Entrepreneurial story: the creation of U.[Lik], the virtual library
For my course of Entrepreneurship,I had to interview an entrepreneur to understand the major obstacles to overcome when creating a venture. Raphael Labbé, one of the co-founder of U.[Lik], has taken the time to share with me the story of U.[Lik] from an entrepreneurial point of view. A lot of Tech It Easy readers probably already know U.[Lik], and this article is not about U.[Lik]’s features or the concepts of social cataloging and collaborative filtering that Raphael contributes to democratize, but more about the creation of the venture in itself. Sorry for the length of the article, but it’s hard to shorten an entrepreneurial adventure!
Background about the entrepreneur: Raphael Labbé
Raphael Labbé is 28 years old and graduated from HEC Paris in 2003. After his graduation and a first venture attempt (already on the entertainment intermediation field), he started working in a private equity fund of funds, which gave him good insights about the relationships between financing sources and entrepreneurs. In 2005, after a long talk with his HEC alumnus friend Mathieu Léronde, they decided to work together on a recommendation system. After a long period (nearly a year) of study on recommendation algorithm and a deep analysis of the new web trends (UGC, SN, Long Tail that turned out to be Web2.0), they went back to the other side of the fence and created U.[Lik] : the virtual library.
Identification of the opportunity
During their 3 years at HEC, Raphael and Mathieu benefitted from 2 aspects of campus life:
The cultural enrichment due to the possibility of going into some friends’ room to discover and borrow new CDs, books and DVDs;
The chance of having access to a high speed Internet connection and a computer network allowing the emergence of P2P sharing long before the mainstream usage, which gives them a glimpse of what the future will be.
When they met again a few years after having left campus, they both felt that this kind of cultural sharing was much more difficult in professional life and that there were few solutions to provide with good recommendations on what to buy. One of them is Amazon’s recommendation engine, which is a sales booster but not really user-customized. Raphael assesses: “It’s far from being like the trusted librarian or friend who knows your tastes […]. Also, my purchase history is not a really good proxy of my tastes: I dislike some CDs I ordered and many books I bought were intended as a gift for someone else. Amazon has actually few social / community features that would entice users to rate items and share them.”
This perception of an unsatisfied need, associated with a strong interest for cultural products -Mathieu’s thesis was about dematerialization of culture – and the ideal objective of connecting people through their tastes, pushed them to create U.[Lik]. U.[Lik] allows users to share their tastes by creating their own online library where they can aggregate their ratings of cultural products (music, cinema, arts, people…). The items users rate are placed on their Hall of Fame (what they can’t live without) or on their Hall of Shame (what they deeply hate) based on their rates.
This virtual library enables users to share their tastes, to meet friends with the same tastes, and to benefit from U.[Lik]’s recommendation system which suggests items users are likely to enjoy considering their tastes.
The business model is based on transaction pay-per-action or pay-per-clic with e-merchants websites: e-merchants are always in need of qualified traffic and are ready to commission on sales this traffic of potential buyers.
First steps of the venture
With this idea in mind, Raphael and Mathieu had to start developing the algorithm allowing U.[Lik] to provide relevant recommendations in relation to users’ ratings. Fortunately, complementarities between the two partners helped a lot: Raphael had a strong background in sociology and Mathieu in mathematics, which allowed them to design the most relevant algorithm.
For example, whereas on all other recommendation systems on the Internet users are invited to rate products with a mark from 0 to 5 (mainly stars), Mathieu and Raphael decided to make users rate cultural goods by allocating these products to one of the 5 categories of their library (Hall of Shame, Dislike, Interest, Like, Hall of Fame). This system forces users to make a significant choice when they rate a product (socio-psychological aspect) while allowing the algorithm to be more efficient when providing recommendations (mathematical aspect).
However, they had no real coding knowledge and had to learn really quickly to develop the first version of their website, as they needed users to start rating products in order to adjust their algorithm. Timing was crucial as network effect plays an important role in this market. But after a certain time, they divided their tasks according to their area of expertise: Mathieu has developed strong coding skills and keeps on working on the technical side whereas Raphael focuses on the marketing and financing sides.
Finding financing: overcoming the timing obstacle
Raphael had the chance to intrigue his previous managers at Finama with his ideas about cultural intermediation, so they followed his entrepreneurial adventure closely and invested at the beginning to help him start his venture. But finding more financing became rapidly crucial to recruit some talented developers and before all, to get an office.
Raphael started looking for business angels. He didn’t only want to get some money but also to add some expertise to his company and build relationships on the long term with people able to really understand the U.[Lik] project despite the early stage of the realization of the vision. Therefore, he approached only business angels having a strong knowledge of fields related to U.[lik]. He finally managed to get on board a few high profile business angels who gave him the money to get an office and recruit the first developers, and also provided U.[Lik] with more exposure. He notices that the deals were based almost only on the team and the fact that Raphael and Mathieu had executed everything by themselves so far, more than on the business plan: at the business angel stage, it seems to him that the human side counts much more than the economic one.
One of the most difficult decisions Raphael had to take was to refuse money from a business angel who had promised €50K and then lowered his offer to €5K. The team and the other Business Angels strongly believed that it was better for the project to give equity only to people ready to strongly back U.[Lik], without half-measure.
One month after having moved into his office, Raphael started a VC round to get money to build his team. The website he had to present was everything but polished, and the value of the venture relied at the time on the perspective of evolution and the capacity to understand Raphael’s vision for U.[Lik]. But the timing was wrong: the concepts of long tail of cultural products, social cataloging and social shopping didn’t mean anything for VCs at this time and U.[lik] was probably too much in advance. The most striking example is one of his meetings with a VC where he explained that he wanted to be able to illustrate with videos all the cultural products that would be inside U.[Lik]’s database. The VC thought that this was completely unrealistic because Youtube was just starting to boom on the other side of the atlantic and VCs couldn’t imagine that so much content would be available so soon. In a word, understanding the venture required being familiar with some nascent concepts and being visionary enough to evaluate the impact of these concepts on consumption of cultural products in the future. Clearly, it was too much at the same time and this round of financing was far too early considering the maturity of the market.
Raphael went back to see VCs one year later and expects to raise money before the end of the year. He received better feedbacks this time and VCs seem interested in participating in this still early project, that Raphael compares (with a smile) to “the creation of a cultural Babel”.
But Raphael still faces major obstacles. The difficulty to do a seed financing round in France and the risk-averse culture make it difficult to trust two HEC alumni (with a business background) to create a technology. On the other hand, he is still confronted to some incomprehension about the project from some Business Angels and VCs who try to force him to adapt U.[Lik] to reduce risk. This involves some tough decisions to break some deals because the team and its backers believe in the project as it is and don’t want to alter its potential.
Building a team with limited resources
Staffing was another challenge Raphael had to overcome when building U.[Lik]. He has decided to recruit only active users of U.[Lik], so that he’s sure that these people are strongly motivated by the company, understand the value of the project and share the same vision. This is how he recruited the 2 content managers, in charge of animating the U.[Lik]’s community and to manage the database of entertainment items on U.[Lik].
But it was much more challenging to recruit technical people, essentially developers. Developers are not really in the target of U.[Lik] (centered on culture, with women being a large part of the user base) and it was harder for Raphael to find developers strongly motivated by such a cultural project. Moreover, the shortage of good developers in France has led to an increase in wages, and Raphael couldn’t compete with other big companies’ packages. So he lost some developers because of their low motivation for the project and their focus on high wages. As he really needs some technical profiles to develop his algorithm, he hopes that the next round of financing will provide him with the resources to offer competitive wages.
Key success factors: Mathieu and Raphael expertises and involvement of U.[Lik] users
Key success factors are probably Mathieu and Raphael’s executed work and analysis of the market, as well as the core community of U.[Lik].
Raphael has become a reference when it comes to social web. His analyses of the phenomenon and his vision about possible evolution have made him a privileged insider of the all social phenomenon. At the beginning, this extensive knowledge was probably a drawback as he recognizes that he was probably in a “bubble” and didn’t manage to understand people who didn’t get the U.[Lik] concept. But now that the notion of social web applications has emerged and been democratized thanks to popular websites, this knowledge is a strong asset, especially when dealing with financial partners.
Mathieu has managed to create a great algorithm proposing relevant recommendations to users and has coded a very advance UI (that needs to become more use friendly) that is able to foster contribution and enables a very quick way of navigating through an ocean of content (that will be, at one point, fully available online)
Another key success factor is linked with the product himself. U.[Lik] is a strongly viral and addictive product: once you start rating some products on the website you can’t stop, and you want your friends to join the service so that you can see their tastes and get better recommendations. This characteristic has helped to expand greatly the user-generated database of entertainment items and starts to become viral. Some heavy users have largely contributed to the expansion of the database at no cost for U.[Lik]; other users have even developed U.[Lik] plug-ins for Firefox. This kind of outside help at no cost for the company has been priceless for U.[Lik], and is possible almost only in the particular sector of social web. It is something that is difficult to take into account, even if this kind of online behavior has been put into lights with example like Yelp, flickr or wikipédia.
Results to date and next challenges to face
Raphael and Mathieu have managed to create a great algorithm proposing relevant recommendations to users. They are adding social features along the time (for example matching users’ tastes to propose them common recommendations, which is useful when you want to decide which movie to watch with friends). But more concretely, U.[Lik] has 21K subscribers and is now a database of 216K references, with more than 800K users’ ratings. With 500 items added everyday by users, Raphael knows that the database of reference items to rate is now growing autonomously thanks to users’ contribution.
The next challenges will be to reach a critical mass of users and to build a technical team to guarantee the development of new functionalities.
In terms of financial objectives, U.[Lik] has just started implementing its 2 different models of monetization: pay-per-clic on the French website and commission on sales on the US website. Raphael is waiting to reach a critical mass of users to start optimizing the monetization model, which will mean determining which model is the most profitable on which kind of web pages. But by starting monetizing the website already, he proves to VCs that the model is viable and has an exponential growth potential.
U.[Lik] is a great example of quite a pharaonic project (creating a platform of tastes sharing) driven by the identification of a basic need and by the strong belief of two entrepreneurs that it is possible to fulfill this need even with limited resources and the necessity to overcome the major obstacle of being in advance on the market. So, good luck to the U.[Lik] team to achieve this goal!
If you want to discover my tastes, visit my lounge here. Christmas is approaching, so please also check my wishlist 
.
Related posts:
Entertainment, Entrepreneurship, Europe, Friends, Internet, Organization, Project Management, Software, Technology, Uncategorized, Venture Capital, Video, innovation, user-generated content, web2.0, web3.0
The concept is excellent! I love it, and this system has a great potential, particularly in the forecasting business. With the list of films in my hall of fame, I can imagine later on a recommendation system what can I watch on TV tonight, see on cinema next week,etc.
In my opinion, if they want to enhance their site, they should focused on:
- the user interface
- the domain name (why not negociate the youlike.com)?
Congrats to the founder of u-lik.com!!!
@Fidji Thanks for this article. It was fun to take the time to look back at what we have done over the last years.
@Morgan Thanks.
UI is on the way. The aim is clearly to reach your TV or your mobile at one point. Oh, and if you have the domain name I will be delighted to have it ;-D
[...] can find the story here on Tech IT Easy. Be careful, the article is really long, but how can you shorten an entrepreneurial [...]