Navigating the World of Digital Mapping

To the cursory observer, online map services like Google Maps or Bing Maps may seem like simple tools, simply placing a searchable compilation of points of interest on a scrolling set of map images. In reality, it’s a very complex business with immense potential going forward, with demand coming from transport electrification, autonomous cars, consumerization of ground logistics (UPS -> Uber/Lyft), and broader use cases for unmanned aerial vehicles, among other areas.

Mapping entails digitizing the physical world, so every map service at its root needs access to mapping data. This consists of the actual imagery – satellite images, aerial photos, and street level photos, for instance – mapped to a digital overlay of roads containing all manner of metadata (e.g. street name, type, traffic direction, speed limit, toll road). Collecting these data is an immensely labor-intensive on-ground task that is never complete (as roads and buildings keep changing), so there are really only a few global players in this space that almost all map-based services ultimately get their map data from – namely HERE (originally Navteq; recently sold by Nokia to Daimler/BMW/VW), Google, and TomTom.

There are a few hybrid players – e.g. Microsoft sources map data from HERE and others, but also had a hundred employees building out their own map data via street vans, aerial imagery, and such (a division recently sold to Uber), and Apple, which recently entered the mapping space with Apple Maps, gets its data from TomTom but is also building out a fleet of its own mapping vans.

On top of map data, you need routing algorithms, address and point-of-interest data, search, and lots more.

Below I will start with an anecdote about my introduction to the world of mapping and then discuss some opportunities in the space today.

TA Maps and Google

After college, I shipped out to India to work at Mahindra, which is India’s largest automaker (and also the world’s largest farm equipment manufacturer, among other things). After moving into my apartment in Mumbai, I realized a few things—one, the Google Maps app, which back home in the U.S. I used quite extensively on my phones at the time, an iPhone and an HTC HD2 (running Windows Mobile 6.5), had incomplete data in some parts of the city, so pretty often I’d be switching between Google and other map apps. Then I upgraded to an HTC HD7, running Microsoft’s rebooted-from-scratch Windows Phone 7 OS (whose story I’ve written about), and there was no Google Maps app in the store at all.

Windows Mobile had earlier conquered the pre-iPhone high-end PDA/smartphone market, crushing Palm OS with a remarkably feature-packed and open OS. So if Google wanted its mapping service in high-end mobile users’ hands, it had to be on Windows Mobile (just as it had to be on iOS later). Yet as many large tech companies often do (e.g. MS ceasing development on Internet Explorer after IE6, having beaten Netscape, only to be woken up later by the upstart Firefox project), Microsoft was busy running a victory lap when the iPhone launched and took a while to respond, by jettisoning Windows Mobile completely in favor of the ground-up Windows Phone 7. Meanwhile Google’s acquisition, Android, launched as a very Windows Mobile 6-like response to the iPhone. By the time Windows Phone launched, Google felt it could forego its biggest rival’s platform entirely and thereby perhaps gain a competitive advantage for Android.

So, with an incredibly smooth Windows Phone 7 device that I wanted to use daily, and no Google Maps in front of me, I sought to fix the problem by writing my own mapping app – TA Maps – that would initially serve as a Google Maps client and then expand to include multiple map sources, thereby solving the constant switching problem I had with Google Maps on iOS and Windows Mobile 6.x. To do this, I sourced map tiles from Google (and later Bing, OpenStreetMap, and others), plugged into their point-of-interest search and directions APIs, and then handled a bunch of curiously complicated tasks like reverse-engineering Google’s compression algorithm for map polylines (e.g. route lines on a map for directions).

With multiple data sources, I solved my own navigation problem and others’ too (e.g. building in OpenCycleMap for bicyclists). In the process, I put the app up on the app store and gained thousands of free and paying customers across the world, learning a ton about mapping in the process (e.g. when customers in China all reported the map as being off by a certain distance, I found that the Chinese government had at some point built a location offset from the (US military-run) GPS system, as a rudimentary security measure ensuring that all non-China-specific maps would be off unless they specifically compensated for the offset).

Then Google began to restrict access to its map data, deprecating old versions of its API and forcing users onto its new API, which required 1) authenticated tokens that identified the particular client requesting map data, and 2) agreeing to ever-narrower usage terms. When the API was updated to essentially ban native third-party navigation clients from using Google Maps, I received a not-so-friendly email from the Google Maps team – not quite a takedown notice yet, but clearly on the way. At that point, I decided to just take down the app (it still had standalone value sans Google, but I was too busy with my actual job to maintain it). Around the same time, another app emerged, as a pure-play Google Maps client that was even (egregiously) called “gMaps” and used a modified version of Google’s own Maps icon as its own. The difference? Those developers were in Russia and had no qualms agreeing to terms that they’d then explicitly violate (and then fight a technical cat-and-mouse war around Google’s API access blocking).

Google clearly saw map services as a tool to gain a competitive advantage in other areas of its business. For instance, when Motorola – then one of the top Android phone manufacturers – decided to use the services of the startup Skyhook Wireless to provide its users better location sensing than Google could provide, Google’s top executives responded with fury to the threat of losing consumer location data, forcing Motorola to switch course on Google’s supposedly “open” Android platform.

A couple years later, in January 2013, I and some others online discovered that Google had begun to specifically block Windows phones from accessing its own Google Maps website—presumably trying to get users to switch to Android. Google somewhat absurdly claimed that this was because Google Maps only worked well on browsers built on Webkit (i.e. Chrome, Safari) – strange, as the site worked fine on desktop Internet Explorer, Firefox, etc.

As I wrote here, if you changed the user agent (UA – a piece of identifying text by which the browser tells websites about itself and the device it’s running on) of Google’s own desktop Chrome browser to pretend that it was running on Windows Phone, it would no longer load Google Maps, and conversely, when a different UA was used on a Windows phone, the site loaded perfectly fine. Eventually the mainstream tech media picked up the story, and having been caught red-handed, Google was forced to re-allow access to its site. (incidentally, so much for “Don’t be evil”)

HERE, Uber, and Waze

Last year, Nokia put its market-leading maps service on the market, by then rebranded from Navteq / Nokia Maps to HERE. This was part of its exit from consumer-facing businesses (selling its best-known mobile phone unit to Microsoft, whose then-CEO Steve Ballmer apparently also wanted to buy HERE, but was turned down by a board so skeptical of any Nokia deal that Ballmer essentially sacrificed his job for it, agreeing to a timetable for stepping down as CEO in exchange for board approval on the Nokia phone deal).

A bidding war ensued for HERE, in which Uber battled a consortium of the German car manufacturers – Daimler, BMW, and Volkswagen. Why would either of these parties be interested in what might seem like off-core-competency offerings for either? The answer is simple – the future of transportation will depend on distributed data collection.

An Israeli startup, Waze, was an early entrant on the consumer side of this space, with the basic premise that if you collected position and speed data via a smartphone app running inside consumers’ cars, and had enough users, you could get a good idea of real-time traffic flows (better than existing sources of traffic data, such as government-installed highway car counters that at best can estimate traffic at particular locations) and use this to provide better traffic-adaptive routing. Waze executed exceedingly well and was acquired by Google for $1 billion.

Waze is dependent on a smartphone running inside a car, though. What if one thought of the car itself as a device—as an increasingly sensor-laden rolling connected device? Every car on the road could provide all of what Waze sees and much more (e.g. road grades, potholes, lane markers, more precise positioning, etc.)? Herein lies the problem for carmakers—platform companies like Google (Android Auto), Apple (CarPlay), Microsoft (Windows Embedded Auto), and BlackBerry (QNX) have designs on moving beyond where they currently play – in-dash infotainment systems – and into the car as a data platform.

Carmakers hate the thought of being reduced to commodity device builders like the no-profit world of Android smartphone/tablet manufacturers. Hence the German automakers’ interest in HERE, to preemptively build out the car as a digital platform and avoid getting marginalized by Google (which is the second largest mapping player and now, with Waze, also the leader in crowdsourced road data). HERE has its own infotainment platform, but more importantly, soon every Mercedes, BMW, and VW (meaning VW, Audi, Porsche, etc.) will provide Waze-like data to HERE, building up a strong, Google-free Waze alternative. HERE’s ambition is to power both tomorrow’s cars and location-based applications of all sorts.

Meanwhile car dispatch apps like Uber, Lyft, Didi Kuaidi, Ola, and such are essentially in the logistics business. The better they can route cars, the faster customers and drivers meet, the more transactions the companies process, and the more they profit, consequently. The business of route optimization, previously limited to delivery companies like UPS (whose in-house routing famously avoids left turns at almost all costs, reducing wait time in turning lanes and avoiding accidents), is now squarely within the sights of Uber and its ilk. Uber’s driver app on Android (but not iOS) currently bounces drivers out to Waze by default for optimized routing. But that’s a ton of useful data that Uber’s feeding to Google instead of itself, and at the same time, Google’s looking to directly encroach on Uber’s terrain (with its own car sharing service), so for Uber, becoming Google-free as quickly as it can is a priority.

One route was for Uber to buy HERE and have a full-fledged mapping business on its hands. With its huge market cap, Uber could probably afford to outbid the German automakers too (which itself is something worth reflecting on). Yet Uber eventually lost that bid and opted for another strategy, which was to make a deal with Microsoft. Under CEO Satya Nadella, Microsoft is focusing heavily on cloud-enabled services and treating everything below that in the stack as a commodity (its own offerings there will eventually just be demand drivers). Part of that is a new strategy for its map services (such as Bing Maps) in which, rather than driving imaging vans around the world, Microsoft will have strategic deals with map vendors like HERE to source imagery while focusing on higher-end services (such as 3D mapping and integrating mapping into other services). So Uber and Microsoft struck a deal by which Microsoft is transferring its surface imaging unit (and the technology entailed) to Uber, and Uber will integrate deeply into Microsoft services like Office and Cortana. With this, Uber can eventually turn its global network of drivers and riders into a huge source of map data that’ll be of value for its own routing but potentially also to others.

Looking Forward

At Mahindra, I eventually headed strategy and tech planning for the electric car venture, Mahindra Reva (a startup in Bangalore that we had acquired). One of my focus areas was building out a vision for the connected car, and as part of that, I looked at areas in which we could build EV-specific experiences. One idea that came to mind was in mapping— electric powertrains are drastically more efficient than internal combustion engines (ICEs), so when looking to improve efficiency and maximize range, one starts to look at things like aerodynamic drag and road grade much earlier than with ICEs (where these things only really matter for racing cars).

Could we create map routing that would optimize energy consumption by, say, sticking to flat or downhill roads? I met with map vendors and realized the idea would be a bit challenging to implement because most navigation apps calculate the crow’s flight distance (i.e. if the land were all flat from a top view), not a 3D-mapped altitude-sensitive true distance. Further, in some regions, grade data were not available at all. We would’ve had to develop grade-sensitive navigation routines in-house, which was beyond our core competence, but the opportunity here remains significant.

There are lots of potential applications in robotic navigation as well – how would an Amazon delivery drone best navigate an urban environment (FAA rules permitting), for instance?

Clearly, much remains to be done in mapping, and it’s quite an exciting field today.

By: Ashish Bakshi


read more

When I spoke with the founders and early employees of famous two-sided marketplaces such as Airbnb and Thumbtack, I consistently heard the same story: acquiring sellers was the focus early on, but it became progressively harder to grow the buyers at a suitable rate. This is a natural progression for two-sided marketplaces, and balancing buyers and sellers is probably one of the most difficult problems in the growth stages of a company.

A two-sided marketplace is any platform where buyers and sellers come together to transact: drivers sell their services on Uber to riders, who act as buyers; hosts sell their room nights on Airbnb to the guests that buy them. In order to get initial traction, the marketplaces need both buyers and sellers to use the platform. But why are sellers more important early on? And how quickly should marketplaces grow buyers compared to sellers?

Onboarding sellers early on is critical because of the frequency of transactions

There are two reasons why marketplaces need to grow the number of sellers more quickly early on:

  1. Since sellers transact more frequently, they are instrumental in creating repeat business early on: Uber drivers give on average 6 rides per day while customers take on average one ride every 8 days (Disclaimer: all numbers are back-calculated from publically available data that are snapshots of different points in time. In addition, these are realized averages that do depend on marketplace supply and demand.). Similarly, Airbnb hosts average one stay every nine days while guests average one stay every two years. Even on Ebay, sellers transact once a week while buyers transact once a month. Startups trying to attain early traction need to dedicate resources to acquiring the users that can provide more transactions in a given time frame.
  2. Idle sellers are less dangerous than idle buyers: Airbnb hosts who have trouble renting out their rooms might choose to post on Craigslist or Homeaway simultaneously, but they are unlikely to leave the platform altogether given the low cost of keeping a posting open. The perception that they might get business once in a while can be enough to motivate them to stay with Airbnb. On the other hand, for a traveler looking for lodging on any given night, she will perform the purchase off-platform if she can’t find a suitable seller on Airbnb. Not only do buyers tend to have more alternatives than sellers because institutional sellers exist (hotels, taxis, etc.), but also buyers are harder to retain because of their less frequent transactions.

Optimal growth in the number of sellers and buyers is determined by customer lifetime value

Optimizing the growth in number of sellers and buyers depends on many factors, including timing and location idiosyncrasies. However, the ballpark numbers can be inferred by looking at retention rates and transaction frequencies.

In a marketplace that fully satisfies sellers and buyers, the following equation holds:

(number buyers) * (buyer lifetime value if all transactions are satisfied) = (number of sellers) * (seller lifetime value if all transactions are satisfied)

 Therefore,

(number of buyers) / (number of sellers) = (seller lifetime value) / (buyer lifetime value)

By the definition of lifetime value (discount rates are ignored for this analysis since it remains consistent for buyers and sellers),

(Lifetime value) = (Gross margin per transaction) * (monthly frequency) / (monthly churn)

The gross margin per transaction is the same for buyers and sellers from the point of view of the marketplace. Therefore,

(number of buyers) / (number of sellers) = (seller frequency) / (buyer frequency) * (buyer churn) / (seller churn)

Interestingly, churn also depends on frequency, though the relationship differs by marketplace. In general, though, the less frequent a party performs a transaction, the higher the likelihood of churn.

What this means for Airbnb and Uber

The frequency of transactions for an Uber driver is much higher than the frequency of transactions for an Uber rider. Similarly, the frequency of transactions for an Airbnb host is much higher than for an Airbnb guest. Looking forward, these companies will have to grow their buyer base much more quickly than their seller base. In addition, as these companies determine their steady-state churn rates on both sides over time, the data could suggest that the growth needs to be more lopsided than before.

This creates an important contrast to the traditional two-sided marketplace: Ebay. Ebay sellers do transact more frequently than buyers, but the ratio is an order of magnitude smaller compared to Uber and Airbnb. Growth in these marketplaces cannot be measured the same way as growth was measured for Ebay.

By: Philip Hu


read more

Uber’s Dead End – Germany

The global expansion of Uber’s transportation services is unparalleled. Within few years, Uber entered multiple markets across continents and today, serves customers in 63 countries[1]. However, conflicts with national transportation regulations have caused Uber headaches in several cities. One particularly difficult market for Uber is Germany. Multiple court cases have first slowed down the German market entry before on March 18 of this year, a nationwide ban[2] on UberPop was imposed. With fines as high as 250,000 EUR (approx. 272,000 USD) per violation, did Uber reach a dead end with its expansion in Germany – a key market in Europe?

[3]Key Challenges for Uber in the German Market

1

Germany’s taxi market is well known for its luxury cars. Indeed, most of German taxis are comfortable Mercedes. Not an easy environment for a taxi service business model that builds on the idea, that private drivers with their own cars provide taxi services. Nonetheless, the attractiveness of the German taxi market is high with yearly revenues of over 4.4bn EUR. Above that, an annual growth of over 4.3%[4] over the last decade underpins the potential for new players in the passenger transportation service market. Therefore, a successful expansion in the German market is quintessential for Uber to grow in Europe. However, there are three main challenges Uber is facing in the German market.

 

  1. Regulatory challenges:

The basis for transportation services is the ‘PBefG’ law in Germany. It regulates the taxi market and requires multiple safety and quality standards from taxi drivers. In the final court ruling (Frankfurt district court) in March 2015, the presiding judge declared[5] Uber violates the passenger transport law, and thus distorts competition. The main argument is that Uber drivers operate without necessary licenses as well as insurance levels to cover Uber’s services are not sufficient. As a consequence, Uber had to cease its UberPop service in the following weeks after the court ruling.

 

  1. Competition from Uber clones[6]2

Regulatory hurdles are not the only challenge for Uber in the German market. In fact, there exists a very strong competitor, mytaxi, which allows customers to call and pay taxis via an app. Mytaxi positions itself as the worldwide first taxi-app. They work only together with licensed taxi drivers and thus, circumvent the regulatory dead end Uber faces. Mytaxi claims to have 10m registered users and a network of 45,000[7] taxis. Thereby, they have a strong focus on business customers and also have partnerships with loyalty programs such as Miles & More. The functionality is quite alike Uber’s: one can see the available drivers in the neighborhood, book a trip, see the rote, and pay with the app. However, a distinctive difference – there is no surge pricing. German’s appear to prefer reliability and no surprises. A nice add-on of mytaxi is the option to request specific drivers.

 

  1. Traditional competitors beefing up:

3

Also the traditional taxi players become aware of the opportunities digitalization offers. Meanwhile, many larger regional taxi companies have launched their own apps. Apparently, the network effects of these apps are limited as they are bound to drivers of the same network. Thus, they are at a disadvantage compared to a mytaxi, who has taxis in every major city and across different taxi companies in the network. As the prices are the same, there is no real incentive for customers to choose a company specific app vs. apps that connect different taxi networks.

 

 

 

Key Learnings for Uber: Learn How Germans Think

First of all, Uber’s dogma ‘rather ask for forgiveness than for permission’ did not work out at all and heavily damaged the image / branding of Uber in Germany. My advice: ‘rather ask for permission than for forgiveness’, because Germans simply not good at forgiving mistakes. Secondly, proactively regulate oneself. Uber can offer an adjusted service that complies with German regulation. Similarly to mytaxi, Uber needs to partner up with licensed taxi drivers. The challenge will be to be more attractive for drivers than mytaxi. Uber could leverage its size to offer special services to drivers (e.g. better rates at car dealers, repair shops, or car washes) and aggressively offer bonuses when joining Uber as a driver (similarly to the 500 USD bonus in the US). Finally, Uber needs to step up to the high expectations of the German market. Linking Uber with Miles & More, the leading loyalty program in Germany, is quintessential (not only SPG as in the US). Moreover, Germans are accustomed to the option to request regular drivers as well as order taxis in advance. The latter option is particularly important for business travelers.

[1] https://www.uber.com/cities

[2] http://fortune.com/2015/03/18/german-court-ban-uber/

[3] http://www.autobild.de/bilder/taxis-aus-aller-welt-3500104.html

[4] Own calculation, based on numbers from: Deutscher Taxi- und Mietwagenverband; BMVI; Deutscher Taxi- und Mietwagenverband – Geschäftsbericht 2014/2015, Seite 113

[5] http://www.bbc.com/news/technology-31942997

[6] http://upload-magazin.de/blog/7859-mobilitaet/mytaxi/

[7] https://de.mytaxi.com/index.html

By: Frederic Rupprecht


read more

Companies that build online marketplaces struggle to solve the dreaded chicken and egg problem—how do you get buyers without sellers and sellers without buyers onto the platform? Many of today’s familiar online marketplaces—like Uber, AirBnB, eBay, Xbox Live, PayPal, OpenTable, and Angie’s List—managed to crack this challenge and reach sustainable scales.  In addition to creating useful products for their users, most analysts would agree that these businesses also benefit greatly from positive network effects between their buyers and sellers. That is to say, each additional buyer or seller on the platform generally has a positive effect on his or her counterparty:

  • More sellers à more options for quality, price, and convenience for buyers
  • More buyers à more opportunities to transact for sellers

Many people are ready to give these companies the benefit of having solved for positive network effects on their platform. In fact, I too fell into this trap until Professor Gautam Ahuja of Ross School of Business (visiting at HBS last year) made me re-examine this belief. Rather, there exists an opportunity to more deeply understand these companies by exploring the network effects across all party interactions. We can gain a better idea of the costs involved in growing the business and what valuable follow-on products should be developed. The basis for this company analysis begins by understanding that multi-sided platforms must inevitably also have multiple network effects. To clarify what this means, it might be easier to first understand what network effects look like on a single-sided platform.

A Single-Sided Network Example: Early Facebook

When Facebook first came on the scene in the mid-2000s, it resembled a single-sided platform. In the time before marketers were peppering the site with ads—weren’t those the days!—users basically joined other users in a system where all users were alike. We can think about Facebook’s network effects by looking at what happened when a single new user joined (an increment of +1 new users).

The incremental user’s content provided other users with more to see and explore. This content came in the form of posts, photos, and music preferences and meshed seamlessly into other existing users’ experiences. An existing user now has more to explore and interact with on the site—like seeing pictures of the new user’s trip to Bali or discovering linked articles the new user shared. Existing users’ content might also get more interactions from the new user in the forms of more likes, comments, and tagged photos. These could all be described as positive effects as they generally enhanced the value for Facebook users.

However, the addition of a new user could make the Facebook experience turn negative as well. The new user might constantly post pictures of their cat and make it difficult for existing users to find the content they really wanted to see in their newsfeed. The new user could also be someone existing users didn’t want on the site, like a coworker or boss, which would make them less inclined to share photos or other content. Quite simply, the network effects induced by one new Facebook user could be positive or negative.

The insight into Facebook’s network effects helps us understand a lot about what followed in the company’s history. Because the effects were predominantly positive, people invited their friends and family to the site virally, reducing expensive marketing and growth costs. The site exploded and reached 1.49 billion monthly active users as of June 2015 and saw 1 billion different users log in on the sam day in August 2015. Facebook subsequently developed products like a filterable newsfeed (no more cat pictures from the new user) and privacy tools to reduce users’ pain points (only share photos with friends or make posts visible to friends “except XYZ” people) that resulted from its rapid growth and negative network effects. These developments could have been predicted by our simple analysis of identifying the network and its positive and negative attributes.

A Multi-Sided Network Example: AirBnB

When thinking about multi-sided networks, the model for analyzing network effects grows more complex. Rather than thinking about how an incremental user will affect the entire network, we should scrutinize who the incremental user is and what network is being affected.

To illustrate this more clearly, let’s consider a two-sided platform like AirBnB. When a new user joins the AirBnB site, we should first consider whether the user is a guest or host. Next, we need to explore how that additional guest or host affects other guests and hosts. For this two-sided network example, there exist 4 possible network effect scenarios:

  • How does an incremental guest affect hosts?
  • How does an incremental guest affect other guests?
  • How does an incremental host affect guests?
  • How does an incremental host affect other hosts?

Understanding network effects can quickly get complicated when dealing with multi-sided platform. The number of unique network effects necessary to consider is n2 for each distinct n type of parties on the platform. As we saw with Facebook, network effects can be either positive or negative, complicating our understanding of two-sided platforms even more. When considering positive or negative effects, the interactions that should be examined are 2n2.

Below I’ve illustrated how AirBnB might experience positive and negative network effects across all 4 of its network change scenarios:

AirBnB Network Effects Diagram

 

An Updated Model for Thinking About Network Effects: Uber

The AirBnB analysis is a useful starting point, but I find it easier to simplify each of the distinct networks into a more manageable characterization. This reduction allows us to quickly understand the dynamics in the networks of a company while maintaining an explainable simplicity. I therefore classify interactions into one of three buckets.

  • Collaborators (Positive)—Parties predominantly enhance the experience of other parties in the network. Examples of collaborators include funders on Kickstarter who together to support an idea or product or gamers on xBox live who play together in multiplayer Halo battles.  Such relationships encourage users to invite other users to a site, and can lead to organic site growth and lower user acquisition costs.
  • Counterparties (Positive)—Parties are involved predominantly in a monetary transaction or exchange that satisfies both sides. Examples buyers and sellers transacting for a deal on Groupon or a mother ordering food via delivery service Sprig. Parties exchange clearly identifiable goods and services, which, when priced at a point such that the transaction clears, creates value for both parties that supports repeat usage and high user lifetime value from multiple transactions.
  • Competitors (Negative)—Parties predominantly compete for resources or opportunities. Examples include applicants applying to Y Combinator where only a select number of applicants are accepted or eBay bidders competing against each other for a baseball autographed by Mickey Mantle. In both cases, competition is likely to give users a worse user experience as they might not secure the opportunity or good or end up paying a higher price. This can result in a lowered user experience, unsubscribing, and high sign-up or reactivation expenses.

By using this framework, it helps me understand the operational costs a company is likely facing and what products they might consider developing in the future. For example, if this analysis were applied to ride-sharing start-up Uber, it might look like this:

Uber Network Effect Diagram

For Uber’s two-sided platform, a large part of the company’s value comes from solving the most obvious network dynamic: matching drivers with riders and riders with drivers. The company was able gain a foothold in markets like San Francisco because cab companies were not keeping pace to satisfy this counterparty need. As more drivers joined Uber, riders benefitted with greater ride availability and more riding options (uberPOOL, uberX, uberXL, uberTAXI, UberBLACK, uberSUV, uberSELECT, uberPOP, uberBIKE, etc.). As more riders sign up, drivers are more likely to match with a pick-up request and earn money for their services. These services led to increased usage by both riders and drivers as value was realized.

However, this dynamic doesn’t necessarily lend itself to growth. Drivers aren’t actively inviting or converting new riders, and new riders aren’t energetically recruiting new drivers. The first time they usually encounter each other is during an Uber ride, at which point they’re both already on Uber’s platform. While a positive experience for each party—a clean, convenient ride for the passenger and a profitable transaction for the driver—will influence who how often the other party uses Uber in the future, they’re not actively growing the platform.

The story gets more interesting when you look at the network dynamic across the rider <-> rider and driver <-> driver dimensions. For current riders, each additional rider chiefly means increased competition for resources. In Uber’s case, this manifests itself in surge pricing when many people try and use the app at the same time—such as during a rainstorm or Friday rush hour. The experience is painful, and users are upset by either the wait time to find a rider or the total cost for the trip. Similarly, as more drivers join the platform, existing drivers face both increased competition for riders and reduced chances for earning surge prices. If driver’s aren’t able to find riders and drive around unoccupied, this cuts into the driver’s ability to earn for time worked.

By understanding the interplays occurring across its networks, it’s easier to identify and appreciate Uber’s growth, marketing, and development challenges over the past few years. While Uber has benefitted from word-of-mouth marketing for its remarkable service, much of Uber’s recent growth depends on promotions and discounts rather than virality. Because riders and drivers aren’t actively working to sign up other drivers and riders without an incentive, Uber bears the burden of these growth costs itself.

For example, Uber offers money to users (riders and drivers) who sign up new riders and use a unique promotion code (mine is below—feel free to join using it!). Additionally, Uber will give new riders one or more free rides upon joining. To sign up new drivers, Uber offers drivers a $500 bonus after completing their 20th rider, $500 dollars for signing up a Lyft driver, and fixed hourly income guarantees to ensure new drivers realize monetary gain immediately—with promotions often varying city by city. These acquisition costs can be large for a company looking to scale globally and help explain why Uber has raised massive amounts of cash to grow operations in areas like China, India, and other parts of Asia. In addition to building and operations costs, a lot of that money will likely go to signing up riders and drivers through aggressive promotions and discounts and launching citywide marketing campaigns. Given Uber rider’s lifetime value from its positive counterparty interactions, such costs can likely be easily justified.

Uber Promotion Code

On the product development side of the business, Uber’s network effects can explain a lot of what the company has focused on building. Paramount to the experience is maintaining a positive rider <-> driver and driver <-> dynamic. Anything that facilitates a quality service has taken precedent in the pipeline to protect the company’s advantages and keep users using the app. Such developments include credit card scanning for easier payment, license plate information to help riders identify drivers, written explanations for 3 star or below reviews to protect drivers’ reputations, and optimal route maps to make sure the most cost-effective route is taken.

Uber has worked to tackle negative network effects inherent to its business as well. Uber launched a fare split feature that aims to make the riding experience more collaborative and less competitive, easily allowing users pay each other and receive ride receipts. Additionally, a feature was added to surge that allows users to be notified once surge has dropped below as certain level, decreasing the pain from increased competition over resources. Finally, uberPOOL matches different rider pairs so that a each group receives a lower fare for carpooling with the other party. These features all subtly aim to shift the experience from competitive to collaborative.

Looking Ahead

In this post, I hope that I’ve helped lay out a new model to help understand network effects for multi-sided online marketplaces. By identifying all the distinct networks that exist and then understanding the interplay of people in those networks (collaborators, counterparties, competitors), one can develop a valuable tool for understanding much about an online business. This insight can be used to analyze a company’s growth and marketing costs—will users sign other users up or do they need to deploy resources? It can also give vision into a business’s product development priorities—what networks need to be protected with better products and which networks need pain points addressed?

 By: Ry Sullivan


read more

Uber for X is all the rage. But can (and should) we try to apply Uber’s model to one of the world’s most stagnant institutions?

Uber

10 seconds. Maybe a little longer if you mistype your password.

That’s how long it takes to call an Uber. Through the app, you can see your driver’s name and rating; you can see the make and model of the car and track its approach; and you can even call if one of you gets lost. After the ride, payment is handled automatically and you can leave a review that helps decide whether the driver gets to continue working with Uber. All this at a fraction of the cost of taking a taxi.

If you’re reading this blog, you’ve probably already used Uber. Since 2009, Uber has grown into a $50B behemoth that offers its services in 300 cities in 57 countries.

Given its tremendous valuation, it’s no surprise that entrepreneurs and investors alike have rushed to support Uber-like models in other industries. Need an on-demand dog walker? Meet Wag. How about a mover? Try Lugg. Left your plane at home? There’s even an Uber for private jets.

UberX

For now, many of the above services are only relevant for narrow segments of the population. Sorry, Glamsquad, but the average American doesn’t need on-demand hair styling. But could an “Uber for X” startup work in a different type of business, perhaps one that could affect nearly every individual in the world?

As an education entrepreneur, I’m excited about the potential impact that Uber-like services (often labeled “on-demand mobile services” or ODMS) can have on the way we learn, from early childhood education to professional development.

 

How education has already changed

First, some context.

classchanged

Don’t listen to the naysayers – education certainly has changed over the last century. In the 1950s, US classrooms focused almost exclusively on rote memorization. By the 1970s, they had begun flirting with the Open Classroom model. By the 2000s, schools were beginning to emphasize softer skills like the 4C’s: communication, collaboration, critical thinking, and creativity. Policies like No Child Left Behind have encouraged more standardized testing, and things like cursive, arts, and recess are mostly absent in today’s schools. Charter schools have transformed the public school landscape by allowing administrators to adapt organization, culture, and curricula to community needs. From a technology standpoint, educators are increasingly using online video libraries like Khan Academy to “flip the classroom” and software like Google Classroom to promote collaboration. Interactive whiteboards have replaced chalkboards in over 2.8 million K-12 classrooms globally. From a pedagogy perspective, 46 states have adopted national Common Core standards which emphasize a quicker path to literacy and deeper mathematical understanding.

Despite progress, in other ways education has been slower to adopt changes seen in other industries. Students don’t have individual learner profiles which follow them around from school to school and capture their strengths, weaknesses, and learning preferences (like electronic medical records in healthcare). We don’t really have a great sense of how to collect, analyze, and interpret student data (like we do for consumer data in e-commerce). Most learning management systems don’t have robust predictive algorithms to truly personalize learning (like the ones powering Netflix or Amazon’s recommendation engine).

So while education has changed somewhat, the pace of this change is remarkably slow compared to other sectors. Can we change education as swiftly and extensively as Uber is changing the taxi industry?

To figure this out, let’s:

  1. Break down Uber into its component parts
  2. Imagine what these components would look like when applied to education
  3. Identify education-specific challenges and way to overcome them

 

Component 1 – Two-sided platform which matches latent supply with unmet demand

sharing

A two-sided platform in education would connect educators who have excess capacity, with students who have unmet demand. Here, educators aren’t teachers in the traditional sense. Uber educators are anyone with the ability to help others learn a specific skillset. They may themselves be experts in the particular area they are teaching, or they may simply know how to help others become expertsin that area.

This looser definition of a teacher is relevant because the US is facing a nationwide teacher shortage, so latent supply for an educational Uber would need to come from other professions (e.g., engineers, writers, musicians). For starters, we could focus on the areas of education with the greatest supply/demand mismatch. Given the increased emphasis on STEM in schools and the increasing number of workers moving to the tech industry, matching experienced  tech professionals with students interested in technology could be one option.

Challenges:

Non-traditional teachers may not know how to teach – You could be the world’s foremost biologist, but that doesn’t necessarily make you more qualified to teach middle school biology. Teaching requires more than just subject knowledge – it requires an ability to connect with students, an understanding of pedagogy, and unbelievable patience, among other skills. Any two-sided platform in education needs to either properly train the supply side in effective teaching methods, or pair non-teachers with teachers to give students a solution that combines content expertise with teaching expertise.

Unions would fight back – Teacher unions are notoriously resistant to outside influences in the education system, which they fear could erode their power. And given teachers’ immense political clout, an Uber-like business in education would be prudent to figure out how to deal with unions before they are overburdened with regulations that stymie its growth. It could potentially position itself as a supplement, not competitor, to teachers, or perhaps employ the Uber approach of lobbying for regulations that fight against incumbents.

 

Component 2 – On-demand, mobile service

This actually has a few sub-components, so let’s examine each individually.

sharing

  1. Exactly what you need – Just like Uber gets you from point A to point B and does little else, an Uber for education should deliver exactly the educational content you need in bite-sized chunks. The best way to do this is to tie content to action-oriented outcomes. In other words, you want to learn how to do X? Let’s match you with someone who can teach you how to X. No more, no less.
  2. When you need it – In order for Uber to match every rider with a driver within minutes of opening the app, and to keep drivers close to full utilization, it needs to carefully connect supply to demand and ensure neither side outgrows the other. Students find that instruction is most effective when it is delivered when they need it most. Perhaps you can expand the capacity of individual educators by making it easier for them to serve multiple students at the same time, without sacrificing quality. And if we assume the teacher-student interactions can take place online, a supply of teachers located across multiple time zones can support 24/7 coverage.
  3. Where and how you need it – True on-demand learning means that students are taught wherever they are most comfortable, whether that’s at the library, in the classroom, or from home. They also need to be able to choose from multiple teaching styles (e.g., video lectures, problem sets, group projects) in order to learn in a way that best suits them. A service could either match specific teaching styles (e.g., expert video lectures) with specific learning preferences (e.g., student wants watch videos on the subway), or make it easy for an educator to convert a lesson into multiple formats.

Challenge:

Students may not be able to outline specific learning needs – What if a student is falling behind in algebra and wants general math help, instead of help with a specific algebra problem? Well just like you wouldn’t order an Uber without having a single, imminent destination in mind, Uber for education may not be right for students who can’t modularize their needs. Any viable service would most likely need to focus on specific, manageable chunks of education, either in the form of specific homework questions, learning objectives, or even tasks at hand (i.e., professional development). If students can’t break their needs up into chunks themselves, the service should be able to help.

 

Component 3 – High-quality, community-rated suppliers

Just like with Uber drivers, any teacher on the platform should be continuously rated and dismissed if their rating falls below a predetermined threshold.

rating

Challenges:

It’s difficult for a student to gauge a teacher’s quality –It is arguably easier to assess the quality of an Uber experience than an educational experience, partly because preferred learning styles and speeds vary vastly among individuals. Instead of giving a rating on a single dimension, an educational Uber could ask for teacher ratings on multiple dimensions like content knowledge, patience, enthusiasm, etc., like AirBnB does by asking guests to rate accuracy, cleanliness, location, etc. The service could also periodically match teachers with “master teachers” and have them be rated more formally, much like they would be rated by an observer in a school.

Digital lessons aren’t as effective as in-person sessions– As anyone who has taken an online course can confirm, learning over your computer or phone isn’t always as engaging, and thus less effective, than learning in an actual classroom. An Uber-like service can mitigate this decrease in quality by focusing on subjects which can be augmented with technology, not ones where technology is just a medium for the interaction. For example, a physics teacher can easily send over simulations to students to help them understand tough concepts, or a programming teacher can do a quick code review with any number of digital tools – activities which wouldn’t be as seamless in-person.

 

Component 4 – Low cost

Uber keeps its prices to consumers low by paying drivers as contractors, not employees, circumventing a taxi monopoly which requires taxicabs to own highly-priced medallions, and using scalable technology in place of overhead like a central dispatcher.

An Uber in education may to keep costs low by having younger, more inexperienced employees (e.g., college students) work as teachers, or by hiring lower-cost teachers from overseas like some online tutoring services do.

 

Maybe we can Uber-ize education, but should we?

sharing

In summary, there is definitely opportunity in creating low-cost, on-demand learning experiences, especially in disciplines where there is a growing demand for teachers and flexible supply (e.g., technology professionals can  supplement the  learn-to-code movement).

However, it is important to consider the effect such a disruption would have on the broader education system. Many students would not benefit from an educational Uber, from reasons ranging from being unable to pay for a premium educational service, to not being engaged by a digital teacher (and perhaps no in-person teachers are nearby).  As a result, the service may miss the students who need it the most, entrenching inequity issues we have been fighting for decades. And just like the success of Uber may leave the lowest-quality drivers left in the traditional taxi industry, we should worry if on-demand education gives wealthier students a monopoly on the best teachers.

Furthermore, the analysis above assumes that the goal of education is to learn a specific, well-defined skill. That premise is itself controversial. When Governor Scott Walker of Wisconsin tried replacing the words “search for truth” with “meet the state’s workforce needs” in the state code for universities, it was met with backlash from academics and educators alike. They argue that an education should provide a sense of social responsibility, encourage discovery, and instill wonderment about the world. Any service that claims to educate should serve this dual purpose of preparing students for work and life.

 

What next?

Several well-positioned players are already moving into on-demand education. Online course provider Udacity has recently pivoted from offering full-scale university courses to bite-sized “nanodegrees” which are proctored by their global network of 300 code reviewers. They claim their best code reviewer can earn more than 8x the monthly salary of a part-time teacher in the US. Startups like MathCrunch provide Uber-like math tutoring services at a fraction of the cost of incumbent Tutor.com. Maybe even Uber, who partnered with Levo League in early 2015 to offer brief “mentoring rides” with influential businesswomen, can themselves make a play in the market.

As for myself, I’m excited about exploring whether an “Uber for Education” has any merit given the immense social value it could create. If you have thoughts on the topic, please share them below!

 

——————————————————-

Image sources:

https://mises.org/blog/how-uber-threatens-our-way-life

https://medium.com/@cschultz/labor-goes-independent-cfc60f5e10e9

http://www.connectsafely.org/lets-not-use-21st-century-technology-with-19th-century-pedagogy/

http://studentsforliberty.org/blog/2015/07/30/the-sharing-economy-creates-growth-so-let-it/

http://kernelmag.dailydot.com/issue-sections/features-issue-sections/10958/the-hidden-cost-of-the-on-demand-economy/

http://www.businessinsider.com/leaked-charts-show-how-ubers-driver-rating-system-works-2015-2?r=UK&IR=T

http://www.kumulos.com/2012/10/18/questions-app-developers-should-ask-themselves-backend-as-a-service/

By: Vibin Kundukulam


read more