Good afternoon, and welcome to the Lyft Third Quarter 2021 Earnings Call. At this time, all participants are in a listen-only mode to prevent any background noise. Later, we will conduct a question-and-answer session, and instructions will be given at that time. If anyone should require operator assistance, please press star then zero on your touchtone telephone. As a reminder, this conference is being recorded. I would now like to turn the conference over to Sonya Banerjee, Head of Investor Relations. You may begin.
Thank you. Welcome to the Lyft earnings call for the quarter ended September 30th, 2021. Joining me today to discuss Lyft's results and key business initiatives are our Co-Founder and CEO, Logan Green, Co-Founder and President, John Zimmer, and Chief Financial Officer, Brian Roberts. A recording of this conference call will be available on our Investor Relations website at investor.lyft.com shortly after this call has ended. I'd like to take this opportunity to remind you that during the call, we will be making forward-looking statements. This includes statements relating to the expected impact of the continuing COVID-19 pandemic, the performance of our business, future financial results and guidance, strategy, long-term growth, and overall future prospects. We will also make statements regarding regulatory matters. These statements are subject to known and unknown risks and uncertainties that could cause actual results to differ materially from those projected or implied during this call.
In particular, those described in our risk factors included in our Form 10-Q for the second quarter of 2021, filed on August 5th, 2021, and our Form 10-Q for the third quarter of 2021 that will be filed by November 9th, 2021, as well as the current uncertainty and unpredictability in our business and markets and economy. You should not rely on our forward-looking statements as predictions of future events. All forward-looking statements that we make on this call are based on assumptions and beliefs as of the date hereof, and Lyft disclaims any obligation to update any forward-looking statements except as required by law. Our discussion today will include non-GAAP financial measures. These non-GAAP measures should be considered in addition to and not as a substitute for or in isolation from our GAAP results.
Information regarding our non-GAAP financial results, including a reconciliation of our historical GAAP to non-GAAP results, may be found in our earnings release, which was furnished with our Form 8-K filed today with the SEC and may also be found on our Investor Relations website. I would now like to turn the conference call over to Lyft Co-Founder and Chief Executive Officer, Logan Green. Logan?
Thanks, Sonya. Good afternoon, everyone, and thank you for joining our call. We had a great Q3. We beat our outlook on every metric and reported our second consecutive quarter of adjusted EBITDA profitability. Demand remains strong, and we've seen a material improvement in driver supply. We are well-positioned for a continued recovery and excited about the solid foundation we've built to continue to scale our business. Let me address a few highlights from the quarter. Revenue increased by 73% year-over-year and was better than our outlook. Active riders grew by nearly 2 million versus Q2 as more riders returned and resumed prior use cases and new riders started using Lyft. Although recovery trends still vary regionally, it's clear that Lyft riders are on the move. During each month of the quarter, we set a new pandemic record for rideshare rides.
Nights out and weekend use cases picked up, and airport rides nearly tripled year-over-year in Q3. In addition, we saw strong demand for bikes and scooters, with bike rides hitting an all-time high in the quarter. Citi Bike in New York is just one example of the exclusive content only available through Lyft. In Q3, Citi Bike rides made up 40% of our total ride volume in the region. It's strategically valuable as we look to deliver increasing value to this installed base. Switching gears, driver supply has materially improved and retention has been strong. In Q3, active drivers increased by roughly 45% versus last year. New driver growth was robust, up 60% year-over-year. Keep in mind, in September, the enhanced federal unemployment benefits sunset, and we had the highest level of new driver activation since COVID began.
Just to be clear, the number of drivers matters, but so does the number of rides they give in an hour. We found that drivers have been giving more rides versus 2019 for some time. In fact, in Q3, drivers gave 20% more rides on average than they did in Q3 2019 on top of innovations in our marketplace engine and higher earnings. When drivers are busy, they can optimize their hourly earnings and support more rider demand. This is a win for drivers, riders, and our business. We built Lyft against the backdrop of a consistently tightening labor market. The unemployment rate reached a multi-decade low just before the pandemic hit. Driver earning requirements vary a lot from city to city. Ultimately, the playing field is level. Our competitors have to navigate the same factors.
As a marketplace, when conditions change, our pricing adjusts automatically and dynamically as an offset. You've seen this dynamic play out this year. We've demonstrated improving leverage even while driver earnings have remained elevated, and I'm confident in our ability to build on the momentum in our business. Turning to Q4, early trends have been positive. October tends to be the strongest month in the fourth quarter for rideshare rides due to seasonality. Brian will talk more about this, but people are typically more mobile in the summer and less in winter or around the holidays. This is especially the case with bikes and scooters. Looking further ahead, although the pandemic continues to create operating uncertainty, we're optimistic that the recovery will continue and drive additional ridesharing use cases as well as fuel Lyft's growth.
John will provide key business updates, but before he does, I'll turn the call over to Brian to review our financial performance.
Thanks, Logan, and good afternoon, everyone. Before I walk through the numbers, let me start with an update on supply. As Logan shared, we are extremely pleased with the significant impact and results of our Q3 supply investments. We entered the quarter determined to improve service levels given growing demand trends. In Q3, new driver activations increased 34% quarter-over-quarter and jumped over 100% versus the number of activations in the first quarter of this year. The growth in new drivers contributed to strengthen active drivers, which increased nearly 20% quarter-over-quarter. This improving supply position helped fuel volume growth that enabled us to outperform our financial outlook. We delivered 73% year-over-year revenue growth and on a quarter-over-quarter basis, nearly tripled adjusted EBITDA to $67 million. We also achieved record contribution margin and revenue per active rider.
Keep in mind that we generated these results despite the impact of COVID variants, which delayed the return to office for many companies. Before I move on, I want to note that unless otherwise indicated, all income statement measures are non-GAAP and exclude stock-based compensation and other select items as detailed in our press release. A reconciliation of historical GAAP to non-GAAP results is available on our investor relations website and may be found in our earnings release, which is furnished in our Form 8-K filed today with the SEC. Let's move to the details. Q3 had strong unit growth. Despite increasing COVID case counts, the sequential growth of Q3 rideshare ride volume accelerated and jumped by over 80% relative to the Q2 growth rate. This was fueled by broad sequential strength across cities.
In terms of specifics, 99 of our top 100 cities generated positive sequential rideshare ride growth in Q3. New Orleans was the sole outlier given Hurricane Ida. Additionally, average daily rideshare ride volume increased each month in Q3. Beyond rides, we saw healthy growth in unique riders. In Q3, the number of active riders increased by 51% year-over-year, and 11% quarter-over-quarter to 18.9 million. New rider activations increased by 47% year-over-year. Revenue per active rider increased by 14% year-over-year to an all-time record of $45.63. Revenue per active rider benefited from a 6% sequential increase in ride frequency, which we partially attribute to improving service levels.
The combination of these trends led to a $99 million sequential increase in third quarter revenue to $864 million, which was above our revenue outlook of $850 million-$860 million. It's worth noting that bikes and scooters provided roughly $10 million of the $99 million increase given their seasonal strength. For the second quarter in a row, we achieved a new record contribution margin level. Contribution margin in the third quarter was 59.4%, which exceeded our outlook of 58.5%-59% and represents a nearly 10 percentage point increase from Q3 of 2020. The outperformance on revenue and contribution margin relative to our outlook helped drive strong Q3 contribution of $514 million. For each dollar of incremental revenue growth versus Q2, contribution increased by over $0.60.
As a reminder, contribution excludes changes to liabilities for insurance required by regulatory agencies attributable to historical periods. In the third quarter, there was no adverse or positive development net of reinsurance recoverables. Let's move to operating expenses. Operations and support expense for Q3 was $103 million, an in crease of 12% year-over-year. Operations and support expense as a percentage of revenue was 12% in Q3, up slightly from 11.3% in Q2. This is primarily from bike and scooter rental activity, as well as growth in background checks related to driver onboarding. R&D expense in Q3 was $109 million, down approximately $20 million quarter-over-quarter, resulting from the sale of our Level 5 self-driving division, which closed in July.
As a percentage of revenue, R&D expense declined to 12.7% in Q3, down from 26.2% in the year-ago period. Q3 sales and marketing was $99 million. As a percentage of revenue, sales and marketing was 11.5%, roughly flat with Q2's 11.6%. Within sales and marketing, incentives were less than 2% of revenue. G&A expense in Q3 was $167 million, a decrease of 18% year-over-year. G&A expense as a percentage of revenue was 19.3%, a decrease of 70 basis points quarter-over-quarter. In terms of the bottom line, our Q3 adjusted EBITDA profit of $67 million was above our outlook of between $25 million and $35 million and nearly triple the $24 million achieved in Q2.
It's worth noting that Q3 adjusted EBITDA included $18 million of benefits related to two items. First, we captured additional gains of $8 million related to Flexdrive selling vehicles. Second, we were able to settle a legal matter and release an accrual that provided a combined $10 million benefit to G&A expense. Without these gains totaling $18 million, our Q3 adjusted EBITDA profit was $49 million. Unrestricted cash, cash equivalents and short-term investments increased quarter-over-quarter to $2.4 billion. Before I move to our Q4 outlook, I want to remind investors that the pandemic is not yet over. Future conditions can change rapidly and may impact our outlook. With that, let me share what I can. In terms of supply, given our success in Q3, onboarding new drivers and expected tailwinds, we plan to taper supply investments in the fourth quarter.
Of course, when it's extra busy, we will use dynamic pricing to fund extra incentives to help retain and attract additional drivers onto the platform. In terms of rides, in October, we achieved our sixth straight month of growth in average daily rideshare ride volume. As a reminder, though, in North America, rideshare faces seasonal headwinds in November and December, given the impact of holidays on demand. In both 2019 and 2020, so pre-COVID as well as during COVID, October was the peak month of the fourth quarter in terms of rideshare rides. We expect the same trend this year. Additionally, there's a population of riders who have not yet resumed their full range of pre-COVID activities. Even though ride volume increased over the summer, in Q3, we were still down over 35% from our peak. The reasons and circumstances vary.
Some people are concerned about the most recent surge in case counts and are waiting for boosters. There are parents who are foregoing certain activities until their kids are vaccinated, and then there are those waiting for mask mandates to end. For some, it's a combination of these factors. Separately, with the summer rise in COVID case counts, many companies postponed a return to office until Q1. This is especially true in a city like San Francisco. As a data point, Q3 rideshare rides in San Francisco were down by more than 60% versus Q3 of 2019, meaning San Francisco quarterly rideshare rides were less than 40% recovered from two years ago.
Given the delayed return to office and other contributing factors, the recovery boost tied to additional ridesharing use cases is more likely a first half 2022 event, especially in key West Coast cities like San Francisco. We anticipate this tailwind will help drive volume next year. It's a matter of when, not if. For these reasons, year-over-year revenue growth for full year 2022 is expected to exceed the rate for 2021. In terms of our outlook, we expect revenue in Q4 of between $930 million and $940 million. This implies growth of between 63%-65% year-over-year versus the 73% achieved in Q3. This outlook includes the typical Q4 rideshare seasonality and the delayed reopening acceleration. In addition, remember that Q3 is also the seasonal peak for micromobility in North America.
In the fourth quarter, bike and scooter revenue is expected to decline by up to $20 million quarter -over -quarter, which is included in our outlook. In terms of profitability, we expect Q4 contribution margin to be around 59%, given the impact of seasonality, among other factors. The midpoint of our outlook for revenue and contribution margin implies what would be an all-time record for contribution, eclipsing the level in Q4 of 2019. We continue to expect that contribution on a per-ride basis will be greater post-COVID than it was pre-COVID. In terms of the bottom line, we expect that Q4 adjusted EBITDA will be between $70 million and $75 million versus the $49 million in Q3, adjusted to exclude the $18 million of benefits. Similar to revenue, we face seasonal pressures in Q4, and for that matter in Q1 as well, that can pressure EBITDA trends.
In 2019, so pre-COVID, our adjusted EBITDA loss increased between Q3 and Q4. Last year, we undertook layoffs in Q4 that obscured the typical trend. The fact that we expect to increase adjusted EBITDA profitability sequentially in Q4 despite the headwinds speaks to the improvements we've made to our cost structure. The Q4 outlook implies adjusted EBITDA margins of approximately 8%. This compares with 7.8% in Q3 or 5.7% without the $18 million of benefits. Separately, based on our momentum, we continue to expect that Lyft will achieve adjusted EBITDA profitability on a full-year basis in 2021, which is an important milestone. In fact, year-to-date through Q3, Lyft has already generated cumulative positive adjusted EBITDA of nearly $20 million.
The midpoint of our outlook implies annual 2021 adjusted EBITDA of approximately $90 million, which represents an improvement of roughly $850 million year-over-year. Let me close. Since the inception, we've overcome a number of formidable challenges by remaining resilient and focused on our execution and strategy. The pandemic is no exception. Over the past 18 months, we have transformed our operating model, achieved adjusted EBITDA profitability ahead of expectations, and are now demonstrating improving leverage. Looking forward, as we emerge from the pandemic, we expect to be a stronger company with greater leverage. We plan on building a significantly larger business as we attack the massive market opportunity ahead of us. We see exciting opportunities to lean into growth to deliver solutions that serve and expand our addressable market.
At the same time, given our growing scale and expected tailwinds from the recovery, we are positioned to unlock natural business leverage. We expect that we can fund these growth opportunities even as we generate improvements to overall profitability. With that, let me turn it over to John to provide key updates on the business and our strategy.
Thanks, Brian. I'm excited by the momentum in our business and by the significant market opportunity ahead. Near term, we're focused on navigating and strengthening through the recovery. This includes relentlessly advancing our technology to optimize our real-time market balance, which makes our network even better for drivers and riders. On the driver side, we are continuing to dive deep on innovation that delivers the best possible user experience. As one example, we've been testing a new app interface that gives drivers more granular visibility into our market conditions before they start driving. The early results have been fantastic. We saw a roughly 25% increase in the number of times drivers engaged with our app and an almost 5% increase in the number of hours they drove.
Likewise, incremental refinements to our Prime Time dynamic pricing technology can support a higher ride volume, improve pickup times and increase driver pay all at the same time. We will continue our diligent focus on this work since these types of enhancements can result in higher driver engagement and retention, better marketplace dynamics, and ultimately tens of millions of dollars in leverage every year. On the rider side, we've scaled new modes that give our riders more options while also providing valuable benefits to our real-time marketplace. A rideshare mode like Wait & Save that allows riders to wait a little longer for a pickup and pay a little less versus a classic ride is one example. It delivers the most value to riders by giving them the ability to prioritize what's most important to them at the moment, price or time.
The mode also helps distribute demand in a way that allows us to optimize driver utilization and more usage of our network. Less expensive options like bikes and scooters, along with different use cases such as with Lyft Rentals, further help us to maximize the overall usage of Lyft's transportation network. These advancements and services add value today and help us prepare the infrastructure for the future as we deliver more and more transportation value to consumers. They also build on our core competencies and deepen our competitive differentiation. The Lyft network is the product of nearly a decade of engineering investment, and we are perpetually achieving new levels of efficiency and functionality with a focus on transportation. The tremendous value of our specific approach will become increasingly more apparent over the next few years.
As we look forward, we continue to see a very long runway to both add riders and capture more of their individual spend on transportation. Our work in addressing the trillion-dollar transportation market opportunity is just getting started, and the critical profitability and product milestones we've hit this year set us up well for the quarters and years ahead. Every year in the U.S., around 4 million people turn 18 and become eligible to use Lyft on their own. These cohorts have digital-first preferences, value green transportation options like bikes, scooters, and the EVs we can offer through Express Drive and have significant lifetime value. Our focused execution will allow us to establish Lyft as their go-to transportation partner. To that end, we currently work directly with more than 150 university and college partners to develop transportation solutions for more than 500 ,000 students.
Through these partnerships, students can get access to fully funded or discounted rideshare rides, as well as to our bikes and scooters, potentially for the first time. For these riders, this can be a low-risk trial period to get to know Lyft. For us, it's an opportunity to cement our relationships with these riders, become ingrained in their daily routines, and grow with them over time. Before we move to Q&A, let me give an update on the regulatory front in Massachusetts. The Coalition for Independent Work reached an important milestone in Q3 with the certification of its ballot initiative. It's worth noting that Massachusetts drivers overwhelmingly support the ballot measure by a 7-to-1 margin because it allows them to keep their independence while also securing historic benefits.
We are now full steam ahead on both a ballot and legislative solution and are highly optimistic we will be successful in establishing an independence plus benefits model for drivers. We're now ready to take questions.
Ladies and gentlemen, as a reminder to ask a question, you will need to press star one on your telephone. To withdraw your question, press the pound key. Please stand by while we compile Q&A roster. For our first question, we have Doug Anmuth from JP Morgan. Doug, your line is open.
Great. Thanks for taking the questions. Maybe first, just on driver supply, you talked about strong new driver trends and supply of 45% versus last year. So I know you're tapering the investment into Q4, but can you kind of help frame what that means more and perhaps explain a little bit kind of around what percentage recovered you might be versus current rider demand? If there's any more color you can add there. Then, secondly, just on Citi Bike, if I heard you correctly, you talked. I think you said 40% of rides in the New York region are related to Citi Bike.
Can you just talk about how that's kind of driving acquisition of customers, you know, between bike and core rideshare and anything else you can add on just acquisition trends, characteristics of those users and what you see versus kind of regular Lyft users? Thanks.
Sure. Hey, Doug, this is Brian. Let me start, and I'm going to hand off to John. Look, we definitely played offense in Q3 to invest in supply and improve service levels, and we're tapering because we see more drivers returning to the platform. As Logan pointed out, they're giving more rides. There's three contributing factors. It's the sunsetting of enhanced federal unemployment benefits. It's this increase in productivity. There's also a pool of potential drivers who are being cautious. Let me just touch on each of these. We believe the sunsetting of enhanced federal unemployment benefits is acting as a tailwind. It's too early to tell the long-term impact. But in September, driver activations, so these would be new drivers, jumped 17% month-over-month and grew further in October.
If you look at the full third quarter, driver activations, again, these would be new drivers, increased 60% year-over-year.
It's important to understand. I feel at times the market doesn't understand this. You know, last year, 85% of drivers drove less than 10 hours per week on our platform. The significant majority of drivers are using the Lyft platform for supplemental income. If you're getting $300 per week from the federal government plus whatever a state added, it really eliminated the need for some folks to drive for supplemental income purposes. Again, the jump in new drivers after the enhanced benefits expired is really providing some organic tailwinds. Second, Logan pointed out, productivity has increased on the platform. When analyzing supply, you don't have to look at just the number of new drivers, but it's also just understanding, you know, how many rides per driver.
We've seen the number of rides per driver increase versus pre-COVID. As Logan mentioned, in Q3, the number of rides per driver was more than 20% greater than Q3 of 2019. Then finally, you know, there's some conscious people who have not yet resumed driving or applied to drive, and there's different reasons. You know, some have concerns about the most recent surge and have a preference to wait for boosters. There are parents who are waiting until their kids are vaccinated, and others just don't want to drive all day wearing a mask. You know, I think conditions are expected to improve over the coming months, and as this happens, we expect to see people shift from delivery to rideshare, where earnings tend to be higher based on historical studies.
This is John. Before I touch on the Citi Bike question, just wanna also zoom out on the question around drivers. One of the data points we looked to was from the Bureau of Labor Statistics, and we looked at what you could call comparable kind of labor markets. We looked at the retail industry combined with the hospitality and leisure industry. What we saw from January through September is that our active drivers grew at a pace 5 x faster than the recovery in the combined retail and hospitality sectors. One of the main reasons I would say that is because of the flexibility provided by the platform in that you can turn on and off the app and earn and earnings whenever you want.
As people come out of COVID, we feel strongly that the type of model we offer is one that workers want. Moving on to Citi Bike, you're right. The stat is that Citi Bike rides made up 40% of our total ride volume in New York. As we've been saying for many quarters, the bike share acquisition we did a while back and the type of contractual long-term category exclusivity we get with a program like Citi Bike is strategically valuable. Kind of to drill into the question you've asked, you asked around potential crossover, we see incredible crossover between rideshare and bikes. People, consumers think of how much they spend on transportation. They think about their transportation options and compare them against each other.
We're the only place you can do that with Citi Bike. Year to date, the number of rideshare riders that tried our bikes for the first time is up nearly 100% versus last year. In New York, it's even higher, up nearly 150%.
Actually, I'm going to answer a question you didn't ask, but I'm going to preempt one that I expect, just in terms of contra revenue, related to some of the driver investments we made in Q3. Incentives classified as contra revenue increased on a quarter-over-quarter basis by $47 million. You know, as we previously shared, to the extent the quarter was stronger than expected, we plan to reinvest more into supply to improve service levels, and we're pleased that we're able to do this and still exceed our outlook on revenue, contribution margin and adjusted EBITDA. It's also worth saying that we believe GAAP revenue, which is net of incentives, is the cleanest metric for investors. Contra revenue in isolation could be a red herring, especially if higher prices are funding the driver incentives.
When it's extra busy and we need more drivers, we're focused on funding the required driver incentives from the elevated prices. Actual GAAP revenue, which of course is net of the incentives, is the clearest metric to measure the true impact of this relationship. We're gonna continue to report the amount of contra revenue, but just realize we're not focused on it. We're focused on GAAP revenue and overall EBITDA. Just keep in mind too that the second quarter, not the third, was the peak quarter for contra revenue on a per-ride basis given volume growth in Q3. In general, again, we're feeling much better on supply and looking forward, given the expected tailwinds and improving service levels, we plan to taper supply investments in Q4.
Thank you, Brian.
For our next question, we have Eric Sheridan from Goldman Sachs. Eric, your line's open.
Thank you so much for taking the questions and I hope all is well with the team. Brian, my first question, I wanted to follow up on what you just said there. As you think out to 2022, and you think about elements of pricing possibly normalizing and driver incentives also normalizing, but there being some inflationary elements around energy prices, how do you think about striking the right balance between incenting supply and drivers to be on the platform versus levels of driver earnings that drivers might have gotten used to during periods of supply-demand imbalance, as we come out of the pandemic? I'm curious for just philosophically how you guys think about striking that right balance of supply and demand as we turn the calendar into 2022.
Also on the consumer demand side, when you look at the gap between active riders getting back to pre-pandemic levels, and you take the comments, you guys had early on in the call about where there are pockets of active riders that have not come back, how much do you think that's just time versus elements where the company needs to lean in and maybe incent some of the demand or lean in on the incentive side to drive active rider growth as well? Thanks so much.
Sure. Look, we're confident that we have the right levers regardless of the pricing environment to drive strong business results. Longer-term, as we fully emerge from the pandemic, we expect for prices to be lower and volumes for obviously to be greater. Yeah, I think it's worth repeating, you know, rideshare rides in Q3 were still down more than 35% from the Q4 2019 peak. As we recover, volume will definitely help us leverage fixed costs within cost of revenue, things like depreciation and to a certain extent, hosting. Also, as we emerge from COVID, we expect ride frequency to increase, and this will help us unlock savings related to aggregated billing, as an example. Then finally, we expect marketplace efficiencies will help us increase monetization, which should help contribution too.
In general, we continue to expect to exit the pandemic structurally more profitable per ride than we were going in. In terms of just the comments around active riders, I mean, we've been very disciplined on sales and marketing. Sales and marketing as a percentage of revenue was below 15% for the sixth straight quarter, and incentives classified as sales and marketing were just 1.9% of revenue, and we feel good about our competitive position. You know, I think just maybe at the risk of repeating, our long-term strategy hasn't changed. You know, we want to win on product innovation, on customer experience and brand preference, not coupons. And we believe that, you know, R&D investments can create competitive advantages and just have stronger ROI than coupons.
We expect that absolute sales and marketing will increase in future periods as revenue rebounds. We expect that sales and marketing expenses as a percentage of revenue will likely be lower post-COVID than it was pre-COVID. For us, I think it's just a matter of time. Then just going back to something you heard, you know, every single year there's 4 million new people who become of age to use ridesharing in the U.S., and that's just a constant inflow every single year. Again, there are structural, you know, growth trends here in terms of driving up that active riders.
Thanks for the color, Brian.
Sure.
For the next question, we have Stephen Ju from Credit Suisse. Stephen, your line's open.
Okay. Thank you so much. Logan or John, knock on wood, you know, mobility continues to improve and driver supply has continued to come back. You have brought back, I believe, Shared rides in certain cities. You know, can you talk about the rider uptake there and whether it's time to more widely roll that out across the country? Brian, I think, since the onset of the pandemic, I think, you know, Lyft's done a lot of to rightsize your cost base, you know, relative to the demand levels you were seeing at the time, you know, particularly in ops and support. You know, do you think you have the resources in place to spool things back up as demand and unit growth begin to normalize? Thank you.
Thanks. This is Logan. We have, as the pandemic sort of came into place, we sunset shared rides, and we launched Wait & Save, which is, in a lot of ways, you know, fulfilled a similar market need. It's a, you know, slightly lower service level where riders wait a few extra minutes to save a little bit on every ride. We did experiment with launching shared in one market in Philly. We have paused any further scaling up, sort of waiting on CDC guidance. There's you know, nothing further you know, imminently planned on the shared front. We will relaunch Shared eventually, but we don't have a date on that yet. I'll turn it over to Brian.
Yeah. On operations and support, this line can be volatile, especially in Q3, because it can be impacted by bike and scooter growth. Given Q4 seasonality for bikes and scooters, operations and support expense as a percentage of revenue should decline by about 100 basis points. We feel good in terms of our ability to scale up there.
Thank you.
For the next question, we have Mark Mahaney from Evercore ISI. Mark, your line's open.
I want to ask a high-level question about rider and driver incentives pre- and post-COVID. I'm wondering if the impact on the financial model. It's a simplistic question, but I think what we found is that the service really has compelling value proposition for a lot of riders, and therefore there's a lot of pricing power. There's maybe less need for driver—I'm sorry, for rider incentives. There may have been some structural changes, and especially the rise of alternative driving opportunities. I'm talking about delivery for drivers such that you may see a structural increase in the need for driver subsidies.
Could you just comment on that, whether, you know, that could be all neutral to the business model that maybe Lyft in the future is less driver and more, less rider and more driver incentives just because of the change brought about by COVID, the pricing power that you've shown, you know, with riders, and then just the, now you're having a, some more competitive environment with drivers. Thanks a lot.
Sure. Let me take the rider side first, then I'll transition to the driver side. Just as a reminder, you know, in the year before our IPO, sales and marketing was 37% of revenue. At the time of the IPO, we said that, you know, long-term, we expected it between 10% and 15%, probably closer to 15%. We just reported a quarter with 11.5%. It was our sixth straight quarter again below 15%. I think that's a pretty good validation. We would rather drive growth through innovation and taking care of riders better than the competition. Again, I think longer term, we expect that sales and marketing as a percentage of revenue will be lower than it was pre-COVID.
I agree with you, I think there is generally more pricing power than anyone ever realized, existed in the industry.
In terms of earnings per drivers, again, you know, there's large buckets of. You know, if you look at a number of historical studies, rideshare tends to be king in terms of the, you know, in many. If you're looking at rideshare versus delivery, earnings tend to be higher because there's just more qualifications. You have to pass a background check to drive. You need a car that qualifies. Earnings tend to be higher. I think there are some reasons why there are some people being cautious right now. You know, the earnings are elevated, and yet there are people still on the sidelines because, you know, they may have concerns about the recent surge. They want their booster. Some have kids, and they just don't want to be driving until their kids are vaccinated.
There are a lot of folks, you know, right now, if for ridesharing, you have to wear a mask. I mean, that is a requirement. We're following CDC guidelines. For a lot of folks, I think they're waiting for that, you know, requirement to drop. I think we will expect more drivers coming in. I think going back to what I just said earlier, you know, I think we've done a pretty good job, you know, navigating, you know, high price, low price, you know, different volumes, et cetera. We have a lot of leverage to manage this and manage growth in the P&L. We feel very good about our position right now.
Okay. Thank you, Brian.
For our next question, we have Alex Potter from Piper Sandler. Alex, your line's open.
Great. Thanks, guys. Maybe another high-level question. As you're probably aware, right now it's pretty hard to buy a new car. It's pretty hard to buy a used car. Generally speaking, it's tough to buy any kind of car. All else equal, that should benefit this sort of mobility model. I'm just wondering the extent to which you're trying to capitalize on this environment, whether you regard it as an opportunity to try to get consumers comfortable potentially with ditching that second car and embracing ride-hailing as a primary mode of transportation, and any opportunities you have to do that. Thanks.
Yeah. I mean, I think it's a good pressure kind of counter to what everyone was asking like, you know, during the pandemic of, "Hey, is everyone going to go out and get a car now?" It provides kind of a backstop to that. We've spent multiple years on building out a fleet business. Primarily, we've been focused on doing that for drivers. Now we see opportunities to do that for riders. We've launched, you know, Lyft Rentals with first-party vehicles as well as third-party vehicles.
I think it just further supports having our broad and focused approach to transportation, where we can offer you a rideshare vehicle, we can offer you a rental, and do that first party and provide the best experience, a bike and scooter, et cetera. Yeah, it's an opportunity when there are shifts in the market, when there are limitations in the market on vehicles. Again, we wanna cover every aspect of transportation, to increase the size of what someone spends on transportation with Lyft, but to reduce their overall spend on transportation. That's something we've been consistent about, we stayed focused on in the pandemic and will continue to pay off.
Okay, great. Maybe one last quick one just on electrification. Obviously, this Hertz and Tesla news has been topical recently. Any additional comments on Lyft's strategy with regard to electrification would be helpful. Thanks.
Sure. Well, I can say that, again, kind of similar to what I was saying, we spent a couple of years building out this part of our business, the fleet business. We have a unique asset through our ownership of Flexdrive, which is our independently managed subsidiary. This gives us a massive strategic advantage as we convert all vehicles on the platform to EVs. You brought up EVs. We're very excited about that transition. We were the first rideshare company to commit to having 100% electric vehicles by 2030. Seeing others follow suit is a good thing for the planet. By having that ability to offer it from our providers gives us the ultimate control. We've actually had EVs available through Express Drive for multiple years.
We feel great about leading that transition with Flexdrive and partners if needed.
Great. Nice quarter . Thanks, guys.
Thanks.
For the next question, we have Ed Yruma from KeyBanc. Ed, your line's open.
Thanks for taking the question. Really interesting to hear the strategic importance of the micro mobility segment, particularly you guys calling out Citi Bike. I guess as you kind of look over the medium term, are there other municipalities that have open tenders that precipitate the growth? Can you get greater density in the municipalities that you're in? I guess just any update on kind of improvements you've made for the longevity of the devices, given that I know people can be pretty tough on them. Thanks.
Yeah. Thanks for the question. In terms of the markets, we have already a relationship in Chicago with Divvy, similar to what we do in New York City on Citi Bike. We have similarly in Boston, and we have in the Bay Area as well as a few others. All the major cities we feel quite good about, we have majority share, and those relationships typically are exclusive. We're really happy with that strategy and happy to see it paying off.
In terms of the actual devices, the fact that our largest market is New York, and being built kind of like a tank, those bikes for New York, think of all the weather changes in New York, and the way and the streets and the way those bikes can be handled. So they've been battle-tested. The great news is that we continue to take all the learnings from the bikes and scooters, and we look at a total cost of ownership model for our devices. So we look at not just, you know, what it takes to serve, or what it costs to buy the bike, but we might actually invest a little bit more in the bike so that it delivers for longer.
Our first kind of ground-up piece of hardware we launched, the Lyft Ebike, is showing massive advantages just in really early testing, in terms of cost to deliver that ride per ride. We have a bigger battery. We look at all the repairs needed, and we can fix the parts or harden the parts so that they don't need repairs. The team has done a phenomenal job, and we feel great about our strategic position as an exclusive provider in those key markets.
I also want to call out something we're doing with Lyft Pink, our membership program. In every market, you know, we sell a local bike-share membership program. It's typically done on an annual basis. For the first time, we rolled that together into, you know, in addition to that, offering a national Lyft Pink membership that gets you unlimited access to our bike and scooter network. We're seeing some great uptake from that program. It's just another example of how we can leverage the businesses together.
You can think about, as Logan's talking about, like, the bikes in New York as our exclusive content, similar to, you know, a key show, Squid Game on Netflix. If you are in New York and you want to get transportation access through Pink or through any rideshare provider, but you use Citi Bike even occasionally, it's gonna make sense for you to join with the Lyft membership because of your access to that exclusive content.
Got it. Thanks for that explanation and for giving the color on Pink.
For our next question, we have Deepak Mathivanan from Wolfe Research. Your line is open.
Great. Thanks for taking the questions. Brian, can you provide an update on where the rides volumes on a unit basis are with the 2019 levels currently? You know that some of the use cases are still lagging. Any update you can call out a few and how much they're lagging so we can think about the timeline for recovery on some of those would be great. And second question, can you talk about, you know, what you see in terms of market share trends currently? You know, do you think there's been any shifts or any change in your strategic approach to market share right now? Thank you.
Sure. Let me talk about use cases. I'll talk about Q4 rides, and then I'll talk about just sort of competitive trends. In terms of use cases, look, COVID is obviously still here, so it's difficult at this time to provide a precise view. I think in general, looking at use cases, you know, we expect all of them to return. We expect commute rides to strengthen as more companies return to the office. I'd say we're beginning to see an uptick in business travel, but it's early, and we expect that this will become more pronounced as more companies return to the office. One interesting data point is that airport rides reached 8.5% of total rideshare rides in Q3.
If you go back two years ago, airport rides were 9.1% of total rides in Q3 of 2019. While leisure has been strong, we believe some of these airport rides is actually the beginning of corporate travel. You know, looking at Q4, you know what I can say is, you know, we've learned, COVID can change quickly, so it's always important for me to start with that caveat. In terms of data points, as Logan mentioned, October was our best month, and last week was our best week since April of 2020 for rideshare rides. You know, we talked through the seasonality. You know, October is the peak month in Q4.
You know, in terms of some of these, you know, as I shared the summer rise in COVID case counts, you know, many companies as a result postponed their return to the office until Q1. This is prominent in a city like San Francisco. We experienced, you know, San Francisco is only 40% recovered. When you think about sort of our P&L and our financials right now, we're only 40% recovered. I'm willing to bet, you know, San Francisco will, you know, regain its former glory here. As we mentioned, you know, we do anticipate that this will be a tailwind to help drive volume next year. It's really a matter of when, not if.
If you missed it on, you know, our earlier comments, we do expect that for revenue growth for full year 2022 is actually expected to exceed the rate for 2021. In terms of our, you know, our outlook for Q4, given the uncertainty with COVID, we see multiple ride growth scenarios. Our revenue outlook is not tied to a single scenario that we, you know, obviously expect rideshare rides to grow quarter-over-quarter. At the midpoint, our outlook implies revenue growth of 63%-65% year-over-year. In terms of competitive trends, you know, in general, if you look at sort of pre-COVID versus where we are today, we haven't seen really any major share changes.
You know, we believe overall there's strong demand, which is industry-wide, and I think in terms of Lyft-specific trends, we were really pleased to see the volume growth accelerate in Q3 versus Q2. As I mentioned, the sequential growth rate in terms of ride units increased 80%. The growth rate increased 80% relative to the Q2 sequential growth rate.
Okay. Thanks, Brian.
Sure.
For our next question, we have John Blackledge from Cowen. John, your line's open.
Great. Thank you. Two questions. First on the driver supply, were there any geos where driver supply was at or above pre-COVID levels? If so, were ride volumes and/or service levels better in those markets? Second question, the new rider activations number was really strong. Just any particular cohorts or geos that drove the new rider activation? Thank you.
Yeah. I would say in general, you know, we wanted to play offense in Q3. You know, we sort of beat the industry to profitability in Q2, which is a big milestone, and we decided we were going to actually play offense. We're very pleased that we were able to invest, and I think you can see it in terms of results, in terms of our, you know, the supply position now, going into Q4. I think in general, if you look at third-party data, you know, you can see that there were some states which sunset federal unemployment earlier. In general, supply came back sooner. Prices came down more relatively more in those markets. I think those are probably early indications. Generally, I mean, the market's pretty efficient. It will balance pretty quickly.
I think that's probably the most I can share in terms of, you know, it's all the use cases are coming back. You know, everything, what we refer to as party hours, which is sort of the late night rides, commute, off-peak, airport rides. Like, we're seeing volume of all of those categories of rides increase.
On the rider activations?
Oh.
Any, any-
Thank you.
Yeah.
Yeah, no. Look, we were very happy in terms of on the rider activation. Again, in terms of for new riders, it jumped 8% quarter-over-quarter and up 47% year-over-year. I think as John points out, you know, we do lean into university programs. You know, typically these programs are either, you know, fully paid by the university or heavily discounted by university, so that's just a great way to habituate someone on the Lyft platform. We love those new digital natives who are going all in on Lyft. Again, it goes to this notion that there's 4 million people who become of age to use ridesharing each year.
The fact that we have bikes and scooters, to John's point, it's just another entry point into the Lyft ecosystem.
The other cohort I'd just repeat that Brian brought up a couple minutes ago is just that return of the business user. Given he provided those numbers around airport travel, you know, we're continuing to see that move in a nice direction, which is an important cohort as well. We have a whole Lyft business team executing to bring in that segment.
As John talks about business travel and airport rides, there's one thing that I think is worth clarifying for folks. You know, we've seen a lot of reports and a lot of studies, you know, allegedly trying to figure out, you know, where prices are versus where they were. I just want to say it's impossible to precisely calculate the magnitude of price changes since ride details can change. What I mean by that is the average time and distance of rides can change, and the mix of cities and modes can also impact average prices. For example, in Q3, they mentioned airport rides as a percentage of total rideshare rides continue to increase. Airport rides tend to be more expensive because they're longer.
First you have a shift in terms of these longer rides that are, you know, if you're looking at credit card data, look like higher-priced rides. Separately, as more people resume their normal lives, traffic is returning. Even if you take a ride that was the same exact length and you compare it to, say, January, that ride's going to be more expensive because it's a longer ride in terms of minutes, and minutes factor into prices. You know, you have to remember COVID artificially reduced traffic. Then for the folks who are trying to do comparisons versus two years ago, it's really important, and John touched on this, our lowest price option, Shared Rides, is back in only a single market, Philadelphia.
Shared Rides were only offered in bigger cities, which tend to be the same cities with these attempted pricing studies. Just as a useful data point, in Q3 of 2019, Shared Rides represented approximately 40% of rides in Miami, and right now it's zero. If you can imagine doing a weighted average, you know, calculation on average price, because of mix shift, that will look different.
Thanks so much. Super helpful.
For the next question, we have Itay Michaeli from Citi. Your line's open.
Great. Thanks, everybody. Just two quick ones from me. First, maybe Brian, can you comment on where your service levels are right now relative to internal targets? Secondly, going back to electric vehicles and the target for 100% by 2030, how are you thinking about managing Lyft's asset intensity, you know, in the next several years in terms of how many EVs are you willing to have kind of in a owned fleet relative to partnering with third parties and remaining more asset light?
In terms of service levels, look, you know, as I've mentioned, we do expect as we get more and more drivers back, again, there's definitely some cautious people who again do not want to be giving rides right now. Today, they are doing some more, probably more on the delivery side. Again, you know, looking at all the historical studies, rideshare, you tend to make more money versus delivery. We do expect that we will see this influx of new drivers again, and I think that will continue to improve service levels. You know, for us, there was a big focus in Q3, and we feel really good in terms of the impact on both lowering average price as well as improving pickup times.
That's an area we continue to focus, and we always want to get better there.
On EVs, again, I feel strongly that the way we're running the fleet business gives us a really important edge. There's kind of two types of third-party. You could have third-party operators, and you could have, like, third-party financing partners. The operations to us is super important to do well, to get right, and to not pay kind of a middle person extra fees. That's what I mean when I talk about first-party. There's lots of smart ways to finance it with third parties, which I think we're gonna get better and better at, and we're exploring, you know, lots of different options to do that so that it doesn't burden us.
Operationally, to win the transition to EVs, doing that in a first-party way in key markets and then getting national coverage through third parties, we believe is the right way to do it.
Sorry, this is Brian. The only other thing I would add, as we look at total CapEx for 2021, we now expect it will be lower than 2020.
Got it. Lower. Thanks. That's very helpful.
For our next question, we have Brent Thill from Jefferies. Brent, your line's open.
Thanks. Just a question around pricing. I know you've been investing, trying to get ride times shorter and pricing down. Can you just give us just a general sense of where you stand on that and how you expect that to play out in the next couple quarters?
You know, I just shared why. There's many reasons why it's really difficult to look at sort of average price. I do think, again, when you look at where we were in Q3, what's interesting is if you look at our revenue, we're actually only 15% off our all-time high quarterly revenue with rides down, you know, more than 35% from our peak. In Q4, if you look at our outlook, I think it's around 8% off our peak. We're doing a really good job in terms of monetization. But again, in general, volume really matters in rideshare. Maybe at the risk of repeating myself, you know, this extra volume that we expect with the recovery helps us leverage fixed costs within cost of revenue.
Unlike our competitor, we include depreciation within cost of revenue. They put it in a different category. It does provide significant leverage for us. Hosting is also one where there's significant ability to leverage hosting costs with volume. It's not a one-for-one relationship with ride volume. One of the areas that we've decided and we're sort of leading the industry on is around transaction processing. One of the programs, if you use Lyft, you'll see this, we're doing aggregated billing, where we're actually, you know, we send you a receipt each time, but we're billing your credit card once every 24 hours. As ride volume comes back, there's just that ride frequency ticks up.
Most people when they're using ridesharing, you know, they'll use Lyft, and they're going to a party, and then they're coming home. They're going to the office, they're coming home, and these round trips add up. This provides a lot in terms of our contribution. There's some definite leverage there. Then finally, you know, I don't think we talk about it enough. You know, we have one of the preeminent teams, if not the most preeminent team on marketplace, when you look across the industry. What they're driving in terms of improvements on monetization, both in terms of revenue and driving down variable costs, should also help contribution. You know, as I said, we expect to exit the pandemic structurally more profitable per ride than we were going in.
Great. Thanks, Brian.
For our next question, we have Steven Fox for Fox Advisors. Steven, your line's open.
Thanks. Good afternoon. Just to follow up on that last point, I just had a clarification and then a question. One, just to be clear, you said that you dropped down about $0.60 of EBITDA for every dollar of revenue growth, and then you said increasing leverage into next year. Does that mean you're saying that drop-down ratio can improve? And then for a question, I was just curious, like, you laid out a lot of obvious leverage options you can pull on. Where does mix factor into your leverage thinking next year and maybe longer term? Thanks.
Let me talk about leverage, and then I may hand off to talk about mix. When we look at Q4, we expect to increase our take rate and generate contribution leverage. Despite seasonality, we expect contribution margin leverage will be nearly $0.55 on each incremental dollar of revenue. If you normalize for the Q3 remarketing gains, the leverage is $0.65 on the dollar in Q4, which is an increase from Q3. Q3 was $0.62. This is inclusive of the $20 million sequential decline in quarterly bike and scooter revenue, which obviously creates a headwind to contribution. It's probably worth calling out, despite this headwind, the midpoint of our outlook implies contribution will reach an all-time record in absolute dollars in Q4.
You know, in terms of, again, we're not providing an outlook on 2022 at this point, but we do expect, you know, all of our modes to be back next year. I think, again, to John's point, we wanna make sure we have the right product at the right time for the right user. And it changes throughout a day, throughout a week, throughout a month, and as folks travel. We're really excited because again, some of these other modes help us attract and really increase the addressable market for us.
Great. That's really helpful.
All right.
Thank you.
All right. Thanks so much, everybody. That is time, and we look forward to talking with everybody next quarter. Have a great evening.
Ladies and gentlemen, this concludes today's conference call. Thank you all for participating. You may now disconnect.