London Taxi Wars: Uber, Gett, Hailo & the minicabs

, 3,779 words

London’s black cab drivers are famous for having strong opinions. Uber? Hailo? TfL? Pick a subject and you’ll get a sharp answer.

As something of a London cab power-user — having got one nearly every day for the past 9 years — I’ve had good opportunity to use the major taxi and minicab systems in London, and to try to unpick what is happening. I’ve found it striking that whilst black cab drivers are right to feel hard done by, much of their anger is misdirected.

Minicabs & Addison Lee

I got a minicab to work, and occasionally back from it, between 2006 and 2011. Shortly after that, in 2012, the arrival of Uber was so disruptive to my local cab office that I dropped them from my daily journey. First to Addison Lee, then to Uber, then to Hailo, and now, finally, to Gett.

Two things seemed to bring about the demise of the local minicab office. The first was the arrival of Uber. Drivers all over London left the local companies to make the switch when they realised they could earn more, and potentially work less, with Uber. Before Uber’s arrival, drivers would complain about sitting around all day in toxic minicab offices, wasting time waiting for jobs that never came in, and struggling with favouritism or bribery in the office. Some drivers would get the good airport jobs, others wouldn’t. Whilst the kickbacks and the cash-in-hand work was good, they didn’t compare to the benefits Uber could offer.

The second part of the breakdown of the minicab office was Transport for London’s failure to properly manage the minicabs. Unlicensed cabs became a huge problem. As Uber arrived, some of the drivers stopped renewing their minicab licenses. I’d frequently find myself travelling over London Bridge, through a routine police stop, where the police would not bat an eyelid that the cab’s licensed had expired. This is one of TfL’s many failings, and it's particularly unfair to black cab drivers who are regulated rather more strictly, and whose interests were better looked after when they were regulated by the Metropolitan Police. And of course, once a few drivers had got away with not renewing their licenses, most of them in the office stopped. Why would a customer pay to travel in an unlicensed, uninsured, illegal cab? I wouldn’t — so I stopped.

For a long time Addison Lee had been the minicabs’ main opponent, winning against them on reliability. When one called a minicab office and was promised a cab in twenty minutes, that could mean half an hour, 45 minutes, or not at all. Addison Lee would commit and deliver.

I only used them briefly, mainly because the service was horrific. I defy anyone to use their online booking system without becoming enraged by its obtuseness. Signup is convoluted and horrible. Booking is convoluted and horrible. Account management is convoluted and horrible. The founders sold out of the business at the right time, and despite the PR, it seems unlikely its new PE overlords know what to do with it. With a declining 10% share and traction only in B2B, Addison Lee is barely an also-ran in the taxi wars.


Moving away from Lee and the minicabs took me to Uber, which I’d used heavily whilst in the US. For a long time, Uber was stuck in West London, and those seeking a cab in the City were in for long waits. But by early 2013, Uber had enough critical mass to have cars all around the city. Surprisingly, it made for a cheaper journey to the office in the morning, but the London Uber experience is not like the Californian one.

American cities largely have their streets on a grid system, and when one is new to driving it’s pretty easy to figure out, short of painful one-way systems such as San Francisco’s. In London, however, the streets are anything but. Small, windy, at angles, prone to frequent closures, congestion and rearrangement, driving well in London requires experience and good knowledge of the roads.

Uber drivers don’t have this. And the automatic route-finding system that Uber uses is disingenuous. An important route named the Embankment provides the main flow of traffic heading from the City to the West End, but when the Embankment is gridlocked with unmoving traffic, the Uber system will instruct its drivers to take it, promising a 15 minute journey. At the same time, it will be obvious — even in Google Maps — that a driver taking that route will be stuck in traffic for at least 45 minutes whilst other routes are available.

The hallmark of my Uber rides in London is getting horribly stuck in traffic with drivers who don’t know the routes, speak such poor English that explaining another route is difficult, and who are scared to drive any way but the way which the app tells them to. In addition to this, they don’t know where anything is. Bond Street? Greenwich Park? Shoreditch High Street? With no exaggeration, these are the last three places I asked for in an Uber, and in each case the driver did not know of them. Rarely does the car move until you’ve found a postcode. How does one even find a postcode for a park? I’ve never understood that.

Speaking of language, it’s ironic that Uber’s international rollout has created the same problem in the UK that it set out to solve in the US. In the old days, getting a yellow cab in New York would give you a driver who didn’t speak English and didn’t know the city. Uber has solved NYC's problem well, but in London taxi drivers already speak perfect English and know the city backwards. If you summon an Uber in London, you’ll likely have the NYC yellow cab experience: someone who has mastered neither the city nor the language.

Trying Uber in other territories — such as Bucharest, where I occasionally visit on business — it is evident that it’s not just the UK where Uber’s quality is poor. A street cab in Bucharest may rip you off, probably won’t speak English, and will have an old car. Uber passengers won’t get ripped off in Romania, but Uber’s enforcement of their English-language and new car policies is shaky at best.

The power of Uber’s army of angel investors is often overlooked. Prolific writers such as Jason Calacanis would long ago have spoken out about the problems with Uber, but by having the good fortune to have invested, he is conflicted enough to have effectively bought his own silence. A number of influential angels have a lot of money riding on Uber’s continued toppy valuation.

Were this small army of influential angel investors not tied up this way, they might be commenting and poking into three areas.

Firstly, the importance of Uber’s international growth. Many of these investors see the clear value that Uber provides in NYC and LA and assume that a similar, or better value is being created internationally, and as I’ve explained that is far from the truth. Uber services internationally are not nearly as well managed, don’t provide the same quality, and in the case of London and other countries where there are well-regulated, knowledgeable, professional taxi drivers, the core benefits of Uber’s proposition don’t apply.

Secondly, there is an attitude amongst many of Uber’s backers that Uber themselves are best-placed to be the “Uber for everything”. Flowers? Groceries? Parcels? Their argument runs that Uber’s valuation is supported by how readily they can do this. This argument, however, pays no attention to the complex nature of operations and logistics. It is not possible: Amazon is already the “Uber for everything”, having invested billions in building out the infrastructure necessary.

Finally, Uber’s surge pricing is a double-edged sword. Without a doubt, surge-pricing is the best part of the service. Uber works because it is reliable. If one needs a cab, one can get a cab. At any time. Without surge-pricing, this would not be possible. The official line is that surge-pricing is deployed to encourage drivers to take to the road in periods of high demand, or anticipated high demand. But can users rely on Uber to fairly and judiciously deploy this mechanism without any regulation or oversight? There have been enough reports of peculiar pricing spikes, dirty behaviour with the competition, and dirty behaviour with the press to think that there is something morally suspect with the company.

When in the US, where possible, I use Lyft instead of Uber. The service is similar, but Lyft is the morally acceptable face of transport-on-demand, and serves the same purpose. They have a less toxic culture, and provide more interesting encounters with the drivers, who they treat better. Uber’s fairly recent change to encourage their users to subject long-suffering drivers to their passengers’ music fills me with disquiet, and it hardly seems fair to drivers.

So my time with Uber in London didn’t last long. After a series of dirt-cheap but ultimately nightmarish, never-ending journeys, I soon dropped them and went back to black cabs. There's a particularly nasty element which crops up sometimes. If a driver can't get to the pickup, rather than cancelling they will sometimes call and ask the passenger to cancel the job. Presumably the driver is be fined for too many cancellations, and of course, the passenger is also surcharged for cancelling after a short period. Several times I've had drivers say they can't make it -- but refuse to cancel -- leaving it to me to cancel and then contact Uber's support for a refund.

At first I resolved that I’d still use Uber when I wasn’t in a hurry. Over the weekend perhaps. But I soon realised that even on a Sunday, it doesn’t matter how much cheaper Uber are when a black cab could do the journey in half the time. Even had Addison Lee’s campaign to grant minicabs access to dedicated taxi lanes been successful, it wouldn’t make up for the black cab drivers’ knowledge of the right routes to use.

Hailo & Gett (GetTaxi)

In most black cabs, Uber is a dirty word. Drivers will tell you Uber has been stealing their business and competing unfairly with them. I’m not convinced that that is true.

Amongst my friends and colleagues, Uber has found a place with the younger ones. People who would typically think a black cab too expensive. Uber have tapped into this, and if black cab drivers think they’ve lost out to it, they’ll lose out a lot more if these users advance in their careers and continue with Uber rather than graduating to black cabs. Uber’s offer of higher-end cars provides for the social bragging rights that are so beloved of the SnapChat generation. A special treat for a night out once per month? Get a high-end Uber car. Whilst a black cab provides a better service than Uber’s cheapest car, it doesn’t offer any opportunity to upgrade the experience, and the results can be variable.

Migrating away from Uber, I started using Hailo as it gained traction in London. At the time, they had around 2,000 black cab drivers using the system. However, Hailo’s announcement that they’d offer a parallel service in-app, with both black cabs and minicabs led to its temporary demise as a workable solution. It seemed that in the two-day period after the announcement, all of its cab drivers dropped it and uninstalled en masse. Hailo ads were torn down from cab sidings, and it became impossible to get a black cab with the service. Drivers’ contempt and hatred for Hailo’s founders was more intense than that shown towards Uber.

As of mid-2015, whilst Hailo have reestablished their network of black cab drivers and the anger in the taxi community seems muted, Gett have sustained the huge lead that Hailo’s misstep gave them. At the time, I switched to Gett (then known at GetTaxi), and whilst their app was primitive and far behind Hailo’s at the time, it soon caught up.

Gett isn’t too bad. The service they provide is as good as a black cab, but with some caveats. In building their network of drivers, they have clearly colluded with them to serve their interests better than those of their customers, and this is obvious in a few areas.

Firstly, it’s hard to rate drivers on Gett. Part of its promise is that you’ll get a good driver, as you can theoretically give feedback to poor drivers and eventually they will be disqualified from the system. But unlike Uber, where users are prompted to rate their driver even several days after their ride, Gett doesn’t work this way. Unless one actually has the app open as the ride ends, rating the drivers is not possible. This is deeply skewed towards the interests of bad drivers, and does passengers a big disservice.

Gett’s bungling with their user-rating system gets worse. As a regular user of the app, I’m frequently bombarded by promotional messages. I can’t think of another paid service where I have to endure unskippable promotional messages to refer my friends. I spend around £10,000 a year on taxis, yet every week or so they degrade my experience using the app trying to persuade me to refer a friend for a £20 credit before they’ll let me use their service. This is foolhardy and short-sighted marketing, and the best I get for my loyalty is an inscrutable ranking system for customers, which sometimes leads me to get “first class”.

What does “first class” mean from Gett? I do not know, but it seems to mean that when hailing a cab through the system, I’m served another arduous promotional message stating that I’ve been “upgraded” to a better driver or a faster experience. It doesn’t take a genius to work out that either this is nonsense, or that Gett deliberately give their new users below average service such that they can send the higher-ranked drivers to power users. Which companies have been successful on the back of a strategy of giving their new customers worse service?

On the subject of poor service, Gett has introduced functionality to pre-enter the destination information when hailing a cab or waiting for the driver to arrive, just as Uber does. But unlike Uber, Gett broadcasts this information to drivers before they arrive. Pre-enter a destination for a short trip, or say, a trip from the City to Canary Wharf during rush hour, and they’ll cancel the ride, leaving one to hail again. Unlike Uber drivers, there are many journeys black cab drivers don’t want to take, and they’ll cancel if given the chance.

Gett's misleading cab times

Gett’s estimated time to arrive as shown in the app is as inaccurate as Uber’s route time estimation. I’m not sure I’ve ever seen it claim a metric other than 1 to 5 minutes to have a cab, yet rarely can a cab be secured in that window. 5, 10 or 15 minutes is more likely, and some weeks it can be regularly impossible to get a cab at all with Gett. Of course, the indicator will still merrily inform would-be passengers that a cab can be theirs in a minute.

Finally, Gett’s attitude towards capturing more demand from business users seems lacking. They approached me months back about setting up a business account, which seemed like a fine thing. But somehow the business account made riding with Gett more expensive — not less — than it normally is. The days of telephone bookings, elaborate customer service and corporate accounts that cost more are long-since over. All one needs for a cab now is a mobile app, and the staff can easily program in a company card and hail cabs without any paperwork. Gett seemed mystified by my refusal of the account, but I suspect the message they got from the market was near-universal. Recently they approached with a much lower, but still increased cost. They need to cut into their margin on this one; this old-world Addison nonsense is no longer good enough.

The future for black cabs

Whilst Uber has a bright future in London, I don’t think things are as gloomy for black cabs as their drivers make out. They provide a service which is fundamentally better than Uber’s, and that is a powerful thing. As many as a third of London’s 12 million population are overseas visitors at any time, and the famous London taxi is well-known and regarded around the world. Many new customers will adopt Uber, but they are ripe to be pulled back as they grow tired of Uber’s problems, and as the black cab ordering systems improve.

London’s roads are often a nightmare. The hallmark of Ken Livingstone’s run as mayor last decade was terrible congestion, and after Boris Johnson came to power — a large and disruptive programme of replacing Victorian sewers aside — London cabbies enjoyed half a decade of pretty easy motoring. The closing year of Boris’ mayoral spell has brought increasingly dire traffic with the construction of the cycle superhighways, but it is to be hoped that this will subside soon enough.

In my view, there are three areas that black cabs and the hailing systems such as Gett and Hailo need to crack to improve their odds.

Understand who the enemy is

Regular black cab protests against Uber are daft and misplaced. They are invaluable for Uber, as they show potential customers just how mercurial cabbies can be, and they go to underline part of Uber’s proposition: you can get an Uber when you want an Uber. You can’t get a black cab all the time. Gripes with lax regulation aside, Uber haven’t done anything that the taxi driver shouldn’t already have done.

The real villians in this saga are RadioTaxis and ComCab. These are the computerised booking systems used by black cab drivers prior to Uber’s arrival, and they had a clear mandate to deliver this technology. However, their heads were firmly in the sand. They had the technology, the market lead, the data, and a colossal opportunity to do what Uber, Hailo and Gett have done, and nothing, apart from their incomprehensible incompetence prevented them from doing so. Forget Transport for London, Boris Johnson, Uber. None of them have done black cabs anywhere near the damage that these two companies have done.

Even in mid-2015, somehow — almost impossibly! — ComputerCab still doesn’t get it. “Open your account with ComputerCab and book your taxi in 6 clicks”, their current advertising runs. Presumably no-one in their office has bought a mobile phone since 2007, or they’d realise that the rest of the market has a simpler promise: “Book a taxi in a single tap, no need to open an account”!

ComputerCab's latest advert

Understand how and where to compete

Whilst arriving somewhere in an Uber car in a timely fashion may not be reliable, the process of getting the car to come and pick up is. With black cabs this isn't the case. Enter the wrong destination, or hail at a busy time, and drivers won’t take the job. When there’s a lot of demand for cabs, say at rush hour in the rain (common enough in London!) drivers will turn off Gett and Hailo and pick up passengers who hail them from the street.

This gives users a clear message: when one really needs a black cab, one probably can’t be found with Gett. Uber solve this with surge pricing, and by insisting drivers take jobs. The likes of Gett and Hailo need to force drivers to take jobs that come their way. If they can’t bake reliability into their booking systems they face increasing numbers of users defecting, or situations where users simultaneously juggle the hailing process with Hailo, Gett and Uber together. I occasionally find I need to hail with both Gett and Hailo simultaneously to get a cab in City rush hour.

Drivers complain that they need regulatory change or enforcement to block Uber. They’re wrong. They need regulatory change so that they too can use surge pricing. Letting black cabs surge price will bring more of them on the roads when their customers need them, and it will make them more money. It will also provide a regulated surge pricing framework which Uber won’t be able to touch.

It's interesting, then, to reflect where the disingenuity comes in both systems. At Uber they have been focused on reliability and ease of use in getting a car. It's a single tap, with no barriers, to hail a cab in Uber, and the user isn't disappointed until the car arrives and the journey starts. With Gett it's the other way around: once the car has arrived, the journey is pretty good, but the process of getting a car is waylaid by in-app marketing, misleading hail estimates, cancelling drivers, and long hailing periods. Consequently, it is not surprising that Uber's growth and valuation have far outpaced those of Gett.

Root out bad drivers, and let users give feedback more readily

Cabbies' fraternal solidarity with their union brothers has not done them any good. Their unions have failed to protect them from their partners' harmful lack of innovation. Passengers need consistent, reliable and convenient ways of feeding back on poor service. Bad driving or poor knowledge of the routes? Uber may have an intractable problem with the latter, but they closely monitor the former, and black cabs have a clear customer service problem. The likes of Gett and Hailo need easier, clearer, and more visible mechanisms for giving and addressing feedback.

Of course, this is just an opinion. I'm just a customer, not an investor in this space. If you have a contrasting view, I'd love to read it.

August 2015 update: I wrote a follow-up to this piece with the best of the comments and news I received since publishing this. See the London Taxi Wars revisited.