LendingTree, Inc. (TREE) Q2 2017 Earnings Report, Transcript and Summary

Good day, ladies and gentlemen, and welcome to the LendingTree, Inc. Second Quarter 2017 Earnings Call. [Operator Instructions] As a reminder, this conference is being recorded. I would now like to turn the conference over to your host for today, Gabe Dalporto, Chief Financial Officer. You may begin.

Good morning, and thanks to everyone for joining LendingTree's second quarter 2017 earnings conference call. As I am sure you have gathered from this morning's press release, we are especially pleased to be discussing this quarter's performance today. Sticking with the typical cadence, I will first discuss the quarter's financial results, and turn it over to Doug to provide his thoughts and speak about our revised guidance. But first, let me provide the usual disclaimer. During today's call, we may discuss LendingTree's plans, expectations, outlooks, or forecasting for future performance. Forward-looking statements are typically preceded by words such as we expect, we believe, we anticipate, or other, similar statements. These forward-looking statements are subject to risk and uncertainties, and LendingTree's actual results could differ materially from the views expressed today. Many but not all of the risks we face are described in LendingTree's periodic reports filed with the Securities and Exchange Commission. On this call we will discuss a number of non-GAAP measures. And I refer you to today's press release available on our website at investors.LendingTree.com for the comparable GAAP measures, definitions, and full reconciliations of non-GAAP measures to GAAP. With that, let's get into the results. Put simply, the Company's second quarter performance was outstanding, as we accelerated top- and bottom-line growth across the business. The results highlight the type of growth we can achieve when you couple secular tailwinds with tremendous execution across all levels of the organization. Starting from the top, consolidated revenue of $152.8 million represents year-over-year growth of 62%, and substantially exceeded the high end of our prior guidance, driven by continued sequential growth in nearly every key product throughout the quarter. Mortgage revenues once again bucked broader industry trends and should further rest any concerns about our ability to grow that piece of the business through varying interest rate environments. At $71.5 million, revenue from our mortgage products grew 28% year-over-year compared to a 9% decline in industry originations, according to the Mortgage Bankers Association. Revenues from both purchase and refinance grew materially and meaningfully from the first quarter levels as lenders continued to turn to LendingTree as a trusted source of predictable, high-quality volume. And non-mortgage revenue grew 112% year-over-year to a total of $81.3 million, comprising terrific performance across a number of key verticals. Momentum in the home equity business continued, growing $6.9 million compared to the prior year. Marking a seventh consecutive quarter of triple-digit growth, this business perfectly illustrates the flywheel concept inherent in our business model. As we continue to add new lenders to the platform and expand coverage, monetization continues to improve, enabling us to profitably extend our reach across acquisition channels to drive more end-market consumers. We believe there's plenty of runways ahead in this category. Credit cards grew to $37 million in the quarter, with year-over-year growth accelerating to approximately 43% on a pro forma basis. The team continues to execute relentlessly as we optimize the business across LendingTree and CompareCards brands. And notably, revenue from our personal loans product achieved record levels, easily surpassing our previous high from the first quarter of 2016. Consumer demand for this product remains robust, as a personal loan requests submitted through LendingTree were up 46% year on year. In terms of profitability, variable marketing margin grew to $48.3 million in the quarter. At 32% of revenue, VMM incorporates our largest investment in offline advertising in recent memory, as we continue to test new creative and educate consumers on the LendingTree value proposition. Adjusted EBITDA in the quarter grew 62% to $27 million and also substantially exceeded the high end of our prior guidance. Relative to our initial guidance for the year, I'm ecstatic that our outperformance through the first half of this year has given us the ability to deliver incremental profits for shareholders while, at the same time, accelerating investments in our business. Doug will provide more color on our revised guidance in a moment, but I wanted to highlight that we are maintaining a disciplined approach to delivering growth in both revenue and profits while maintaining healthy margins and reinvesting in the business for the long term. Looking now at GAAP results, we recorded $8 million of net income from continuing operations in the second quarter, or $0.59 per diluted share. Similar to our first quarter, our results were impacted by a charge of $9.4 million associated with the markup of the contingent consideration liability on the balance sheet in conjunction with expected earn-out payments in the CompareCards acquisition. Simply put, CompareCards continues to deliver outstanding performance, making full achievement of their earn-out more likely. We're now carrying that liability at $41 million against the $45 million total potential payable, so we would expect future charges of the same nature to be less impactful going forward. Removing the impact of that charge and other extraneous items, adjusted net income per share was $0.90. From a cash flow perspective, we generated a robust $31 million of cash flow from operations during the quarter. Lastly, looking quickly at the balance sheet, where our unrestricted cash position sits at $356 million as of June 30. Thanks to continued positive cash flows from the business and the successfully executed convertible debt offering during the quarter, we remain well-capitalized to continue evaluating strategic acquisition targets. Having now executed four acquisitions of varying sizes throughout the past year, we've been thrilled with the success we've had thus far in acquiring terrific businesses at reasonable prices, and integrating those businesses to derive meaningful synergies from the LendingTree platform. With that, I'll turn it over to Doug.

Thanks, Gabe, and thank you all for joining the call today. Our second quarter topped off an exceptionally strong first half of the year, where we continued to diversify the business, achieved record revenue and profitability, and accelerated our growth trajectory. I couldn't be more thrilled with our performance, but what's even more important to me is the way we're growing our business. It's happening through the continued acceleration of what we've always called the flywheel. Our revenue increases because LendingTree is working for more and more lenders, which improves our monetization while also being incredibly profitable for our lending partners. You see that clearly in the pace at which we grow versus the mortgage industry, and the pace of our home equity business, which is growing from almost 0. You saw it in the personal loans business; and you are still seeing it in personal loans, and we have many more examples in the years to come. In home equity, for example, where we've added 14 net new lenders to the marketplace in this quarter alone, our triple-digit growth was driven by a combination of strong volume and consumer demand, coupled with significant improvements in match rates, which improves the monetization. Six months into the year, all of our marketplace are performing exceptionally well, which is also enabling us to invest heavily in our brand and focus on the execution of our strategic plan. I'd like to walk you through the progress we've made on those initiatives, and then I'll discuss our updated guidance and future outlook. Since our Investor Day in 2016, you've heard us talk about our strategic plan to advance LendingTree's leadership position in all categories where we compete. These initiatives are: to expand into new categories, strengthen the consumer relationship, improve the consumer experience, and fully optimize our conversion funnel. I'm happy to say we're making significant headway, and this quarter's results are proof of our strategy's efficacy. In June, we announced LendingTree's acquisition of DepositAccounts.com, which allows consumers to compare rates on deposit products from thousands of financial institutions through proprietary data aggregation technology. With rates expected to rise, we felt this was a timely opportunity to enter the bank deposit space, a move we've been exploring for several years. We wanted to do this because it is countercyclical to the refinance industry originations, and it benefits in a rising rate environment. We feel this acquisition fully aligns with our strategy, with the deposit space being a natural product expansion for LendingTree. Along the same lines, we acquired MagnifyMoney, a consumer-centric media property providing unbiased information, tools, and resources to help consumers compare financial products. While this isn't necessarily an extension or expansion of our current product offering, we feel that the acquisition, coupled with the acquisition and integration of CompareCards, accelerates our product and marketing channel diversification efforts, giving us greater market presence in credit cards through content and organic search traffic. The bigger picture here is the continued execution of our M&A strategy, where we are able to identify and acquire complementary companies with reasonable valuations, creating mutual beneficial opportunities for us and the people we are acquiring. We look forward to providing you all with more detail on these businesses as we continue to expand into 2018. Our second initiative is to strengthen LendingTree's relationship with consumers, and this is where we see My LendingTree playing a major role. In the second quarter, the number of new enrollments in My LendingTree accelerated significantly, growing 128% year-over-year and 41% sequentially, now surpassing 5.7 million users with a steady user engagement. Revenue contribution from My LendingTree continues to improve, growing 60% year-over-year and 36% in this quarter alone. We have also added home valuations and home equity estimates for the My LendingTree users, providing customers with unique data points on their personal finance based on credit files, automated valuation models, and third-party data. And in the second half of the year, we are going to continue to improve monetization through continued optimization efforts, which is a direct result of our lenders and the value we're adding for consumers on that product. We continue to make progress on reimagining the consumer experience, with a primary focus on creating a fully digital mortgage experience. We're currently working on digital integration with over 50 lenders on our network, including 13 lenders who are fully integrated and another eight lenders who are in the process of becoming fully integrated. Our fourth and final strategic objective is to fully optimize the conversion funnel from ad unit to loan funding, and our optimization efforts continue to bear fruit, as you see in our results. Continuous A/B testing across our ad copy, landing pages, and form flows is yielding meaningful improvement in pull-through, monetization, and lender closings. And we're seeing significant progress with a newly formed teams [ph] solely dedicated to optimization across all channels. Before moving on to our revised outlook for the remainder of the year, I'd like to once again thank the LendingTree team for getting us here. Our execution every day and against each of these strategic pillars has been nothing short of spectacular. With two quarters of outperformance under our belts, we're once again revising our full-year guidance, providing substantial increases to both revenue and EBITDA. For the year, we're increasing revenue to $580 million to $590 million compared to our prior guidance of $535 million to $545 million, and reflecting year-over-year growth of more than 50%. Variable marketing margin we're now anticipating $190 million to $195 million, up from $180 million to $185 million. And adjusted EBITDA is now expected in the range of $103 million to $106 million, an increase from the prior range of $95 million to $99 million. Consistent with the increased full-year guidance, our third quarter outlook reflects continued substantial growth and is provided as follows: revenue of $155 million to $160 million, reflecting year-over-year growth of 64% to 69%; variable marketing margin of $51 million to $54 million; and adjusted EBITDA of $28 million to $30 million. Now you'll likely notice the implications for the fourth quarter suggest modest sequential declines from Q3 levels, but if you know our Company, this shouldn't surprise you. Every year we discuss anticipated seasonality in the fourth quarter. But as the business grows, and especially as our mortgage business becomes more purchase as opposed to refinance, we're projecting seasonality could have a more pronounced impact than in years past. And by the way, if you know our business, a shift to purchase is a very, very good thing. With all that -- so we have laid an incredibly solid foundation to achieve continued growth, and I couldn't be more proud of our results this quarter. The flywheel is spinning, as both Gabe and I have said. It's spinning perfectly. And we are optimistic about LendingTree's prospects for growth in 2017 and beyond. With that, let's open the call for questions.

[Operator Instructions] Our first question comes from Mark Mahaney, RBC Capital Markets. Your line is open.

Two questions, please, one specific and one broader. On the personal loan business, could you just go through the factors that drove that growth there? And then secondly, in terms of optimizing the conversion -- optimization kind of across your product -- traffic into leads, into revenue -- could you just talk about where you are now versus where you could be long-term? And I know that's an iterative process, and there's never an end to it, but I know you've seen a lot of improvement in conversion rates. How much -- how do we think about how much more upside there is in terms of conversion improvement over the next, call it, 2 to 3 years? Thanks, Doug.

I'm going to have Gabe -- Gabe is going to answer the personal loan question. I'm going to answer -- I'm going to talk to you about the funnel and conversion rates.

Yes, we are actually really pleased with the personal loans business performance this quarter, and actually this year. As most of you will remember last year, the industry faced a lot of headwinds. We actually managed pretty well through that, through a pretty rough cycle in the personal loans business. And LendingTree actually was able to weather that storm pretty well. As we moved into Q1 and then into Q2, we've seen more lenders coming onto the platform. We're seeing firming up of pricing, increases in revenue per lead, and continued strong consumer demand. So you'll notice that we've materially increased our lenders, we've materially increased our revenue per lead, and we've also materially increased our volume. And so we're seeing kind of a nice rebound in that business. And as we mentioned in Q1, we continue -- we expect continued growth there. The one thing to note is there is definitely seasonality in Q4. But the business is fundamentally healthy, it's fundamentally growing, and we feel like we've passed the dark days of last year.

And on your next question on the conversion funnel, we're not going to give necessarily a lot of data on this, because we consider it highly proprietary. But you -- very small changes. We actually have this in our IR deck -- make very big boost to our monetization. So we've got a team completely dedicated and focused on this, and if you -- just as a couple examples. At the end of our conversion funnel, for example in match rates, that's that percentage of people that we can match with at least one lender. In Google terms, and if you think of LendingTree as a search engine, this would be like in the early days of Google, how many people were you able to deliver a paid search result for a given search query. We basically took personal loans over the past few years from matching 50% of customers to matching 70% plus of customers. That improves monetization 20% right there. And when you do that all the way down the funnel, and you do it with matches per, and you can match somebody from two lenders to three lenders, that's another 50% monetization. So that coverage, and what you think about the long tail of being able to get increased coverage, that's basically that -- that is the same analogy for LendingTree. In terms of lender conversion rates, they are all over the map. And sometimes lenders have -- based on their strategy, have very different conversion rates. But I will tell you this: we have lenders in mortgage -- then the neat thing is, in mortgage for example, we have lenders that convert at less than 1% from a lead to a close; and we have lenders that convert north of 15% from a lead to a close. The beautiful thing is we actually can help the 1-percenters become the 15-percenters, just like Google did through helping them with technology. And we're continually doing that with our lenders. So how high can they go? I think they can go -- I mean, over time, incredibly high. Because if you can fill out a form, see the lowest rates in the country, and why the heck wouldn't you take one right there if we can actually make the process smooth and easy? So, how high can they go? I hate to say, like 5X. But it's not crazy because we see lenders that are already there. And we see the things that they're doing are not things that we can't teach to other people. It was a long answer to your questions, Mark, but hopefully you find it informative.

Okay. Thanks, Doug. Thanks, Gabe.

Our next question comes from the line of Kerry Rice from Needham & Company.

Great quarter, guys. Two questions; first on marketing. Obviously you guys do a lot of marketing. How do we think about that maybe in a long-term strategy? Is it just continuing to do what you do? Is there any shift between maybe offline and online? Any kind of update or color maybe you can provide there. And then the second question is -- you rolled out some tools to help subprime consumers. And I think that was kind of an attempt to also help drive traffic to My LendingTree. Can you talk a little bit about an update there? What you are seeing, if that's also helped lead to that strong growth in My LendingTree? Thanks.

Sure. Gabe is going to talk about marketing, which is perfect, because he's our former CMO as well. And I'll take your subprime question.

Great. Look, for us, marketing is gas: it's what fuels the business. And I'm actually thrilled. You guys probably know this from the press release; we just hired Brad Wilson to be our CMO. And he brings a tremendous amount of experience from the travel industry. It's a marketplace, it's a complete analogue; and we've been basically saying lending is travel, just 10 years less developed. And so he is kind of the perfect person to step in and take that travel playbook and put it into the lending space. That said, we would expect over time to absolutely increase our marketing spend. We want to squeeze our competition out of any profitable media placement on the planet. We think we can do that because we have better monetization and a better brand. Brand plays a really important and strategic role with LendingTree. It's one of our core competitive advantages because people know LendingTree. And because it's financial services, and trust matters, people are much more likely to engage with us and complete an application with us than they are with any other shopping site on the planet, because of the trust in the brand. So we would expect to increase brand spend over time. We think we can do this profitably, and quite profitably over time. We're going to be led by Brad or me evaluating our core positioning and really doing a whole lot of segmentation and insight gathering, and finding those positioning and messaging statements that are going to electrify our audience. And so we would expect over time to grow offline investment significantly and very profitably. If you look at MoneySuperMarket in the UK, they are public. You can download their financial statements. They spend a tremendous amount offline and extremely profitably. So that's a key piece to our strategy going forward. On the digital front, I would say I would put us up against anyone on the planet right now, and we're just getting better. We've made tremendous strides with the team. We've got an amazing talent there, and we're seeing really good marketing breakthroughs and wins. And so I feel like there is just enormous legs still on the digital side. And then finally on the SEO side, it's an area where we've been pretty underdeveloped. Magnify is a great SEO shop, and we're investing in our SEO team; and so we would expect to see, over time, some pretty good progress there. So, marketing is our gas. I think we're better positioned than anyone. And you should see a healthy blend of offline/online.

And the only thing I'd add on the marketing side is that I always sometimes can correlate the success of -- or the health of our business with how much TV advertising I'm seeing. And I talked about the increase of media and of offline. If you -- if we're spending money in offline, it means that there is a lot of demand for our business from lenders. Because offline is the most expensive channel to acquire customers, but it's also the one where you can actually scale up the most. So that actually is a very good indicator of our business. In the subprime space, this is an area that is both a passion of our Company and one that can be very profitable. We believe that consumers should all get banks competing for their business. And if you've got a bad credit score for whatever it was in your life, you should still be able to get lenders competing for your business. And you certainly shouldn't get screwed over by the check-cashing store or the guy down the street. So, we have rolled out products that are screened through us in things like debt settlement and credit settlement and credit repair. It's also a key part of our My LendingTree strategy because people can join us at low credit scores, and then we'll help them get to higher credit scores. So this is one of those great things where we can do something great for consumers, it's great for lenders, and it's profitable for LendingTree. And any time we can build a product with those three circles together, we'll do it all day long. Next question?

Our next question comes from John Campbell from Stephens. Your line is now open.

Congrats on the continued great results. Just back to the HELOC business, Doug, it sounds like there is a lot of things that are driving that outperformance. If you could maybe just give a little bit more color. Is it more driven by price? Is it match rates, better lender coverage? Just curious about the main notable driver.

Yes. So the question is about our home equity business. And again, that business has performed extremely well for us. And we've mentioned this before. It used to be a very profitable and very significant business for us in the mid-2000s era. And we're seeing that trend continue currently. So we're very excited about the business. And the reality is, it's not -- there's no single factor. We've added quite a few lenders. We've improved our coverage pretty significantly. That has helped us a lot. We've also increased volume a lot. We have really been spinning up our marketing channels, and we've seen some really good big breakthroughs there. So I wish I could say it's X factor, but it is actually kind of broad across-the-board improvement in volume and lender coverage and monetization.

Okay. And then on the lenders on the network, can you give us a rough idea of how many are there today? And then any kind of color around the mix of larger versus smaller lenders?

You mean on home equity our overall?

On home equity.

I think right now, rough numbers, it's over 80. And the good news here for us -- and I think this is a nice competitive advantage for us -- is a lot of our mortgage lenders can move into this product. So the fact that we have got hundreds and hundreds of lenders on mortgage with very deep relationships and lots of tech integration enabled us to launch this business much faster than I think anybody else could have.

Okay, that's helpful. And then on the acquisitions, any way to just broadly size up contributions this year? And then just any early thoughts about synergies as you get into next year?

Yes, so, these two acquisitions, as Doug mentioned, are -- they are strategic. They really each satisfy a different goal for the business. They are profitable. We feel like they were good and reasonable margins -- or multiples in the acquisitions. They are not going to add that materially to the back half of the year. The reason why is that they are still pretty small companies, and we're starting to bolster their management teams and invest in those businesses so they can actually scale and accelerate. So I think they are absolutely strategic to our business, but I wouldn't count on too much contribution in the back half. But we would expect that to be a really nice contribution as we go into 2018 and 2019.

Okay. And then I got to throw this in there. Any thoughts on the Bankrate and Red Venture deal? It seems like they got it at a pretty good price. Is that something you guys ever looked at, or any thoughts on what that new competitor might look like?

Yes. Here's the thing: Bankrate has obviously been on the market for a while. It's something that we knew about. And I think it's probably safe to say, if we saw the value in that merger, at $0.12 over that price, we could have had it. It's not a good fit for us because we were beating them in the marketplace, and we continue to beat them in the marketplace. But at the same time, I can tell you that Red Ventures is a really great company. It's a run by fantastic leadership. They've got a great business model. They are in Charlotte, and it's a natural extension for Red Ventures. And I think it couldn't find a better home. And I think it's pretty cool that two of the biggest fintech companies are right here in North Carolina.

Absolutely. Okay. Thanks, guys.

Our next question comes from Mike Grondahl from Northland Securities. Your line is open.

Congrats on a very strong quarter. Could you talk a little bit about what drove the acceleration at My LendingTree, up to the 5.7 million?

It's really just continued adoption. We're able to do some paid marketing on it, and because of the increased monetization of it. But it's really the increased adoption from people mostly going through our flow from the rest of our loans, from our rest of our loan products, and people sticking around longer and maintaining their subscription. So, I don't know that we'll always see that type of acceleration, but the minute we can feel comfortable about doing offline, you're going to see those numbers increase materially. There are obviously other competitors in the industry out there with free credit report products that they advertise offline. And we think with our brand -- just like our brand name is better in lending, our brand name will be better at advertising on that product. And we'll get -- as we get better monetization, then we think we will take a leadership position there as well.

Got it. And then home equity, the last 18 months, has really ramped up quickly. And you guys have commented the size of the business that that used to be for you. When do you think you can eclipse that? And what's your 3- to 5-year outlook for home equity?

Home equity is an interesting one. And I'm not going to give you -- I'm going to be vague on this, and I'm going to tell you I'm going to be vague. Home equity is a business that years ago, before 2007, was highly, highly, highly automated by the top lenders in the country using automated home valuation, drive-by appraisals. And obviously we're not seeing those types of credit standards again. However, the home equity -- to close a home equity loan, while not that easy now, is easier than a first mortgage loan. And lenders can adjust them to their own -- because it's a bank held loan, for the most part -- they can adjust their credit standards and their technology on their own. They are not dependent on the secondary market. So, we will -- the way I would think about it, without a projection, if I were doing the math, I would look at the overall originations in home equity. I'd probably start compared with originations and mortgage, and then I'd see it go from there. Because home equity is an area where we feel like we have a real competitive advantage. We've already got the brand. We've already got the lenders, they've said because they can go from mortgage to home equity. At the end of the day, though, I would always caution people to think about not to think about one product type. If we were sitting here a year and a half ago, we would have never been talking about credit cards. If we were sitting here four years ago, we would never have been talking about personal loans. Now we're talking about home equity. Guess what? They are all independent flywheels, and there's more to come.

Got it, great. Thank you.

Our next question comes from Hamed Khorsand with BWS Financial. Your line is open.

First off, I just wanted to ask you, as -- with the changeup in the acquisitions, how are you going to look at more acquisitions in the SEO space? And how does that change up any kind of risk that you might be taking, given that SEOs are looked at differently from a Google perspective?

Let me give you the history of SEO at LendingTree. LendingTree has never been, call it, a trick company. And by the way, as old-time SEO was all about tricks. So we wanted to get real customers to come to LendingTree because they really wanted to come to LendingTree. And we started that, fortunately, with TV advertising. And we grew digital as a -- very successfully -- much later, under Gabe's leadership when he got here many years ago. And then we became a digital powerhouse. We were never great at SEO because it was all about tricks. So the acquisition we made is not about that. It's about producing very, very, very high-quality content that consumers actually want, that is meaningful to them, that actually helps them; it just so happens, Google likes. So that's the way we approach SEO, is we want to create high-quality content for consumers, and if it just so happens Google likes it, and Google likes it because lots of other people like it and link to it, that's where we want to go. So I think we will, over time, build an earned media operation here with more and more people who can speak honestly and clearly in specific areas. And that's our strategy.

Okay. Is there any intention to integrate them with the My LendingTree brand and try to get them onto My LendingTree? What's the road of merging the two or three?

Sorry, merging what? I got the My LendingTree part.

Merging the different as SEO acquisitions you've made with the My LendingTree brand, or are you just going to keep them separate?

So, it will be both. There's no reason to ditch a brand when you have one, but then you can move the product. So I would assume, for example, just like we did with our credit card acquisition, we compared -- we continued to run the CompareCards brand and we migrated -- we gave the LendingTree brand to them, too, because two brands are better than one; but, boy, are we going to use the My LendingTree brand. And then we get CompareCards for free, if you will. The same thing would be true with Magnify Money. We've always wanted, at LendingTree, a high-quality content team that could create great content that is usable for consumers, like I talked about. So yes, we'll do that under the LendingTree brand. And in deposit accounts, you'd expect to see us [indiscernible]. That's the reason -- one of the key reasons we can obviously do these acquisitions so profitably and pencil them out really well, in addition to the great management team, and our leadership on doing this. And by the way, we look at a hell of a lot more than we buy. But it's because we have the LendingTree brand and we know that we can put these products on LendingTree, and it's going to add incremental revenue to the Company with no additional cost.

Okay. My last question is what would it take in the business, from your standpoint, to get EBITDA margin to 20%?

Good question. Gabe, you take that first.

Yes, look; over time, we have been able to expand EBITDA margins. We have done a really good job controlling how we scale our staffing and our cost as a business. We have definitely grown that more slowly than we have revenue and VMD. Over time we would expect those -- and we've said -- to be kind of in the mid-20s in the long-term. And we still think that's achievable. Right now, we're really leaning into growth in offline investment and things like that. So that puts a little bit of a short-term squeeze on VMM margins, which affects EBITDA. But over time, those should expand back to normal levels. And as we -- some of our businesses mature a little bit, we expect those margins to expand as well. So I think it's completely on track to our long-term goal.

And the other thing I would add to that is, like some other successful companies, like -- even like an Amazon, we could have operating margins where we want them to be, and we can almost decide, and we will always communicate it with you. I don't like -- this Company, one of our philosophies is we don't look at percentages; we look at dollars. And unlike a lot of investment-laden other companies where, for example, those -- your cost of goods sold, which for us is marketing, effectively -- variable marketing margin as a percentage, I would drive down as long as the dollars are going up. Because what that shows is that we can find more profitable ways to advertise, that we're advertising more and more offline; but more importantly that we have crazy demand from the lenders. So that margin going down but the dollars going up is actually a good thing. And it's a good thing for other leading Internet companies that have pulled this off. I actually think the same thing on the operating margin line, is that we should actually focus on dollars, not percentages, but we should always be very clear as to why. So for example, as Gabe said, I would expect operating margins going on our business to continue to improve. At the same time, we really care about how -- the after-tax cash flow that we're generating for this Company and how fast that is growing. And if we could find investment opportunities that might decrease our percentages, but they're going to increase the pace of our dollar growth, we will do that. But we will always call it out and make sure that investors understand it. And as I've said before, I view this not only as a CEO, but for me it's also a significant investor. So I'm sitting on your side of the call, too. And I'd hope you guys would have the same view, that the higher the profit growth is, the better.

Thank you.

Our next question comes from Jed Kelly with Oppenheimer. Your line is open.

A large percentage of purchase mortgages are still referred directly through realtors. Do you think you need to do continued investment to change consumer behavior for new home buyers to think about comparison-shopping when they are buying a home?

Oh, gosh. You just hit on one of the age-old challenges of LendingTree that goes back to probably 2000, when we bought a company called HomeSpace, which had a realtor referral network which we turned into RealEstate.com, which was one of the earliest home listing sites that unfortunately isn't part of our business anymore. Purchase mortgage -- because we had realtor referral network because of the problem that you just talked about. And it's not going to take a lot of investment, but expect to see some of those tools coming back. So for example, many customers come to LendingTree -- over half of our purchase customers come -- and, by the way, I've been working on this one for 20 years, so I know this one like the back of my hand -- they come without having a home identified. Those people we have to incubate. And our incubation tools -- some of which we had before, we got to regain those muscles -- will come back again, where we're able to continue to engage customers. And a very key part of that is what you said, which is keeping the realtor informed. So expect to see that through technology in keeping the realtor informed. And then the lenders now are applying a lot of those same tools. And now you've got realtor networks and other things that exist out there that we can also leverage. So I believe we're -- 18, 19 years ago, even 10 years ago, probably even two years ago, where purchase mortgage conversion rates were always inevitably low -- the fact that lenders are now so focused on it, the fact that that's where they -- that's the pond they have to go fish in. They are improving their processes, they are improving their pricing, they are improving their technology, and they are listening to us. They are listening to our advice on our end, and it's getting better. So yes, there is a bit of consumer meet. I don't want to change consumer behavior. What we actually need to do is merge in with the existing consumer behavior. Because I don't think you're going to change the consumer behavior of using a realtor.

And then just on a revenue per loan request, it looks like it was up 9% year-to-year, based on our math. Nice acceleration; anything to call out there?

There is so many different loan products in there, it's kind of hard to use that metric reliably to understand pricing. There's all kinds of mix shifts and other things. I'd say, in general, we are seeing good traction on the revenue per lead, the coverage, the match rates, matches per. And so we feel really good about our monetization overall. I would caution you of ever using that revenue divided by loan requests as a meaningful indicator, because there are just too much mix shifts going on in there.

And just to add to that, from my IAC days, and when we had a search engine -- granted it wasn't Google, it was Ask.com. And even with Google, you'll see revenue per query -- which is the measure there, which would be analogue to us -- can bounce around so much because of mix shift. So for example, you could see revenue per query -- for revenue per request go down for us because we're driving a lot more personal loan business highly profitably. And that's gone up versus our mortgage business, but our mortgage business might be going up, where revenue per request is much higher because it's a higher loan product. So there could be noise in that. But that is one number that we've actually been talking about and giving that in con -- because we can put context around it for like some other companies in the search space or in the travel space, which are the closest analogues to us.

Thank you.

Our next question comes from Michael Tarkan with Compass Point. Your line is open.

Just broadly speaking, are the results you are seeing now just more of a function of you guys taking more share from other digital players? Or is this more a function of more lenders moving online, moving to digital channels? Like, how would you frame that?

So, the way I would think of this, and there's actually -- I would refer you to a much more detailed page in our Investor Relations deck where we talk about the core market, the server market, and the expanding market. What you are seeing right now is basically what you saw in travel in 2002. And after September 11, travel companies needed to go online because the travel market was in the dumps. Lenders now need to lend. And over the past few years, as they now need to lend coming off of 2007, they are increasingly turning to digital. If they go to digital, they can do their own search advertising. But they bump up against people like LendingTree who have been doing it for 20 years and have better online brands and better monetization. Or they can come to people like LendingTree, and we can just give them the customers that they want at the price they are willing to pay at the volumes they want. Which, that's a pretty darn good deal. It's basically the same deal we have with Google when we're doing search marketing there. And it's what we do with all of our other advertising partners. We're basically an advertising partner. So you keep adding more lenders to that, and then you keep adding -- and then you keep improving conversion rates, and that's basically how it works.

That's helpful. And then just digging in a little on mortgage, can you just give us a sense for whether the growth there is coming from more pricing or more volume, maybe both?

Absolutely. So pricing is dictated by the lenders. But to your market share point, that is actually a really key indicator, and expect to see us talking about that. We look at market share in a couple ways. We look at it as our percentage share of total originations. We know what our lenders are originating, but they are closing. And we know what the originations are, because they are reported. We have seen share significantly increase of the share of originations. That roughly from 0.7%, for about 15 years in mortgage, to now a little over -- right around 2% of all the mortgage in the United States. And we could take you through each one, but I don't have them in front of me. That's really the biggest change. And that's happening because of lender shift online and because of online moving from search and other types of advertising into the comparison-shopping businesses.

Got it, okay. And then last one for me --

And then pricing, you asked about. Pricing is dictated by lenders in a bidded model based on their conversion rates, which as I said, have been improving.

Got it, okay. Last one for me. Do you have a sense for how the credit performance is of the product that you're facilitating is performing? And the reason for that is we heard Discover this morning talk about maybe shifting away from unsolicited channels. And I'm just kind of wondering if you guys have done any work on that.

It's funny; we've seen this -- we've talked about this with lenders over the last 20 years. And in many lenders over the years have done studies looking at credit performance. I could actually refer you to one lender in the home equity space, interestingly. His view -- we've never heard credit performance from LendingTree being worse than overall credit performance. And I've heard it be better many times. And I think that is really because of the filters. It's because you can, through LendingTree's technology, which goes back to what Rick Stiegler built 20 years ago, is the fact that you can match precisely the types of consumers that you want. And you can -- and obviously you are using learning in that from where your credit from the best credit losses. So I think, on average, its better.

Thank you.

Our next question comes from Brandon Dobell, William Blair. Your line is open.

The first one, if you look at the lenders on the network for the mortgage product versus lenders on for HELOC, how should we expect over the next, I don't know, year or so that -- I almost called it the matching -- to progress? So lenders that do mortgage but don't do HELOC, or vice versa, how would you expect those to converge the next year or so?

Honestly, I don't know. And I'm going to give you a big, fat I-don't-know on that one, because we're always honest with you. I would bet -- if I'm guessing, every mortgage lender can do HELOCs. And, however, the mortgage companies doing HELOCs in the past they are basically, in the most instances, selling those to banks. What we saw in the past was banks doing HELOCs at very low rates and highly automated. It's what I talked about before. So every mortgage company can do them; I don't know how many mortgage companies will. What I'm [indiscernible] is the return of highly automated home equity loans. And when you see that, then this thing is going to sing. Now, in the meantime, it's singing already because basically any mortgage lender can switch over. And these are easier loans to close, as I talked about before.

Right. Okay. And as you think about new products or even nascent products, what are the top, I don't know, two or three things that you have learned from the ramp-up and maybe the acquisition in credit cards and the ramp-up in HELOC that you can apply to those new products? And I think particularly insurance. How do those -- what kind of best practices learnings, just across the whole supply chain, do you think make the most difference?

Oh, gosh. There's so many. I think for what the specific thing that you say that -- said -- is I think our biggest learning is that we should be in everything, even if it's small. Because someday it's going to change, and we might make a change, or the industry might make a change. But we got to place some small bets because they don't cost anything to do it. And let me illustrate that for you. Home equity loans have been on our site forever. There was no home equity business, roughly, from 2007 until a couple years ago. But we didn't take home equity off the site. It was still a good little business. Auto loans are not a fast grower for us because of the interference with auto dealers. Guess what? Someday we're going to solve that. And the auto -- we're going to be sitting here talking about the auto loan business taking off. Personal loans have been on our sites since 1997. It was one of the first loan products. And we used to have one lender that we had to fax them to and they got back to you five days later, so we didn't have very good conversion rates. But personal loans have been on our site. We never had credit -- the only difference there is we never had credit cards on our site. And we didn't have credit cards on our site because LendingTree is true to our consumer model. And the credit card technology was never far enough along where consumers could actually really compare multiple options. And it's still a bit -- it's still a click-out model, but it's getting better and better every day. And there's more -- and the tools are better. So we looked at credit card for years, and we didn't do it because it took till now that the consumer model is right. The same thing is true in insurance. You can't, in the United States, like you can in other countries, get fully bindable quotes from multiple insurance companies online. So that's the LendingTree model. So to fill out a form and give you the names of four insurance companies, like, that's not too exciting. So we did a partnership with Answer Financial, and we think they have the best user experience. And we think, as the user experience comes around, the minute you can do what you can do in the UK in the US, LendingTree will be, I think, the biggest, baddest thing in insurance on the Internet.

Okay. And then final one for me, you mentioned at some point moving to offline marketing for My LendingTree. What's the thought process that you are going through around when that right time might be? Is it metric-driven? Is it feel? Is it you've got lenders telling you, okay, we're ready for it? How do we think about the process that leads into that decision to get more aggressive offline?

So, it's essentially this: we're constantly -- every time we're doing advertising, we're looking at the expected value of the revenue stream and we're -- and the revenue coming in. I could tell you right now, if LendingTree wanted to, we could sit here on the call and say that My LendingTree makes money on advertising, but it makes it over time. However, a LendingTree form, when we advertise mortgage, that makes money in a second or in a minute. So we want to continue to improve the monetization of My LendingTree and get the product precisely right before we jam on the advertising. And that's because we're pushing our teams internally to not take the easy way out to say, oh, we're just going to do a bunch of brand fluffy stuff that will pay off over a couple of years, because we're more disciplined than that. But when we do it, we're going to be the leader in it. And it's basically going to happen when the expected value gets to the point where we think we can step on the gas. We don't -- the good news is, to your second point, we don't need lenders to be ready for it. Because the leads coming off of My LendingTree will not only be exclusive, because we're just telling you the deal that can save you money; they will also convert much higher for lenders. And we're already seeing that, so they are going to be much more value to us. So back to those three circles, it's better for the consumer, it's better for the lender, and it's more money for us. And we're just going to keep plugging away until the monetization gets good, and then we're going to step on it like a NASCAR car.

Got it. Okay, thanks a lot.

Our next question comes from Rob Wildhack with Autonomous Research. Your line is open.

The guidance implies a pretty decent sequential decline in 4Q. And I know you mentioned increased seasonality, but is there anything else you can call out that may be contributing there?

This is Gabe. I can take that. No, absolutely not; it is absolutely -- all you need to do is go on to Google Trends, do a search on personal loans, do a search on credit cards, do a search on buy a home, and you'll see that Q4 is for our three large categories, very seasonal in Q4. And so the guidance in Q4 has zero to do with any sort of negative headwinds or anything like that, and 100% to do with just completely expected and normal seasonality.

And the only other thing I'd add to that, too, is media prices go up typically in Q4 around the holidays, so you get squeezed out of media. And so you end up with lower volumes. And then lenders know that, so lenders staff down for it, and that's Q4.

Got it. And then on personal loans, you mentioned more lenders on the platform. Is it marketplace lenders that are spending more here? More traditional banks? Anyone in particular you can mention?

It's everybody. And refer you to our IR deck. We got a number of names in there. The Goldman Sachs guys are recently up. I think what we're seeing in the personal loan space is just like what we saw in migration over the early days. It starts off with the early adopters, which in this -- well, the early adopters were the traditional guys years ago that I referred to, so let's leave them out. But the early adopters of the Internet were the so-called marketplace lenders. And now that same business model can move to other entities. A lot of Internet companies who are looking to create a marketplace lending platform, they switched to technology platforms. They are selling these to lenders. They are not that hard to build. And as technology gets better, more and more lenders can do these profitably. And then it doesn't become a processing question; it becomes a risk question for the lender. Do they think there's good enough risk and return based on their own credit analysis? So the beauty of this industry over time -- which we have betting on, and we've seen -- is that the cost to transact a loan transaction goes down because people get replaced by technology, and it gets smoother and smoother. As that cost go down, conversion rates go up and lender desire goes up because they are making more money with the advent of technology. It's just like what you saw in Google. It's just like what you saw in travel. And as technology is now being applied in this industry, lenders now can open this channel in the same way that we can open the TV channel, because that's profitable for us. So that's the analogy there, and we're just continuing to see it.

Got it. Thank you.

Thank you. And ladies and gentlemen, this does conclude our question-and-answer session. I would now like to turn the call back over to Doug Lebda for any further remarks.

I just want to once again thank all of our shareholders for your trust and faith in us; thank our employees for the great work they've done; thank our lenders for just continuing to improve, and for their partnership and trust; and obviously the millions of customers we serve every day. And other than that, I have nothing I can say, because I think the numbers this quarter really speak for themself. And so, thank you, and we will see you again in three months.

Ladies and gentlemen, thank you for participating in today's conference. This concludes today's program. You may all disconnect. Everyone, have a great day.