Tuesday, October 21, 2014

Leaked Uber's internal revenue & ride request nmbrs 12/13




Leaked: Uber's Internal Revenue and Ride Request Numbers


Leaked: Uber's Internal Revenue and Ride Request NumbersEXPAND


Valleywag has obtained detailed internal financial information indicating that Uber, the e-hailing app that has spawned a hundreds of half-baked pitches, is making a lot more money than its investors had anticipated.
Above is a two-week-old image screengrabbed from what appears to by Uber's internal dashboard, sent to us by a tipster. The image shows raw numbers, including revenue, signups, active clients, and ride requests versus ride completions for the self-stylized bad boys of Silicon Valley between mid-October and mid-November of 2013.
According to the anonymous tip, the screenshot was "taken while one of [Uber's] employees helpfully logged into their admin console from my computer." We reached out to Uber representatives last night to confirm that it was from their dashboard, but CEO Travis Kalanickdeclined to comment.


The domain in the screenshot, however, appears to be a custom built tool off the Uber.com website. Even without a username and password, the log-in page partially loaded for us twice, showing the same drop-down categories as the screenshot, as well as what looks like a picture of Kalanick, on the left, in an all-white lightbulb costume. Google's reverse image search turns up no matching results for the festive photo.
Leaked: Uber's Internal Revenue and Ride Request NumbersEXPAND
The screengrab only shows data for "Uber Global." But it's not clear whether that includes all of Uber's products—Uber Taxi, UberX, Uber Black Car, Uber SUV—in all the cities and countries where Uber is currently operating. For example, they just launched in a second Chinese city, Shenzhen, last week.
But assuming the "Uber Global" covers all or the bulk of Uber's offerings, it looks as though the company is on track to make
around less than** $210 million in 2013, based on our very, very rough calculation*. (For all the numbers crudely crunched in this post, I'm omitting the last two columns because the screenshot was taken on November 20th, which makes the data for the "2013-11-18" column incomplete, and the final column "Last 7 Days" overlaps with the prior week.)
When Uber raised a stupefying $258 million from Google Ventures and TPG Investments back in August, Kara Swisher reported that Uber was on track to make $125 million in revenue in 2013, "much higher than had been projected, and that number is growing quickly." Those investors plowed hundreds of millions of dollars into Uber assuming the company was worth $3.5 billion. If the $210 million figure is correct**, projections are 68 percent higher than sources told AllThingsD in August.
Of course revenue numbers don't reflect profitability.
Uber's rumored plan to use the new funds to "become the exact thing it hates" by building its own fleet of cars and monopolizing the car-for-hire market will be costly. The company has thus far primarily operated as a middleman, building partnerships with black car companies or cab drivers. Nothing stopped drivers from working for Uber and its competitors at the same time, which is why last month Kalanick "partnered" with General Motors and Toyota to offer drivers discounted financing or leasing on cars in exchange for exclusivity. Then there's the cost ofUber's lobbyists, the cost of driver strikes, global expansion, the lawsuit over skimming drivers' tips, frequent promotional offers for free rides and the occasional mewingsoft serve attempt atsoft power.


The document also breaks down Uber's data by categories like Active Clients, Eyeballs, and Zeroes.
Over that five-week period in October and November, Uber averaged 79,000 new signups per week, or 316,000 signups per month and 3.8 million annualized. It's not clear whether signups refers to users who have downloaded the app or if it also includes people who merely filled out Uber's online sign up form. The average number of "eyeballs" per week was 1.9 million. The average number of rides requested per week is 1.1 million. For rides completed it's around 800,000. The average percent of rides requested that were also completed was 70.5 percent, which seems high if you've ever used Uber to hail a yellow taxi in Brooklyn. Uber's average number of "active clients" for the five full weeks depicted was 430,000 per week.

It's possible that "active clients" refers to users who made a transaction through the app, and that "eyeballs," a common term for online traffic, refers to people who merely opened the app or visited the website. An industry source said active clients are probably people who have used the app within the past one to three months, adding that "zeroes" likely refers to no shows for requested rides as well as non-users.
If active client correlates to active users—a very important metric for the future of a company—a source told Valleywag a handy rule of thumb: To account for returning repeat traffic, numbers for daily active users (DAU) usually represent 25 percent of monthly active users (MAU) and that weekly active users represent 50 percent of MAU. Using that back of the envelope calculation—and assuming active clients equals active users—Uber's MAUs would be in the neighborhood of 860,000.
Regardless of what this random screenshot that showed up in our tips box proves, Kalanick wants his console to get a lot more complicated. Uber is so much more than car rides, he assured Bloomberg last week: "Uber is a cross between lifestyle and logistics. Lifestyle is gimme what I want and give it to me right now and logistics is physically delivering it to the person that wants it . . . once you're delivering cars in five minutes, there's a lot of things you can deliver in 5 minutes."
*That $210 million figure is based on the combined revenue of first five columns, which shows an average of $20.5 million per week. Annualized, that's $1.068 billion for 2013. However, considering that Uber's last reported revenue figure from August was an estimated $125 million for the year, it seems most likely that the billion figure shows the total payment for the full cost of the rides, rather than the money coming to Uber. Last week, Kalanick told Bloomberg to "generally think of [Uber's cut] as 20 percent," which gets us to $213 million in revenue for 2013. However, Uber could end up making a considerable amount less than that since the startup has added more cities and countries and likely grown its top line over the course of the year. In the same Bloomberg interview, Kalanick said Uber Black, its much pricier option, has grown "13 or 15 percent month over month" whereas Uber is "seeing more robust growth or maybe just more insane growth on the lower end than we are on the high end."
**Update: As we noted directly above, Uber could end up making a considerable amount less than $213 million. But AllThingsD offers a crucial data point, noting that in the second half of 2013, Uber went from 35 cities to 60 cities worldwide. Thus, annualizing this five week period, particularly whenNovember is likely a strong month for the service, would skew revenue much higher. Kalanick argues on Twitter that November is Uber's "slowest month of the year."
If you want to take a stab at analyzing the screengrab or have a better understanding of what these numbers reveal, please comment below or email nitasha@gawker.com.

8 65Reply




Compl = the number of completed trips

Eyes = the unique number of people who opened the Uber app

Zeroes = the unique number of people who opened the Uber app but when they did so they saw no cars available. A zero can occur because there is not enough supply (drivers) on the road or because the person looking at the Uber app isn't within a certain mileage range of the available cars.

Avail Drivers = number of drivers on the Uber network in total (receiving either commission or hourly payment)

Hourly = number of drivers on the Uber network receiving hourly payment

I just heard the same thing about "Zeros": that it's when you open the app and see no cars. Uber sees it as a way to track whether they're accurately predicting consumer demand in a location and whether they're operating at capacity with the fewest number of wasted cars. Since not every Zero would complete a ride. In the screengrab, Zeroes are roughly 10 percent of eyeballs, which seems like a good ratio.



Matt Durham, an analyst at an ecommerce company crunched some of the numbers. Here's what he found [table below]:
1. Revenue per active client (all in the 46-48$ range, which I found to be quite a bit higher than I would have expected)
2. Active client to signup ratio (again here, they maintain about a 5:1 ratio on clients who are active to new signups, pretty amazing)
3. Completed request % (not sure what the reasons are for a request NOT to be completed, I’m sure there are some that are Uber’s fault and some that are the customer’s fault), but they keep a 70-75% completion rate which is huge).
4. Eyeballs to signup ratio (quality of their traffic), for every 23 people who see the site 1 signs up.
5. The craziest stat to me is the revenue per completed request, around the 24-26$ mark. I highlighted in yellow the add’l assumed revenue they could have gotten each week by completing just 2% more requests. The number seems small but in the 6 weeks in question that would have increased revenue 2.1 million $, crazy to think about. Every ride counts, I suppose.
For me, the revenue per completed request is crazy. That basically says for every ride they give someone, they are averaging 24-26$. I would not think the average person who takes a taxi spends that much. That says to me that their ease of process, customer satisfaction, punctuality to appointments, etc. all lead people to a) use their service and b)probably spend more or be more generous with a tip than they would in a normal cash transaction in a taxi.
Also possible that Uber, as a premium service, attracts wealthier clients than taxis do, who are more willing to go longer distances. This is probably doubly true for an urban area like San Francisco and environs, which are more dispersed than NYC; nobody in their right mind making middle-class salaries would cab to Oakland or Marin, but I suspect a wealthy tech-worker wouldn't think twice about it.
Another explanation for the high average cost per ride could be that people are using Ubers more and more for billable rides. It's so much easier to deal with expensing an Uber, since you can print the receipt from your email rather than getting a crappy scrap of paper from a driver. I almost exclusively use Uber when I'm billing a ride to my company or a client, and I'm usually going from an event or my office in DC to my home in Virginia, making my average ride easily $20-$30.
Tips are already built into the Uber fare so riders are already being generous with their tips compared to a taxi. Since UberX undercut taxi prices in San Francisco they have replaced my taxi usage for times when I know that I want a ride 5–10 minutes in advance. It'll be interesting to see if the cab companies can adapt to this threat to what used to be a protected monopoly (lumping all taxi companies together)…



A couple of points to consider:
In SF at least Uber turns off the Taxi hailing portion during "surge pricing times" despite the proven fact that nearly 35%-45% of the entire city fleet is empty at any given time.
The vast majority of SF residents I have spoken with actually would prefer riding in a real taxi vs any Uber product (or their competitors) as the drivers of taxis actually know where they are going without looking it up on the GPS.
Uber may be winning in the short game but when it really comes down to it they are just the Walmart of transportation. Kill the small business and get all the same folks to come work for shittier pay and no benefits or job protection.
I seriously hope Travis K gets run over by a taxi.
Where did you hear this?: "In SF at least Uber turns off the Taxi hailing portion during "surge pricing times" despite the proven fact that nearly 35%-45% of the entire city fleet is empty at any given time."
That's interesting considering a comment Kalanick made at the recent Business Insider conference"When it’s 20% cheaper and more reliable to use Uber X, why would you use a taxi?"
I am a taxi dispatcher, driver and industry insider. We have been testing this regularly and comparing it with known data from the SFMTA and taxi companies about taxis being available. In fact at the extreme peak times when Uber says there are no taxis available you can see for yourself by flipping over to Flywheel and see that there are at least 200 taxis available roaming around the city. UberTaxi... Nothing available. Yet Flywheel (the taxi hailing app in most taxis in the city... 3 min to pickup...
Also as a dispatcher watching the status of all the cabs in one fleet I can see that almost all times of day 40% or more of the fleet is empty. I suggest reading the report done by Hara and Associates for the SFMTA on managing taxi supply....http://www.sfmta.com/sites/default/...

No comments:

Post a Comment