The role of experience in the fall of Autolib'

After 7 years, Paris' famous electric car sharing service Autolib' suddenly ceased to exist. What went so wrong that this once much-praised pioneer service was dealt the death blow in just a matter of days?

The end of Autolib

From fame to debacle

Autolib, the famous electric car sharing service launched in 2011 by the Bolloré group in Paris, was once the biggest one-way car sharing service in the world, exhibiting impressive figures: 4000 cars deployed around 1100 self-service docking stations throughout a 665 km2 metropolitan area, serving a total of 155,000 users, of which 110,000 yearly active subscribers, undocking a car every five seconds and travelling some 50 million kilometers per year, all this with an amazing 92% satisfaction rate!

The revenue model of Autolib was built on a combination of yearly subscriptions and trip fees and should have broken even with 80,000 subscribers using the service on a regular basis. And yet, Bolloré recently turned against the Paris region municipalities for whom it ran the service under a public service delegation contract, asking them a mighty €233m to make up for its budget shortfall. Not only did the latter refuse to bow to that demand, they in fact decided to discontinue the service altogether!

How can this be? What went so wrong that this once much-praised pioneer service was dealt the death blow in just a matter of days?

The dominant explanations

Everyone, from the press to actual users of the service or even people who probably never stepped into one of Bolloré’s Bluecars has had his or her say in the last couple of days, providing a substantial exegesis of this true disaster.

Autolib didn’t make its revenue numbers because paying subscriptions started to stagnate and even decrease in the last years, but above all because the number of trips/subscriber dropped substantially. Several reasons have been advanced to explain this: 

  • The number of trips/subscriber fell simply because an increasing number of users also meant a decrease in the number of available vehicles;
  • The number of trips/subscriber fell because of the dirtiness and poor maintenance of the vehicles;
  • The one-way architecture of the service promoted a complementary use rather than a substitute to other mobility solutions, making it intrinsically more volatile and thus very sensitive to a competition that did not existed when it launched but is now booming (1);

At the same time, Autolib saw its costs growing well above initial estimates:

  • The one-way service architecture made the service too costly to operate in such a vast conurbation as Paris and the 102 communities participating in the service (cost of relocating the vehicles);
  • The public service delegation contract made it impossible for Bolloré to manage the service effectively (closing, for instance, unprofitable stations).

All these statements are true, but the main reason behind this debacle is to be found elsewhere.

The people and experience view

The main motivation for Bolloré to create Autolib was not to become a successful car sharing service operator but to build a vitrine for its unique Lithium-Metal-Polymer batteries. Born out of technology, literally developed around the battery in just a few months, the attention and consideration given to future users never went beyond making the service just work. As a result, Autolib ended up being a rather complex system relative to other solutions, but one that indeed did work, delivering great functional value for its users. 

In contrast, the Paris region communities behind Autolib looked at it as a way to improve the livability of the city by offering its citizens a complementary mobility solution, accessible, affordable and clean, when and where relevant. In other words, not a service that just targets a stable base of regular users but one that aims at supporting an ad hoc use of the service by the largest possible number of people (including – why not? – tapping into the unique touristic potential of Paris to support those travelers who would have been keen to experience the thrill of driving around in the City of Lights).

But this would have required to design a service experience that drives interest, promotes engagement and trial, supports first-time and one-time use, while of course doing its best to encourage repeated usage and promote positive word of mouth.

Not one that just works. 


To understand why the actual service experience failed to achieve any of this, one needs to look at a few facts we know about people's behaviors, attitudes, needs and expectations towards mobility:

  • People exhibit increasingly unplanned, extempore, ad hoc mobility behaviors;
  • People expect instant service access & availability ;
  • People expect zero-threshold experiences, as simple as possible and involving as little learning as possible;
  • People expect to be emotionally engaged, not just being provided functional relevance;
  • The fact that experience is being increasingly valued over ownership doesn’t mean that people’s needs for self-expression have disappeared!


In view of the above, let’s now take a look at the Autolib experience:

  • Discovery of the service takes place primarily through Autolib’s Bluecars, whose characteristic and uncommon grey has made them stand out from day one as dull and utilitarian rather than inviting and creating an urge to step in and be seen in them. So much for the emotional engagement and self-expression expectations! (2);
  • Engagement towards the service takes place solely through its digital channels as the physical elements of the service, the cars or the stations, have not been leveraged to support this crucial phase. And these digital channels make a very bad job at it, doing nothing to make the service easily accessible (what, I need a subscription?), drowning the disconcerted user with unnecessary details (different subscription plans, price and distance calculators, maps) and providing no clue as to what one needs to do to get started and just give it a try. So much for the zero-threshold expectation…
  • Activation is in fact the first big hurdle on the Autolib journey, requiring users to discover that they need to physically go to specially-designed registration stations, realize that there are just but a few of them scattered around the city, and make the actual effort of going there. So much again for the zero-threshold and the instant accessibility expectations. For those that do make the effort, the experience at the stations, however, is quite well handled, through a live video dialogue with an Autolib employee guiding you through the process, scanning your documents or answering your questions;
  • Once registration is completed comes finally the time to use the service for the very first time. And this first-time use of the service is yet another major obstacle to overcome. Information is scarce and, above all, fragmented. No support is being offered to be guided and accompanied, step by step, through a journey whose perceived and actual complexity is objectively extremely high, especially on first-time use: from booking a car on the web or through the mobile app to going and finding the station, identifying and authenticating at the station, finding and undocking the car, opening it, discovering the unconventional interior layout, finding the hanging key to start the car, starting it, wondering if the car did indeed start in view of the suspicious lack of engine noise, overcoming the apprehension of driving an electric vehicle (for most users, this first experience with a Bluecar is also in the vast majority of cases their first experience with driving an electrical vehicle altogether), driving it, discovering and understanding how to interact with the car (HMI, navigation & infotainment system), booking a parking place, navigating to it, docking the car back and getting confirmation of the successful closing of the rental. Not to mention the potential risk of having encountered a situation where no car was available or, on return, no parking place was free, of having picked up a dirty car and/or one with defects (broken back mirror, inoperative navigation system, etc.) coming in the way of a pleasurable first-time use. All in all, so much again for the expectation for zero-threshold, instant accessibility, instant availability and being emotionally engaged in a positive way instead of being put through a roller-coaster of feelings…

Returning users, however, are being rewarded with a growing love for the service as they start feeling increasingly at ease with it, as perceived and actual difficulty decreases and as they discover such simple yet hidden gems as a radio system that remembers their favorite radio stations or a navigation system that memorizes their frequently visited or favorite addresses… 

In conclusion

In summary, the Autolib experience is thus one that, technically, did work, but one that remained an objectively very complex one, in actual and, worse, in perceived terms, effectively deterring the vast majority of potential users to even consider giving it a try. 

Could have it been improved to attract & grow a much larger user base to use it on an ad hoc basis? Yes. Are these observations of relevance for its operations in other cities or for other car sharing & other mobility services? Definitely yes! 

Mobility is first and foremost about people. Vehicles, infrastructure and services are just a means to an end: to carve experiences that align with people’s behaviors & attitudes and meet their needs and expectations, both functionally and emotionally. 



1. Competition gradually built up with the arrival on the same year that Autolib was launched of Uber, soon followed by other ride-hailing services such as AlloCab or ChauffeurPrivé, direct and often more flexible car sharing solutions such as Zipcar or Ubeeqo and, more recently, Blablacar’s new ride-pooling service, free-floating motorbike scooter services such as Cityscoot or Coup, not to mention Lime’s standing scooters that just debuted a couple of days ago!

2. In the word of physical products, and especially cars, aesthetics play a fundamental role in creating attraction and desire. The Dallara Stradale, recently launched and designed by Granstudio, is a testimony to this, making every head look round at it, eyes filled with delight and desire!

(Xavier Blanc Baudriller)

Article published in
June 2018