peak. That's a very important metric and usually, we believe is a leading indicator 18 months out of good conditions. Due to the depressed energy prices, Houston has been the exception, although even there, net absorption has been slightly positive for the first three quarters of the year. The apartment sector is performing solidly, with net absorption up 4% year-to-date compared with the same period in 2014. Falling home ownership is still a key driver of demand, the growing number of senior citizen renters and higher-income renters suggests the market's shifting to more renters by choice, which could become a very key driver of future apartment demand.
Despite a 10% increase in construction deliveries year-to-date, which might be due to our plan, strong demand allowed occupancy to increase by 10 basis points from a year earlier, and rents increased by 4.8%. Apartment sales volume is up by 10%, with several cities posting average prices per unit well above the last cycle. In the office sector, net absorption of 68 million square feet year-to-date is 14% higher than one year earlier. While office completions are up 40% to 41 million square feet, that's actually a very small number, and strong demand growth has allowed the office vacancy rates to fall by 60 basis points to 11% over the past year. Because of tightening in supply, office rents hit a business cycle high of 4.3% year-over-year growth. Office sales volumes grew 30% compared with a year earlier, which is well above the 20% increase for all real estate transactions.
The story of strong demand, high occupancy, and high investment sales volumes is repeated in other real estate sectors, including retail, logistics, light industrial, hospitality, and specialty, such as bowling alleys. The broad-based strength in both fundamentals and sales has helped attract increased demand for CoStar products and services and foretells a good operating environment for the next year or so. We're happy with that. In conclusion, 2015 has been a landmark year with our very successful move into serving the immense marketing and information needs of the apartment industry. I believe that 2015 is the year that successfully expanded our TAM, our total addressable market, by $2 billion with this expansion into the apartments marketplace.
While we had to invest and draw from other areas in our business to make this expansion such a success, I believe we are following a well-trod path of great long-term growth companies as we make these moves. Accelerated by the new apartments products, our second and third quarter 2015 sales results are tremendous. In the second half of the year, we've begun to achieve significant margin improvement, which I anticipate will continue into the full year of 2016. In fact, we anticipate our EBITDA to move upward by about $100 million in 2016, and by roughly that amount incrementally in each of the next three years overall, as we remain committed to reaching our goal of 40% margin by the end of 2018. Now that Brian Radecki is no longer here, I can actually get the costs of this business under control. That's a joke. Brian's here with us.
I'm right here, Andy.
I will now turn the call over to Scott Yinger, our very capable acting Chief Financial Officer.
Thank you, Andy. As Andy mentioned, we are very pleased with our performance in the third quarter of 2015. The service offerings we have introduced throughout the year, including CoStar Market Analytics and the relaunch of apartments.com, accompanied by the national marketing campaign, are driving strong sales results and are contributing to top-line revenue growth in 2015 and beyond. In the third quarter of 2015, the company reported $189.1 million of revenue, an increase of 23.5% compared to the third quarter of 2014. The core business continues to grow at a consistent rate, while the apartments.com revenue growth accelerated to 20% in the quarter and is expected to be in the range of 25%-30% in the fourth quarter of 2015.
Gross margin was $135.4 million in the third quarter, or 71.6% of revenue, which includes expenses in the quarter related to Apartment Finder's print and distribution services, including a $1.7 million charge to terminate the Apartment Finder print agreement. As Andy mentioned, we are no longer printing books and expect to be fully out of the print business before year-end, when all remaining books and racks are collected from the markets. We have been very aggressive in transitioning away from print at Finder, and we expect the resulting cost savings to be evident in our gross margins in 2016, beginning in Q2, after seasonally higher expenses we always see in Q1. Adjusted EBITDA was $35.5 million, or 19% of revenue for the third quarter of 2015, and non-GAAP net income in the third quarter was $17.2 million, or $0.53 per diluted share.
Net income in the third quarter of 2015 was a loss of $5.4 million. As expected, we reported an increase in all of our earnings metrics in the third quarter of 2015 versus the second quarter, as marketing expenses began to decline following the peak apartment rental season, and we realized planned cost reductions related to the Apartment Finder acquisition. Reconciliation of non-GAAP net income, EBITDA, adjusted EBITDA, and all of the non-GAAP financial measures discussed on this call to their GAAP-basis results are shown in detail, along with definitions for those terms in our press release issued yesterday and are available on our website at www.costargroup.com. Cash and investments were $391 million as of September 30, 2015. Short- and long-term debt outstanding totaled $370 million as of September 30th. As Andy noted, the integration of Apartment Finder is ahead of schedule.
We terminated all Apartment Finder social media contracts in the third quarter of 2015 and eliminated the costs associated with that business. Our conversion out of the print business is nearly complete as well. Shutdown of the print and social media platforms is expected to result in cost synergies totaling $20 million in annualized expenses by year-end, seven months from the date of acquisition and well ahead of expectations. We expect to gain additional cost synergies in 2016. Now, I'd like to give some additional color on a few metrics to highlight our strong performance in the third quarter of 2015. As of September 30, we had approximately 595 total salespeople across the company, compared to approximately 624 at the end of the prior quarter. We are actively integrating our sales resources to ensure that our field sales teams are appropriately sized and managed in each of our markets.
We've introduced consistent job titles, compensation plans, and training initiatives to create a cohesive organization across our sales force, focused on driving sales growth in all of our services. Revenue from subscription services on annual contracts was $121.4 million for the third quarter of 2015, or 64.2% of total revenue. For the trailing 12 months ended September 30, 2015, subscription revenue from annual contracts totaled $442.1 million, up 18.4% from $373.4 million for the 12-month period ended September 30, 2014, reflecting our continued success in growing these annual subscriptions faster than our non-subscription services. Just to repeat that important point, that's $442 million in annual subscriptions growing at 18%. Renewal rates for annual subscription revenue remained high during the third quarter of 2015.
The 12-month trailing renewal rate for CoStar subscription-based revenue was 90.3% in the third quarter of 2015, while the 12-month trailing renewal rate for customers who have been with us five years or longer was 96.4%, roughly in line with the prior quarter. As Andy noted, the renewal rate on CoStar information subscriptions has remained essentially stable for the past couple of years. LoopNet contracts have always historically renewed at lower rates than CoStar subscriptions, and as we continue to move larger percentages of the LoopNet customer base onto annual contracts, there's a small product mix impact on the trend of our overall renewal rate. I will now discuss our outlook for the fourth quarter and the full year of 2015. We are raising our full-year earnings guidance based on great progress integrating Apartment Finder and the fact that we're reducing costs faster than we expected.
We expect non-GAAP net income per diluted share in a range of $1.74-$1.78 for the full year of 2015, which is up $0.10 at the midpoint from the range we provided last quarter. For the fourth quarter, we expect non-GAAP net income per diluted share of approximately $0.79-$0.83, which includes approximately $1.5 million-$2 million of marketing expense or $0.03-$0.04 of non-GAAP net income per diluted share that shifted from the third quarter of 2015 to the fourth quarter of 2015. For the full year 2015, we expect revenue of approximately $709 million-$712 million.
As the integration of Apartment Finder continues, we believe we're ahead of schedule in discontinuing non-core services, and we have revised our estimates for both the in-year impact of these actions, as well as for the core Apartment Finder revenue that we expect to carry forward into 2016. Faster than expected shutdown of Apartment Finder services is contributing to the increased 2015 earnings outlook, while having a small negative impact on revenue. The updated range includes Apartment Finder 2015 revenue of approximately $39 million-$41 million. The company's strong sales in the third quarter of 2015 are expected to offset any impact on overall revenue related to the accelerated Apartment Finder integration. As Andy noted, the Apartment Finder core online advertising business is now estimated at approximately $65 million in annual revenue as we finish the transition away from print and other non-core services.
For the fourth quarter of 2015, we expect revenue of approximately $190 million-$193 million, which includes expected fourth quarter seasonality in the LoopNet marketplace. The sales results since the launch of the new apartments.com website and the start of the national consumer marketing campaign have been impressive. This business is seasonal in nature, and we may see lower sales and traffic volumes in the fourth quarter. With our clear leadership position among apartment ILS competitors, we expect sales momentum in early 2016 as apartment communities prepare for the 2016 rental season. The growth trajectory of the business and our financial goals remain unchanged from what we previously communicated. We still see the core business growing in the 11%-13% range, with potential for uplift resulting from changes to the LoopNet premium searcher transition beginning to become evident in 2017.
We still believe we can grow apartments.com in the 25%-30% range we previously communicated for 2016. At Apartment Finder, we expect revenue of approximately $65 million. We're working on our budget for 2016 and expect to give guidance ranges early next year, consistent with our practice in prior years. Hopefully, this gives investors and analysts some insight into our preliminary views of 2016. In terms of expenses, as Andy noted earlier, we currently expect to reduce our marketing expenses by approximately $20 million in 2016 while maintaining our leadership position in traffic. In addition to marketing expenses, we continue to believe that we can reduce the cost base of our combined apartments business, and we still expect the Apartment Finder acquisition to be highly accretive to the bottom line in 2016 and beyond.
We believe we can add approximately 8-10 percentage points in adjusted EBITDA margin in 2016 from our full year 2015 estimate of 17%. This puts us on a path to achieve our previously stated goal of 40% adjusted EBITDA margins exiting 2018. In summary, I'm very pleased with CoStar's financial results for the third quarter of 2015, as well as with the strong sales results we have achieved throughout the year, beginning with the introduction of CoStar Market Analytics and the relaunch of apartments.com. We're making great progress with the integration of Apartment Finder and are on track to relaunch that website, as Andy said, later this year.
We believe the current sales trends and plans for continued integration of our multifamily platforms keep us well-positioned to achieve our stated financial goals of $1 billion in revenue in 2018, exiting that year with 40% adjusted EBITDA margins, and a $1.5 billion revenue run rate at even higher margins exiting 2020. I'll now open the call to questions.
Thank you. Ladies and gentlemen, if you'd like to ask a question, please press star then one on your phone. You'll hear a tone indicating that you've been placed in queue, and you can remove yourself from queue at any time by pressing the pound key. If you're using a speakerphone, please pick up the handset before pressing the numbers. To get through all the questions today, we please ask that you limit yourself to one question and reprompt for any follow-ups. Once again, for questions, please press star then one at this time. We'll go to the line of Andre Benjamin at Goldman Sachs.
Thanks. Good morning.
Morning.
My question is in terms of the math on the amount of spending at apartments. There are the two buckets. I believe there's $75 million to $80 million in incremental marketing and branding spend. I think that was supposed to go down by $20 million next year. The rest of the cost, my math, is about $150 million or so, but that could be off. What should we be assuming that those costs do year-over-year? Would those be flat, down, or up, and any color on how much?
Yeah. Andre, this is Scott. I think what we're saying is that in total, the apartments marketing spend, we should be able to manage down $20 million year-over-year. As you said, there's various components in there. There was roughly $80 million in incremental spend that we announced around Apartments.com, and then we put some SEM spending behind Apartment Finder as well. In aggregate, we expect that we can bring the marketing budget down about $20 million year-over-year.
Thanks.
Thank you. We'll go on to the line of Bill Warmington with Wells Fargo.
Good morning, everyone.
Good morning, Bill.
I wanted to ask about the very strong annual subscription net new number at $31 million. We had been looking for something in the low twenties, and I just wanted to check, because one of the beauties of the model is, of course, you get this net new and you can just fat it out over the next four to five quarters in terms of the contribution. That works, but the only exception to that would be is if there's a component of that which is being converted from existing clients, because then you already have that revenue in your model. That's my question is, maybe give us a little detail in terms of that strong $31 million. How much of that represents net new business that was not previously revenue in some form within the revenue base?
Yeah, Bill, we net out in that number, we net out prior billing amounts. If somebody was on a month-to-month contract and now moved to an annual contract, that's netted out of that number.
In fact, it is as big and as impressive and as beautiful as you think it is.
All right. Well, that's helpful. The other, just as sort of a note, are you guys going to be dressing up for the Halloween party tomorrow? Because I heard a rumor that Andy was going as J.D. Tucker.
It's sort of rough when the Washington Post writes an article about the fact that everyone in town should drop by and see our Halloween party because it's really good. That puts a lot of pressure on us and could possibly divert us away from our business, it's really our field research team that does the really good Halloween party. Yeah, I'll be participating in the spirit of it, for sure.
All right. Thanks a lot.
Yep.
Thank you. Next we'll go up to the line of Sara Gubins with Bank of America Merrill Lynch.
Hi, thank you. Sorry if I missed this earlier, what was the core CoStar growth in the third quarter? How did that compare to the trend in the second quarter?
That's the 4.3% sequential quarterly.
4.2
4.2
total growth.
On a year-over-year basis, if you have it?
On an annualized basis, the core business in the third quarter was running in that 10-13 range.
Yeah. 11%-13% range. The core business continues to be in that range. I think absolute year-over-year was sort of the low end of that range this quarter. As Andy pointed out in his comments, sequentially, if I strip out Apartment Finder, the business was up 4.2% sequentially, just organic growth quarter-over-quarter.
The prior quarter, Q2 or Q1, would be 3.2%. It climbed from 3.2%-4.2% sequential quarterly.
Okay. Just a question on M&A. Now that you've done the Apartment Finder deal, should we rule out more brands in the apartment space, or do you think that's a likely area of focus?
On behalf of the 115 software developers who've been staying in various hotels north of Atlanta, Norcross, Georgia, for the last three months, I am not going to say we're about to do something else. They would like to have a little break after finishing this integration job. I think that this probably has changed the calculus a little bit, so that I think it would now look like we have been able to integrate in another good brand in the apartment space, get a good reaction from the customers, retain the revenue, and achieve significant cost efficiencies. As you look at other deals, they would obviously have to be priced correctly. We would want to capture the value for our capabilities ourselves, not give them to someone selling a business.
I would feel much more comfortable about our ability to do these deals and manage the risks of conversion.
Thank you.
There are opportunities, but there's nothing specific right now.
Thank you. We'll go next to the line of Andrew Jeffrey with SunTrust.
Thank you. Good morning.
Good morning, Andrew.
Rich, actually, I was hoping to hear you cry on this call, but maybe next quarter.
There's still time.
With regard-
He could ring in and we'd all cry.
True. With regard to the EBITDA growth cadence, and appreciate your color on 2016. As we think about the 2018 target of 40% exit margin, is it going to be pretty ratable, the EBITDA growth improvement or the margin improvement, sort of 2016, 2017, 2018, or would you anticipate more of a hockey stick as we get closer to the end of that three-year period?
Well, this is Scott, Andrew. I think the kind of guidance we just gave you, I think, will show you that we're making pretty steady progress next year. Exactly what the trajectory is beyond that, it's hard for me to give any detailed guidance on it. I think it's fairly steady. Quarter to quarter, it will vary based on the seasonality of the marketing and things like that. I don't think every quarter it's up and to the right by a fixed percentage. Annually, I think we gave you a pretty good pointer. I just said I thought we could be up 8 to 10% margin points next year. I think if you sort of plot that out over the next couple of years, it gets you to that 40% exiting 2018.
Thank you. Next, we'll go to the line of Sterling Auty with JPMorgan.
Thanks. Hi, guys.
Yes, sir.
You gave a number of points around kind of macro supply, et cetera. I just wanted to see if you could tie them together in terms of how do you feel where we are in terms of the cycle? You talked about the planes flying over and isolating construction, how do you see the timing of new construction coming in impacting 2016?
I think our thought is that, well, first of all, it's great news for apartments.com and for Apartment Finder to have that construction coming in because those folks tend to buy our most expensive ad packages. That's a good driver there. There's definitely concerns out there about the amount of construction on the multifamily space specifically, and with incomes being relatively flattened down for a lot of the demand side, will we be able to continue to see rent growth with a lot of supply coming in? Realistically, the absorption is matching the supply still, our feeling is that you still got 18 months to two years before you're really going to see any big problem develop. Things are obviously priced to perfection in a lot of areas. The office side is actually $40 million in construction. The office side is nothing.
The other sectors are pretty darn stable. One metric that I put a lot of weight on, we've seen in two prior cycles, we've seen it be a good leading indicator, is the going down to the sub-market level and counting the number of sub-markets improving versus the number of sub-markets degrading. Right now, 65% of sub-markets improving. We're still feeling that things look pretty solid. Let's call it a very rounded flat mountaintop, a plateau mountaintop. Does that answer your question at all? No. Feel free to redirect if you think I didn't answer the question properly.
No, I think that's good. Thanks, guys.
Okay.
Thank you. We'll go to the line of Mike Hong with Needham & Company.
Thanks very much. Good morning, guys.
Good morning, Mike.
Nice quarter. Let me see if I could kind of get you to give me a little color kind of inside that bookings number. I mean, great number. Were there any kind of large deals in there? Maybe you could kind of give us a flavor for kind of deal sizes and how that might be trending. Then, I'm not sure if you're prepared to answer a question like this yet, but given that you're going through your budgeting cycle now, given the strength of bookings over the last couple of quarters, are we comfortable to call that a trend now? How should we be thinking about bookings growth rates going forward? Thanks.
On the first question on the nature of the deals coming in, for sure, we are seeing a wide array of deals that are significantly larger than the deals we traditionally see. They're not like CB Richard Ellis scale deals or anything like that. They're just good, solid $10,000 a month, $20,000 a month deals at the upper end of this. The real bulk is $1,000 a month here, $1,000 a month there. That is dramatically higher than doing LoopNet deals at $395 a month. These ASPs are pretty good. There are a couple of things that are floating around out there that could move significantly above that number, but those are not really forecast pipeline or anything like that. The ability to think that we have runway on this is high.
We have been like the duck swimming smoothly across the water, the feet are churning under the water. We've been moving really quickly here, integrating these things, developing new software, switching teams around, blending teams, a lot of things happening here. We're now moving into having a little bit of the luxury to begin to refine and to optimize, and I am really struck. We just finished a round of focus groups with renters, small landlords and large landlords and property management companies in Dallas, Los Angeles, and Washington. I have to say, when I compare what those folks were saying to us two or three years ago when we were researching prior to our move into Apartments.com, and I compare that to what they're saying now, you would be giddy to hear what they were saying. We really have good products here.
The people who have CMA or CoStar Market Analytics are giving us very good feedback on it. Folks are giving us very good feedback on the Apartments.com site. Finder still is not a factor with them just because it hasn't had the new site launch. Hearing a small apartment manager who has 20 units in Washington say that in the spring, his Craigslist marketing stopped working for him, he wasn't getting leads. He started asking the millennials in his community where they were looking for apartments, and they all said Apartments.com. He's been looking at advertising there.
We're really at the beginning. When I look in all those focus groups and I see there's only two people in the randomly selected group that are CMA clients, CoStar Market Analytics clients, they're all expressing a need for the same kind of product, I think we have a long runway. As a CEO, I'm more frustrated by the fact that we have only reached 5% of the people we should reach, I can't be unhappy with the results we've got so far. Definitely it's got legs for sure, in my view.
Thank you.
Does that sound optimistic? Okay.
We'll go next to the line of Brandon Dobell with William Blair.
Thanks. Andy, just so you know, I've raised $80 million to spend on marketing, and I'm going to go after the things most likely to rent market for me, just myself, as a rental opportunity. Just me. Just so you know, you'll see that out there next year, going to be a nice competitor for you guys.
Nice competitor.
I guess I'd focus on the core for a minute, maybe some color around, Scott or Andy, your comments around that 11%-13% being sustainable, but more importantly, how you get that to drift towards the 13% as opposed to.
Sure
letting it drift down to nine or 10.
Yeah, sure. It's a good question. When we say the core, at some point in the next couple of quarters, we're going to probably have to redefine what is the core, because ultimately, the core will probably be defined as LoopNet information products, CoStar information products, LoopNet marketing products, and then CoStar advertising products being one set the way to look at apartments. You have to break out the CoStar Market Analytics targeting multifamily, maybe. Okay, the core probably will ultimately involve LoopNet and CoStar. There's no question about when we made this commitment a year and a half ago to go into apartments.com and to really try to expand our addressable market by taking a share in this relatively new area.
We definitely gave up the ability to get two or three extra points in our core business while it took your sales leadership, your software leadership, everyone to try to grow the scope of the long-term opportunity. As you can tell from this call, we're feeling that we've made very good progress in this, and there are a bunch of different ways to measure that we're now number one. You can see the ability to grow margin in it as well. I think probably one of our highest priorities in 2016 is to go back and to strengthen the relationship between the LoopNet information products and the CoStar information products, and use that integration to dramatically strengthen the quality of the product, and be able to take what is today two very large audiences who are very committed.
One audience is very committed to CoStar, one audience is very committed to LoopNet information, and bring their resources and their eyeballs and their content into one ecosystem rather than two, and then to also eliminate some of the unacceptable price irregularities in the LoopNet information business. Firms doing $150 million a year in revenue relying entirely on our information systems, but only paying us a couple thousand dollars a year. That'll be a high priority for us, and I feel, it'd be a good venture. I think we'll be able to get some additional margin improvement out of the LoopNet and the CoStar business together, obviously very profitable now. We'll be able to get some additional operating leverage there.
I think the more exciting thing is we'll be able to build higher quality products, better content, better customer satisfaction, and then we'll be able to monetize that by getting more consistent pricing between the two. Now, that's a lot of work, and we'll give you more color on it in Q2. Software-wise, it'll take us through the year. We're working on that right now, and we're excited about it and very optimistic. I would be surprised if, as we shifted some of our best resources or balanced our resources back towards the core, you wouldn't see that number come up in the next year.
Thank you. We'll go next to the line of Brett Huff at Stephens Inc. Please go ahead.
Good morning, guys. Thanks for the additional disclosures as we look into 2016. That's very helpful.
All you need is every analyst asking for it, then we are very responsive.
We appreciate it. My question is just to dive a little bit more on the core number again. I want to make sure I understood what the 22% core info net new annualized sales bookings number comprised of. First of all, is that the right number that I'm referencing?
Yeah, Brett, that's correct.
Okay. Second of all, that includes what we think of as the original database business, as well as Loop, as well as the cross sales of the new market analytics business all in there. Is that the right pieces?
The way we define that did not have Loop in it. That was basically the core information analytics. It would include the new CoStar Market Analytics that we sell too.
Okay. It's core info and then the new market analytics product. The core analytics product you said grew $10 million. In net new annualized year-over-year, Andy, you mentioned a number, and I didn't quite get it. How much of that 22% was the old business, if you will, and how much of it was selling this new market analytics piece?
The $10 million was just bookings this year on CoStar Market Analytics, two apartment-oriented businesses. It was specifically just that CoStar Market Analytics piece. The bookings in the quarter was roughly $14 million for the core business alone, just in the quarter, not the three years, about the three quarters of the year. The number would be much larger than that $10 million you're talking about.
Thank you. We'll go next to the line of Peter Lowry with JMP Securities.
Great. Thanks. Have you seen any changes in terms of strategy at RentPath since they hired their new CEO in July?
Who's RentPath?
Exactly.
No. That's a good question. We obviously watch that. We definitely observed the change of strategy on the part of their board when they put a new CEO in place. You look back at his historical activity of trying to migrate back-end systems, I believe, from AutoTrader into the front-end marketing systems. I think that's interesting. I'm not sure the same parallel exists here, because the interaction between the property manager and the owner and the accounting systems and the variety of systems involved here and how heavy they are and how switching is so difficult. You can look at RealPage, where they entered into MyNewPlace.com to try to get synergy between the back-end systems and the front-end systems.
I have a lot of respect for RealPage. I would say that they would probably agree with me say that MyNewPlace.com was decimated by that effort. I don't see them executing the same strategy. The strategy they've had year to date, I would say, is really challenging because their Apartment Guide is where all the traffic is for them and all the customer goodwill is around Apartment Guide. Rent.com, I believe, has much lower customer goodwill. I think somebody interested in URL names pushed the Rent.com moniker. The traffic is on Apartment Guide as well. They've got to switch the customer loyalty from a well-known and liked brand over many years to one that has been less liked by the industry and has less traffic. Again, those numbers on unaided awareness are real, and the traffic numbers of 3x on Apartments.com versus Rent.com are real.
I'd say they're going to have to think about some sort of reset. I take them all very seriously. They're good people. They're competent people. They're not going to just go to sleep and go away. So far, I'm still waiting to see what the effect of strategy is, waiting with bated breath. We're going to keep moving forward. I would say there might be some evidence our strategy is working well. We'll try to move to the next strategy and keep one pace ahead. I do think they probably had a short-term reprieve. I guess you noticed their debt was downgraded the other day. That might have been the source of your question. From that document, you can try to put together the fact that they're not making the numbers they thought they would make this year.
I think they had a little bit of a reprieve because we were very motivated to move swiftly through the Finder integration. We weren't going to print books for another two years in order to keep 99.999% of the customers. They probably picked up a little bit of business from us doing that transition every three months. That's not sustainable. That's a short-term event. We come through the integration much stronger, with a much stronger sales force and very motivated. I hope that answers your question. If not, circle back for another one.
Thank you. We'll go to a follow-up from the line of Andre Benjamin with Goldman Sachs.
Thanks. Had to jump back in there with the one-question rule.
Yes.
I'm wondering, with the pricing on Apartments.com and Apartment Finder, I was wondering how that's trending, both absolute and versus competitors such as you just mentioned, RentPath, ForRent.com and Zillow Group. I'm really just trying to think about how that delta's trended over time and how we should think about you closing the gap, contributing to next year's growth versus the underlying traffic and customer growth.
Sure. Well, looking at Trulia first No, teasing. Again, one of our factors in looking at what companies we were more inclined to acquire, one of the factors was we wanted to make sure we acquired a company that did not have a legacy pricing momentum from their print origins. Many, many years ago, CoStar produced print commercial real estate directories back in the early '90s. One of the things I noticed was as we switched from print to digital ads, we carried the price schedule right from print into electronic. Ultimately, the truth is you've got a less direct cost associated with the ad, so prices can drift down a little bit. We felt that there was probably some downward pricing pressure in some players in the industry.
We felt very comfortable going with apartments.com, which was middle lower part of the pack for pricing. Right now, we're comfortable with that. We feel that the pricing is one that will allow us to grow volume, and volume grows our digital data flows, and it grows a stronger customer experience. We're less interested in trying to nickel and dime each property manager as a client into getting the whole thing where literally people used to charge $25 extra a month to say pets were allowed on the ad. We're not doing that. We're trying to get more volume, so we get broader and broader participation. We build a stronger network. We like the fact that we have the wind at our back on pricing and that others have a headwind on pricing.
The other thing is that we're not just trying to make the money on an advertisement. We're making the money on selling both the pricing information, the comparable information, and the marketing solution. We can go in there and have bigger contracts overall, but much better pricing. I just like having the best product and the lowest pricing while we're trying to grow share. I think the industry has suffered from having these little 5% shares fiefdoms. The renters and even the owners would like to have a real clearinghouse here, and that's what we're trying to build. Bill, did you come in there and disappear? Well, at this point, if anyone has any other additional questions, feel free to give us a buzz this afternoon. I'm sure we'll hear from you anyhow. Thank you for joining us for the call.
Thank you, Brian, for joining us for the call. There was no crying. It was a very good call. Thank you very much, and we look forward to updating you next quarter with a lot of exciting news. Thank you.
Thank you. 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.