Elena Perron - Head of Investor Relations Glenn Kelman - Chief Executive Officer Chris Nielsen - Chief Financial Officer.
Tom White - D.A. Davidson & Co. Jason Helfstein - Oppenheimer & Co. Ryan McKeveny - Zelman & Associates Stephen Sheldon - William Blair & Company Carter Trent - Stephens Inc. Brent Thill - Jefferies Group LLC Brad Erickson - KeyBanc Capital Markets Thomas Champion - Cowen and Company Mark Mahaney - RBC Capital Markets.
Good day, and welcome to the Redfin Corporation Q2 2018 Earnings Call. Today's conference is being recorded. At this time, I would like to turn the conference over to Elena Perron. Please go ahead, ma'am..
Thank you, Jonathan. Good afternoon and welcome to Redfin's financial results conference call for the second quarter ended June 30, 2018. Joining me on the call today are, Glenn Kelman, our CEO; and Chris Nielsen, our CFO. You can find the press release on our website at investors.redfin.com.
Before we start, note that some of our statements on today's call are forward-looking. We believe our assumptions and expectations related to these forward-looking statements are reasonable, but our actual results may turn out to be materially different.
Please read and consider the risk factors in our SEC filings together with the content of today's call. Any forward-looking statements are based on our assumptions today and we don't undertake to update these statements in light of new information or future events.
During this call, the financial metrics, unless otherwise noted, will be presented on a GAAP basis and it includes stock-based compensation, as well as depreciation and amortization expenses. You will find reconciliations of non-GAAP measures discussed today to the most comparable GAAP measures in our earnings release.
All comparisons made in the course of this call are against the same period in the prior year unless otherwise stated. With that, let me turn the call over to Glenn..
Thanks, Elena, and hello, everyone. Redfin's market share gains again accelerated in the second quarter, and our revenues and net income exceeded the range we gave you in May.
Our second quarter revenue was $142.6 million, an increase of 36% over the second quarter of 2017, excluding new businesses revenue from our agents and our partner agents were up 30% over the second quarter of 2017.
We're now forecasting slower revenue growth for the third quarter, based on an unexpected drop in Redfin's bookings growth in the past three weeks, slowing traffic growth and a weakening real estate market. Our second quarter net income was $3.2 million, down from $4.3 million in the second quarter of 2017.
These profits are seasonal, as we still expect to run at a loss for 2018. Net income didn't improve from 2017 to 2018, because operating expenses grew 31% year-over-year, while gross profit grew 22%. Gross margin declined from 35.2% in the second quarter of 2017 to 31.7% in the second quarter this year.
Some of this is the result of investments in personal service from Redfin's buyers' agents that we don't expect to pay off until 2019. Some of it is due to growth in mortgage and title businesses that we expect to operate at margins similar to our brokerage business, but not for several years.
And some of it is because of Redfin Now, our business of buying a home on our own account and then selling it. Redfin Now grew from $2 million in revenue in the second quarter of 2017 to $9 million in 2018. But none of that was profit.
Redfin Now's margins will always be low because we account for the entire price of each home we sell as revenue, but expect then only 1% or 2% of that price is likely to be a profit, even years from now. Fortunately, Redfin Now sells homes for hundreds of thousands of dollars. So 1% or 2% would still be plenty of money.
Even without our new businesses, our core business is growing. Redfin's year-over-year gains in market share accelerated to 0.19 percentage points, up from 0.15 percentage points in the first quarter and nearly double the rate of market share gain we saw at the beginning of 2017. Redfin participated in 0.83% of all second quarter U.S. home sales.
We still have plenty of room to grow, but believe we will soon have enough of our own listings in key markets to drive more buyers to our website and ultimately to our brokerage. As we invest in proprietary ways to market and sell Redfin listings, our bet is that share will beget more share.
Featuring our own listings on our site could, for example, eventually be a major data advantage over portal sites, as we have more information about our own listings and other homes.
But for now, our traffic depends on how quickly Redfin.com loads, how listings from Redfin and other brokers are available on that site, and how accurate our recommendations and price estimates are.
Visitors to our website and mobile applications grew 18% in the second quarter of 2018, less than the first quarter's 28% growth, but still higher than other major sites. We have long anticipated slowing traffic growth, as we're now one of the top destinations for real estate search in most major U.S. cities.
The site is now ahead of us in traffic or well established. It's hard to arrest traffic from real estate portal than from other real estate sites, not only because of the portal's high brand awareness, but also due to likely improvements in their listing data access over the past year.
We still expect gains in search share to continue at a good pace, driven by our recent machine learning innovations and by our increasing market presence. Our growth over the next five years will be driven not only by our online traffic, but by the quality of our service and by the range of services we provide. This starts with repeat business.
When a customer decides to sell the home that she originally bought via brokerage, her likely to use that same brokerage is now 69% higher if she bought that home with Redfin rather than through another brokerage. For 2016, it was only 42% higher. So Redfin's advantage in customer loyalty is widening.
Our advantage in customer's likelihood to refer a friend to use our service is also growing. We measure this likelihood as a net promoter score, which takes the difference between the number of customers who love us and the number who don't, then divides that by the total number of customers who respond to the survey.
In May 2017, our net promoter score was 32% better than other brokerages, but a year later, that gap has now widened to 50%. In six surveys over three consecutive years, we've gotten a similar result.
In that time, we hired three different research firms to survey between 1,500 and 4,500 people each time who try to buy or sell a home, some with Redfin, but most with other brokerages. These surveys teach us how our market is changing, but also measure our performance where it matters most, and our customers' willingness to refer Redfin to others.
The same research we use to measure customer satisfaction also tells us our brand awareness is low but increasing. When we ask people buying or selling homes in the markets we serve to name three real estate brokerages, only 8% named Redfin, up from 4% in December 2016.
With customers that are so much happier at a price that is so much lower than our competitors. We feel confident that increasing American's awareness of our brokerage will increase our sales. Already where we've invested in mass media advertising, we've seen an increase in the folks who named Redfin as a brokerage.
The highest awareness is where we've advertised the longest and have the highest listing share, in our hometown market at Seattle at 19% and the Washington D.C. area at 13%. We just have to remember that not all of the sales from that increased awareness will happen right away.
Most of our mass media audience isn't in the market for a home the day an ad airs and those who are will often take until next year to close on a sale, spending months on the home months on the home search alone. We spent $12.3 million on advertising media in the second quarter, a 53% increase compared to the second quarter of 2017.
About half of the spending was for building mostly on television, radio and billboards. And the rest was largely a direct effort to get a sale mostly from Facebook and Google. Our direct marketing campaigns generate the majority of their sales within six months of when an ad runs and a measurable number for another six months after that.
We expect the direct marketing ads that we ran through the first half of the year to breakeven in 2018, and to make more money thereafter, but nearly all of the benefit of our second half spending will spill into next year. Our mass media ads will take even longer to make money.
Over the past three months, we completed a new analysis based on four years of testing TV ads in some markets but not others, and some markets running for only one year, and then others running every year. This analysis has convinced us that a 2015 TV campaign is still lifting 2018 sales.
We now believe the campaigns we launched in the next 12 months can have an effect that reaches 2021 and beyond. On our next earnings call, we'll present our plans to invest more in 2019 marketing, explaining at a high level of the awareness improvements we expect by the end of 2019 and the revenue gains we expect over many years.
The success of our marketing campaign depends and part on the success of our customers. If the people who see our ads, contacted Redfin agent for service, but end up buying or selling a home through another brokerage, we've wasted our money without helping our reputation much either.
This is why from the first quarter 2002 - excuse me, from the first quarter of 2018, we started investing in more personal service for our home buying customers. Homebuyers in today's market are frustrated by how few homes there are to buy, and how few of those are affordable.
Bidding wars, while declining somewhat recently, became common in 2018 across almost every Redfin market. Adjusted for household growth, the number of listings in the U.S. is 38% below the average from the past 36 years of which data is available.
The postwar explosion in new construction and transit infrastructure to accommodate baby boomers hasn't happened for millennials, who are coming of home-buying age with lower wage growth, higher credit, more congested roads and more student debt than previous generations.
This younger generation of would-be homebuyers is bewildered, many are living in their parents basement. Until the market becomes more balanced, they will take longer and be harder for these folks to find a home.
From the second quarter of 2017 to second quarter 2018, we've lowered the number of customers of Redfin buyer's agent serves by an average of 9.4%. We've asked our agents to meet customers more often, and not just during home tours, but in coffee shop strategy sessions to review the market and to agree on what it will take to win.
We've built new software alerting an agent when a customer she needs to meet wants to tour a home, and made it easier for us to broaden the online searches of customers to neighborhoods that are less competitor. Already in the second quarter, we saw the first year-over-year increase in homebuyer engagement in 34 months.
As measured by new customers likelihood to sign up for second meeting with the Redfin agent. This increase was only 1.4% in April, but it's strengthen to 2.2% in May and 4.5% in June.
The markets that have executed to at least some degree each of the major components of our strategy for delivering more personal service, only reach about 20% of our customers. But we are projecting that those markets will improve their customer success rates by about 8%.
It will probably take us until December 2018 to get most other markets onboard, as we rollout new software to make meeting customers' easier and as we train agents and managers on new tactics for our agents to engage our customers. Despite this encouraging customer response, investor should understand our challenges.
First, many Redfin agents want to meet more, not fewer buyers to improve their odds of finding just one who can afford a home. This is why adoption has been hard. And second, we still haven't seen more engagement result in more winning offers.
In the second quarter, the rate at which engaged customers submitted a winning offer declined slightly year-over-year. We nonetheless believe that an improvement in homebuyers overall success rate is a matter of time, as more engaged customers have in the past almost always led to more sales.
Meanwhile, our home selling business continue to grow very fast in the second quarter for more than a decade we sold homes on behalf of our listing customers using technology to get a higher home price than traditional brokers, while charging a lower fee.
For the last 18 months, we've also sold homes by buying the property from the owner and selling it on our own in an experimental offering called Redfin Now.
Redfin Now had outstanding growth this quarter with purchases increasing from $6 million on the first quarter to $17 million in the second quarter, and with sales increasing from $3 million to $9 million. All of the 17 homes we bought in the first quarter have now sold. In every home we have ever sold have sold for a higher price than we paid for it.
But there is a difference between the gain on a sale and profit. Once you account for the cost of our labor, renovations and capital, Redfin Now has been about breakeven.
The good news is that we expect that as we get bigger within a market, and build more software to manage the renovation process, we'll be able to spend less time and money, bringing each Redfin Now home to market. Encouraged by these results, we're now investing in Redfin Now as a long-term business.
This will involve hiring more software engineers to build online tools for meeting owners for pricing properties and preparing offers, for managing renovations and then selling the homes we buy.
We'll also start expanding beyond Southern California, with field operations led by Jason Aleem, who already has nearly eight years of experience running our Texas brokerage. The Redfin Now team will be busy. One reason, we've raised $247 million last month was to fund expanded operations for Redfin Now.
But we expect the capital buy homes will, in 2019, come from lenders who allocate money specifically for this purpose, using the homes we buy and not Redfin's bank account as collateral. Our goal will be to insulate Redfin's investors to some degree from the markets' ups and downs of actually opening - excuse me, of actually owning properties.
The premise to your Redfin investment should be to own a business that facilitate the purchase and sale of properties. In committing to Redfin now long-term, we eyed our competitors wearily as many are well run, some are willing to take risks we aren't and seeming all of them are well funded.
But we decided we can win, because we believe nobody is better at selling homes than we are. We believe that nobody in real estate is better at combining local service and technology. Newer entrants don't have the online audience we do, and pure websites don't have as much operational expertise. We believe few companies have our spending discipline.
This audience field experience and penny-pinching should let us acquire and sell homes at a lower cost, which will let us offer homeowners more money. For the person selling her home to Redfin Now that money is almost all that matters, and a war on price is the one we feel best prepared to win.
In committing to this business, we also recognized that Redfin Now may become less appealing to homeowners since the market slows. Rising rates often lower homebuyer demand, the cost of capital is thus likely to increase at the same time that buyers become more careful about buying Redfin now's listings, pinching us from both sides.
In such a market, owners would have to pay Redfin a larger premium for buying a home ourselves rather than brokering a sale to a buyer. Some owners will thus be less likely to sell their home to Redfin Now. But in a softening market, others may value even more highly the certainty of an immediate sale.
We'll seek to grow Redfin Now quickly, but as a company it doesn't matter much to us, how many homeowners will choose to sell a home through a Redfin agent or through Redfin Now. The price difference between these two offerings will move with the real estate market and the interest rates in ways we can't control or predict.
And we'll do our best to shift staff between the two businesses. As investors and analysts, you will have to remember that the revenues from Redfin Now sale and brokered sale are very different, but the profits should be much less so.
What we are sure of is that over the next five years many homeowners who want to choose between an immediate sale and a brokerage sale, and then offering both choices ourselves is how we can be the first company homeowners call, when considering a sale of their home. As we invest more in Redfin Now, we'll keep you apprised of its cost and risk.
With this decision, we're excited to offer customers a more complete solution than any other brokerage website or institutional buyer can on its own. Before turning the call over to Chris, let's address changes in the real estate market. Year-over-year U.S. home sales declined in June. We expect U.S.
home sales growth to have improved slightly in July, then to weaken again in August and September. What's striking about this change is that it seems to have been driven by dissident demand from homebuyers, not just a low supply of homes for sale. Nationwide, there were still 5% fewer homes for sale in July 2018 than in July 2017.
But in Seattle, Portland and San Jose where prices have increased the most, the percentage of homes selling in the first two weeks on the market declined in June from 61% to 52%. And the percentage of listings that dropped their prices increased from 31% to 33%.
June sales were down in these markets by double-digits and inventory was up also by double-digits. The trend is continuing in July and reports are now coming in from Washington D.C., Boston, Virginia and parts of Chicago as well, that homes there are getting harder to sell. As U.S.
home prices have increased faster than wages for 70 straight months, buyers in markets like these have finally had enough, at least for now. There are still plenty of markets where homebuyer demand is strong.
But for the first time in years, we are getting reports from managers of some market that homebuyer demand is waning, especially in some of Redfin's largest markets. In July, the percentage of U.S. homes that sold above their listed price declined year over year for the first time since March 2015.
As a result, even as we seek to expand Redfin Now to more markets, we're being very beady-eyed about which homes Redfin Now buys in the shoulder of the 2018 season. We're also watching whether a slightly more balanced market increases our homebuyers' likelihood of winning an offer.
And we're ensuring that our listing agents have good strategies in place to sell homes in a more difficult market. With that, I'll turn the call over to Chris..
real estate services, which reflects our combined brokerage and partner business; and properties, which reflects revenue from buying homes directly from homeowners and reselling them to homebuyers. Redfin Now results are now reported in the properties segment.
We believe this breakout provides investors with the level of visibility into these businesses' revenue and gross profit to be able to model each one appropriately. We also broke out inventory on the balance sheet, so you can see the dollar value of properties we own.
Along with moving Red Now to a long-term business, we increased our capital limit which includes properties we own and also those that we are committed to purchase at any one time from $25 million to $35 million. Real estate services revenue, which includes our brokerage and partner businesses, grew 30% year-over-year.
Brokerage revenue or revenue from home sales closed by our own agents was also up 30% on 27% growth in brokerage transactions. Brokerage revenue per transaction grew 2%, reflecting our continued effort to offset the mix shift of transactions from homebuyers to home sellers as well as a higher average home value per transaction.
Revenue from our partner agents grew 34% on 14% growth in partner transactions. Revenue per partner transaction increased 17%, as we continue to benefit from phasing out the $500 refund to certain partner program customers at the end of 2017. For the properties segment, we record the entire value of a home sale's revenue.
In the second quarter, the properties segment contributed $9 million of revenue, up from $2 million in the second quarter of 2017. The businesses accounted for in our other segment include mortgage, title services, and Walk Score, our neighborhood walk ability tool.
Combined, these businesses generated revenue of $2.8 million, up from $2.3 million in the second quarter of 2017. Gross profit was $45.2 million, up 22% year-over-year. Real estate services gross margin was 34.8%.
Our year-over-year decline was driven by 100 basis point increase in transaction bonuses and an 80 basis point increase in personnel costs, including stock-based compensation, due to increased headcount each as a percentage of revenue.
Our properties segment, for our properties segment, cost of revenue includes the homes' purchase price, as well as capitalized improvements, selling costs and home maintenance expenses. In prior quarters, we commented on Redfin Now's contribution to gross profit solely on the basis of the cost of the homes that we sold.
Based on the complete definition of gross profit, which takes into account operating expenses in addition to the cost of homes, the properties segment gross margin of negative 1.1% improved from a negative 2.5% in the second quarter of 2017, as the business continue to scale and our fixed operating cost continued to leverage.
We continue to invest in the businesses accounted for in our other segment, combined these businesses had a gross margin of negative 7.4%, an improvement from the negative 9.3% in the second quarter of 2017. Similar to the properties segment, we are beginning to see the benefits of greater scale result in increased cost leverage.
Having said that these businesses are still nascent, and as such, we do not expect them to contribute to gross profit for some time. Operating expenses increased 31% year-over-year and represented 30% of revenue down from 31% one year ago.
Our selling and marketing expense grew faster than revenue increasing 42% year-over-year on higher advertising spend to drive greater brand awareness. We are expecting a more elevated year-over-year growth in marketing expense in the third quarter.
Technology and development and general and administrative expenses grew slower than revenue consistent with our operating model of leveraging our fixed costs.
Technology and development increased 29% year-over-year, general and administrative expenses increased 23% year-over-year, including 7 percentage points of growth from outside services related to public company operations and Sarbanes-Oxley compliance.
Our stock-based compensation increased 81% year-over-year to 3.4% of revenue, up from 2.5% of revenue in the second quarter of 2017. Our net income including stock-based compensation and depreciation was $3.2 million compared to $4.3 million net income in the second quarter of 2017.
Diluted net income per common share with $0.04 for the second quarter of 2018 compared with $7.15 diluted net loss per common share in the second quarter of 2017. The prior year loss was primarily driven by the fair market value feature of our redeemable convertible preferred stock, which converted to common stock in connection with our IPO.
Excluding the impact of the redeemable convertible preferred stock, which is now converted to common stock, adjusted diluted net income per common share for the second quarter of 2017 with $0.06. Adjusted diluted net income per common share includes stock-based compensation. Now turning to our financial expectations for the third quarter of 2018.
Revenues expected to be between $137.1 million and $141.3 million representing year-over-year growth between 25% and 29%. We expect our properties segment to account for $10.1 million to $12.0 million of that revenue. Net income is expected to be between $1.2 million and $2.8 million compared with $10.6 million in the third quarter of 2017.
For the third quarter, we expect real estate services gross margin to be down year-over-year by about double the percentage point decline we saw in the second quarter. Typically, our real estate services revenue increases from the second quarter to the third quarter.
With real estate services revenue that is more flat with the second quarter, this will put additional pressure on those gross margins. On July 23, we closed public offerings of common stock and convertible senior notes due 2023.
We received proceeds of approximately $139 million from the stock offering and $108 million from the notes offering after accounting for underwriting discount, but not offering cost. These additional shares and the accretion of convertible debt will be reflected in our third quarter net income and earnings per share calculations.
Our guidance includes approximately $5.6 million of stock-based compensation, $2.0 million of depreciation and amortization, and $1.6 million of interest expense from the convertible notes that we issued in July.
It assumes among other things that no additional business acquisitions, investments, restructurings or legal settlements are concluded, and that there are no further revisions to stock-based compensation estimates. And with that, we'll open it up for your questions..
Thank you. [Operator Instructions] Our first question comes from Tom White with D.A. Davidson..
Great. Thanks for taking my questions. Maybe one on the guidance, so if I - on the revenue guidance - if I punch $11 million in properties is revenue into the model, it seems to kind of imply that real estate revs, you're basically slowing from kind of 30% growth in 2Q to kind of 20% growth. Seems like a steep slowdown.
Is that all macro? It just seems like you guys have a lot of momentum on the share gains. And then, just on Redfin Now, Glenn, maybe if you could talk a little bit about the rollout schedule.
And also just curious to hear your thoughts about how you got comfortable with kind of balancing, growing that business and your responsibility to the homebuyers and sellers that you represent on the brokerage side. Thanks..
Sure. So, first of all, we aren't entirely sure how much of it is the market and how much of it is us, because our guidance is based on a slowdown that only occurred in the last few weeks. It was a significant slowdown. It may be that we have a good week this week and a good week next week, and we can outperform it.
But we are seeing a significant change. My guess is that only some of it is driven by the environment. It is definitely changing. We're hearing things from our real estate agents that we haven't heard in 3 years about homebuyers stepping back from high prices.
But we've also tried to be clear in the call that we are seeing slowing traffic growth and that we want to see more progress on improved close rates as we lower loads on our real estate agents. So my guess is that we can do better, but also the market is creating some challenges.
As for Redfin Now, we got comfortable with that, because all the homes we bought had sold for a gain. So our experience so far has been very good as a holder of these properties. We feel like we've been able to liquidate them fairly quickly, which has limited our risk.
We've been able to validate that there's significant demand from homeowners to sell their home immediately. And we've also seen some synergies with our brokerage.
It was galling to hear that Redfin agents, in markets that Redfin Now currently doesn't serve, are referring some of our home-buying clients to competitors for an instant offer, just because we have people who can't move up until they get rid of their old place, and they need to do it immediately.
So I think that that coupled with the fact that many people who are contacting us about Redfin Now end up listing their home for sale through Redfin's brokerage, convinced us that this is a business that really works nicely with the brokerage..
Great, thank you..
Thank you. Our next question comes from Jason Helfstein with Oppenheimer..
Thanks.
Two questions, Glenn, can you talk about the importance of having a full-featured offering in integrated services, such as mortgage, title insurance, escrow, electronic documents for Redfin Now; and kind of what you see in kind of efficiency with and without that; and kind of any further thoughts about building that out? And secondly, I just wanted to clarify.
You think it will take probably a year to see the financial benefit from the decreased broker capacity initiatives, so kind of a year from now we start to see that in the numbers potentially? Thanks..
Let's start with Redfin Now, and mortgage, and title. It's an advantage that we have all three of those businesses. I don't think it's as large an advantage as others do.
The main thing we have to be really good at is meeting people who own homes, buying those properties for the right price, renovating those properties very quickly, and selling them very effectively. And I would say that the key capability that we need to develop is renovations.
We can't have these homes sitting on our balance sheet, because somebody is driving down to Home Depot, looking for the right faucet fixture. So being able to do that at scale for a large number of properties is going to be really important. Obviously, having mortgage and title is going to help.
But I actually think there's more synergy with our brokerage business, and mortgage and title, than with Redfin Now. That's where we have more customers who are buying homes through Redfin Agent, then asking to get the Redfin of mortgage and the Redfin of title.
As for your second question, you had asked about when we're going to get a financial gain from personal service and lower loads..
Correct..
I think we should see gross margins go up or down depending on whether that strategy is effective starting in the back half of the year. We are not going to offset the entire cost of higher loads immediately. That cost is significant. But I think we will see some improvement in gross margins, if in fact we can increase customer success rates.
And I just want to remind everyone that there are three reasons we invested in the strategy, two of which are going to affect our short-term results. One is to drive growth. For a given number of people who are trying to buy or sell a home, we should get more sales from that group.
So we should be better at driving growth if we can improve close rate, even if that close rate improvement doesn't offset an increase in the cost of labor. Second, if we can actually increase the close rate enough to offset that cost of labor, we should actually get more efficient in terms of our gross margins.
There the goal hasn't been to significantly increase gross margins. It's been to get to a point where we do no harm. It is a very strong seller's market.
It may be weakening slightly, but we expect to see more efficiency gains in terms of how we sell houses right now, than in representing buyers, because on the buy side the sales cycle is just getting longer. And then the third benefit is obvious, but just wanted to make sure everyone understands it.
We have such a high customer satisfaction rate, such a high repeat and referral rate, making more of our customers successful and taking better care of them will pay dividends long term. So I think this is the best platform for building a higher growth long-term business. But it is going to cost us and it's costing us right now.
We would have been more profitable this quarter, if we loaded up our agents with more customers. They wouldn't have had quite the same close rate, but they would have had more closes overall..
Thank you..
Thank you. Our next question comes from Ryan McKeveny with Zelman & Associates..
Hi, thanks a lot guys.
So a quick question on the gross margin, can you talk a bit about just the balance between growing the agent count productivity and the flow through to gross margin, because I think we all see this year there is heavy investment in agents and that's weighing on the margin? I think there's some debate around ultimately how much productivity matters for the long-term outlook of margin improving.
So can you give some additional thoughts around that? And I guess, tied in with that is just the hiring strategies as you're into 2H, and the first-half agent count up about 40%, 42%. Any thoughts on how that can trend in the back-half of the year and the flow through to margin? Thank you..
So your assumption is correct, that agent productivity is going to be the most powerful lever for increasing or decreasing gross margins. The other factor you should just consider is how much of our revenues come from new businesses that are too nascent to have higher gross margins.
So the more we get from Redfin Now title, mortgage, the lower our gross margins will be as a percentage, just because those businesses are still new. So our outlook on agent hiring is really affected by our outlook on the overall market. We're not hiring agents this quarter to close sales this quarter or to close sales next quarter.
All of our hiring is going to be about 2019, most of it will happen in the fourth quarter. We ramped earlier last year, because we knew we are going to invest in more personal service. This year, we don't have to start quite as early. So I would expect agent hiring to happen in Q4.
And honestly, the more agents we hire, the more you should assume we're optimistic about 2019. But we don't anticipate reducing the number of customers per agent further. The ratios are going to remain the same and the only difference will be how much demand we expect in the first and second quarters.
Chris, do you want to add anything?.
No, that's fine..
Thank you. Our next question comes from Stephen Sheldon with William Blair..
Good afternoon. I think, you've talked before about a sizable portion of your agent hiring has come from people that are relatively new to the real estate industry.
So just curious within this group that doesn't have significant prior experience, can you maybe talk about what trends you're seeing and what do client outcomes look like? Are you seeing them ramp their productivity? Just any color there on how they're progressing?.
I'm glad you asked, because we actually just presented analysis of this question to our board, and it was based on the presumption that hiring from the industry as we put more emphasis on customer success rate would be more effective. We'd had market managers, who believe that, and they were wrong.
The data shows us quite convincingly that agents who come from outside the industry, but start as associate agents, these are contractors who provide property access for Redfin, do very well if they decide to apply for full time employment here.
They have higher customer success rates, lower attrition rates, higher customer satisfaction rates and higher productivity. So the difference isn't so huge that we're no longer hiring from the industry.
But it convinced us that the training programs we have in place, the technology we have in place is really good at helping someone get up to speed fairly quickly. And that is sufficiently distinct from what a traditional agent does, that some of the work you do as a traditional agent doesn't necessarily help you thrive here..
Thank you. I appreciate the color..
Thank you. Our question comes from John Campbell with Stephens..
Hey, guys, this is Carter Trent taking the place of John. Two questions. First on Redfin Now. Zillow recently said, they're expanding their home buying segment, I know, they're only in a few markets currently.
But you guys see their competition from them? Do you think that would put pressure on sales or gross margin in the future? And second, what is the plan for buy-side rebates? Is there an inflection point for real estate gross margins for you guys, ultimately see yourself eliminating buy-side rebates at some point?.
Well, first of all, we haven't competed directly with Zillow, and its instant offers program in any market, we just don't have overlapping markets. My expectation is that, it will be a price war.
It will probably put more pressure on sales volume than gross margin, because rather than take more risk on a property, and offer a price that we're not sure we can beat when we flip the home. We would just have to step back and let someone else have that sale and take that risk, that's the discipline we want to have.
It's really easy to get your juices flowing to go after Redfin Now purchases, to offer price that you might later regret. And we are just trying to be very careful about that discipline.
It doesn't mean we're not aggressive about the business that we're not acquisitive about getting more properties, it just means that you can't buy the business and take on too much risk. So that's the answer to your first question. And then the second about the commission refund we offer home buyers, we still think it's an important incentive.
Obviously, the savings we give people who sell their home is even more effective that's why we've been so aggressive about shifting toward the seller and that is also shifted our business toward the seller. It used to be that a third of our sales - or 25% of our sales were to home sellers and the rest were to buyers.
And that has been shifting more towards sellers. So what that means is that someone who is selling one place and buying another will get the savings as a seller rather than as a buyer. So my guess is, is that balance shifts more and more toward sellers, it will put more pressure on saving money for homebuyers.
But I still want to keep the savings for both types of customers..
And this is Chris, just one more note on that, investors should expect that as the sell side of the business continues to grow faster, we would continue to reduce our buy-side refund for customers. That's the commitment in general that we've made, which is that we'll hold revenue per brokerage transaction consistent or flat as we have shifted..
Okay. Got it. That's helpful. Thanks, guys..
Thank you. Our next question comes from Brent Thill with Jefferies..
Thanks for the question. Glenn, I just want go back to the comment about the last three weeks, and I think you made a comment, that this was the worst you've seen in the last three years.
I'm just curious, if you could just add a little more color around what caused the sudden slowdown, why you're not confident that this might kind of bounce back after perhaps holiday schedule or something that's happened in terms of a quick summer slowdown.
What is giving you more caution in your voice and the comments around this forward-looking quarter?.
Well, what I owe the analysts and the investors is for you to see exactly what we see as best as we can transcribe it to you. And in the past three out of four weeks, we saw a significant sales slowdown. We still have growth year-over-year, but it's much lower growth than we're accustomed to seeing.
And some of that is because the environment, I want to be clear that when I said we haven't seen this kind of environment in three years, I wasn't talking about Redfin sales in particular. Even our worst week, we had better sales that week than we had the same week in any prior year.
But what's different is that if your real estate agents saying, I put a home on that normally would have sold in a week, and it's still on the market a month later. I expected to get eight competing offers, I got one and it was below the asking price.
And then when you look at the data to see if that supports the anecdote, it does, and this is a nuance point. But days on market isn't changing that much that how long sold homes take to sell. But the percentage of listings to sell within two weeks is decreasing.
What that means is that for the homes that sell, they're still selling reasonably fast, but more and more there are homes that we thought would sell that don't. And I would say that's concentrated in some of our larger markets.
One of the things that we're sensitive to is whether this is a decline in the overall sales volume in the United States or a shift. So for example, if Seattle or San Francisco has just gotten too expensive and now everybody's buying in Phoenix and Denver, that doesn't really affect U.S. sales volume.
And we think some of that is happening, a place like Pittsburgh still has strong sales growth. But we also think that there is probably going to be a slowdown in U.S. sales growth, if not a reversal in August or September. And there's a case to be made that the whole market will get better there's a case to be made that it won't.
I'm not presenting this as a fact, but if you want to know what we think is the most likely outcome that's our view of it. And then I think we've tried to be clear that we're not just going to talk about the market. We're not a boat that goes up and down with the tides. We've got a sail, we've got a motor, we need to make our own headway.
So I think we're focused on the effectiveness of our marketing efforts, the effectiveness of our website at getting traffic from Google. And the effectiveness of our agents at getting sales from our customers and making them happy. And those are areas where I think the only thing that's changed significantly has been traffic.
We have started to become one of the top two search sites in many major American cities. And it's given us less room to grow really fast. We're still taking search share from pretty much everyone, but it's just going to be harder to do, because we're taking it away from really well-established brands..
Thanks for the color..
Thank you. Our next question comes from Brad Erickson with KeyBanc Capital Markets..
Hi, guys, thanks for taking the questions. Just a couple. First on the headwinds, you guys are calling out, just maybe like on a run rate basis of what you've seen in these last, I think two - three out of the last four weeks you said.
Can you give us any sense of the magnitude of the slowdown you're seeing that led you to reduce the forecast? Is it 20% less on a run rate basis? Any help there would be great..
I can't do that. And I think you know that. But what I can say is that what's surprising is how sawtoothed it is, actually I flubbed a line in the script. We've had three bad weeks and one good week in the past four weeks. And the good week was really good and the bad weeks were bad. And it's made the guidance exercise for us quite harrowing.
We haven't seen this kind of week-to-week volatility much. I've been through it all, baby. I've seen some bad stuff. I've seen some weird stuff. But usually, it's consistent week-to-week in a way that it hasn't been this time.
And so, we're here to tell you what it is or that we're seeing it, I think getting into our daily sales bookings and showing you that graph, it isn't going to help anybody..
Got it. That's fair. And then second, might you potentially look a bit further downstream to lower home price points, where you obviously see - can see higher unit volumes just so overall in terms of a market. I know you look that operate at some kind of sweet spot, where you can be profitable on some baseline price per home.
Just curious, if you look to affect the mix there to offset some of these headwinds you're facing? Thanks..
That's one of the levers we have in the business to take demand that we would normally route to partner real estate agents, keep it for ourselves to make sure that our agents are always busy. Idle agents are deadly for gross margin and we don't expect that our agents will be idle and virtually all of our markets. So I'm not worried about that.
But when you do that you're still pressuring the P&L in a couple of different ways. You're reducing revenue in the partner business. And you're taking lower revenue per transaction in the brokerage business. So I think there's a way to cushion gross margins to some extent by avoiding idle agents and to keep revenues somewhat level.
But what's really important is making sure more people come to our website, more people contact our agents, more of those folks end up successful buying and selling a home..
Got it. And then, last one if I could, just to squeeze one more in, just along the lines of the ad spending you talked about particularly in the next year.
Philosophically, maybe just talk about how that's intended to either largely drive more traffic to the site or just create sort of better brand awareness? I'm sure, it's a bit of all of those things, but talk about maybe any more specificity of how that ad spend might be targeted. Thank you..
Sure. So it's not really focused on driving traffic. We actually have plenty of traffic relative to the number of people, who try our agents as customers. We really want to build awareness of ourselves as a brokerage.
And Chris and I sat in the room we're doing this earnings call maybe two months ago and an analyst took us through probably an 80 or 90 slide deck. There was a tour de force saying that a market where we ran a TV ad four years ago is still seeing an increase in sales. And then, in fact, we shouldn't have turned it off.
We should have kept going with it. And we just went through market pair aftermarket pair aftermarket pair of places that had TV and places that didn't. And what we saw is that there's a long-term revenue impact that we hadn't fully accounted for.
Many technical folks really have a crush on direct marketing, because Facebook and Google put a pixel on somebody's browser, they show up on Redfin site. You can directly attribute that sale. And people like that kind of assurance.
But for an emotional purchase like this, when your brand awareness is in the low-single-digits, you really need to invest in making sure that people understand who you are. And that has become especially true now that other players are spending so much in the space.
So I think we're going to be much more aggressive about it next year and the way will measure isn't through increased traffic, but increased trial. More people who are on the site will end up converting to become Redfin customers. More people will know that there's a better way to buy and sell a home. Most Americans still think Redfin is a website.
And that drives me crazy, because I want them to now that they could save 10,000 bucks selling a house through us, get better service and get a higher price to the home. So that's the message, not that the website is read or that it has better listing data or something like that..
Got it. That's great. Thanks..
Thank you. Our next question comes from Thomas Champion with Cowen..
Hi. Good afternoon. Back to guidance, I think we've talked about revenue, but just curious if you could hit the net income side. The guidance sort of implies a different trend from what we saw last year. And you've alluded there is some incremental expenses from newer businesses.
I'm just curious if you could talk through how you arrived at the guidance for net income. And then, Glenn, I think you talked about customer loyalty on the last call. And you touched on repeat customers again today.
And I'm wondering if you could just give us a sense of what portion of transactions are represented by repeat customers or any color you provide there. Thank you..
Sure. So, on the gross margin front I think the most relevant piece is just that, as we're going into a period of time where typically we have revenues increasing. And our guidance suggests that revenues in the third quarter will be more consistent with the second quarter.
That puts additional pressure on gross margin, just because we don't have that additional pop of volume. At this point, we've hired the agents we want to have on staff for these key summer months. And so, really what you see reflected in our real estate services gross margin is exactly that impact.
And I think the other thing really then to call out just is on the advertising spend front, where we do expect in a similar kind of way to the pickup in spending that we saw in Q2, that we'll want to spend more dollars on a relative basis in advertising in the third quarter.
And just a little bit back to Glenn's earlier comment, that is more likely on the direct or online advertising side of things as opposed to broad-scale media..
And then the referral business, it's - excuse me, the repeat business is growing faster than our other segments. We don't report on what percentage of our sales come from repeat..
Okay. Thank you..
Thank you. Our next question comes from Mark Mahaney with RBC Capital Markets..
Okay. Thanks. Two questions, and one, I think one of the initiatives that you've been stressing out working on has been kind of increased advertising spend in different cities. I think I've seen more Redfin advertising around San Francisco a few months ago.
How do you think about that? And the kind of return you've seen on that, is that something you feel more encouraged by, less encouraged by? And then, I think you talked about it earlier, but I'm sorry if I missed the back and forth on it.
Your thoughts now on adding more agents to the extent which you think that's been a successful strategy and maybe it's impossible to tell, given what's happening macro-wise. But your thoughts on whether the right reaction now is to continue to accelerate agent-adds? Thank you very much..
Hi, Mark. Thanks. So TV advertising is something where we evaluate its impact over more than one year, because it has a direct effect on some people who are about to go into the market or who are already in the market. But it seems to also have a very durable effect on folks who end up selling their home two years from now.
We have been happy with both the direct result, and the brand result. We reported on both today, but really emphasize that brand awareness that basically doubled over the space of a few years in the markets where we've done media.
So we just think that we're going to continue to reap those benefits, but we won't get them all on the year that the ad runs. And then, around agent hiring, we just don't plan to do any agent hiring in Q3. I know that we did last year. But if you go back to that transcript, you'll hear us say how unusual that is.
Almost every other year, we don't hire in Q3, we hire in Q4. You're often hiring in Q3 against October, November, December customers, and that's really our idle period. So I don't expect that we'll be hiring agents in Q3. The real question is how many we hire in Q4 and we just want to get closer to 2019 before we do that.
If we could wave a magic wand and have all those agents fully trained in one day, we wouldn't even hire them in Q4. We'd hire them in January or February of the following year..
Okay. Thanks, Glenn..
Thank you. Our next question comes from Heath Terry with Goldman Sachs. Hello, sir, your line is live..
Hi, this is Adam on for Heath. Sorry about the delay. There was on mute. Thanks a lot for the question and taking the time. Just a quick one, on the brokerage revenue per transaction, you had mentioned that it was up, it looks like, 2% year-over-year. And that some of that was due to being in markets with higher home prices.
And the other piece of that was on sort of the mix shift. If you could parse that out for us, that would be great. Thanks..
Yeah, so the difference, as compared with what we would have expected really even last year, is primarily that U.S. home prices are rising. And so, that does create some updraft on that metric for us. It's a positive trend from a financial standpoint as we get that.
And really the mix shift piece of that is just us doing a good job of forecasting how much the mix shift is going to be, so that we've targeted any changes in our pricing structure to keep that relatively level. But again, the prevailing thing here is just the updraft from U.S. home prices..
Great.
And then one quick follow-up on the - just on the traffic side, anything else to call out besides sort of the law of large numbers and the competition between some of the other top of funnel portals that have also seen traffic slowdown in recent quarters?.
I think some of it is related to inventory. What drives traffic is pretty pictures of houses. And when we put up fewer pictures, we get fewer visitors. This is true of all e-commerce sites by the way. The simplest way to increase traffic is just to put - create more product pages.
Some of it for us is related to improvements the competition has made in listing data quality. I'm not 100% sure that they've made those improvements. I try not to make statements about competitors, but I doubt they will quibble with this one. They've probably gotten slightly better there. And then, some of it is related to just really tough comps.
I mean, we were blowing it out last year, so just holding on to those gains and adding to them has been an accomplishment. We'd obviously like to do better. But I think what you're mostly seeing is the law of large numbers..
Great. Thank you much..
Thank you. And now, I would like to turn the conference over to Elena Perron for closing remarks..
Thank you, Jonathan. And thanks, everyone, for joining us today. We appreciate your interest in Redfin. And look forward to speaking with you again next quarter..
Thanks..
Thank you, ladies and gentlemen. This concludes today's teleconference. You may now disconnect..