Whats IS a Startup???

The logical path while starting writing a research paper about startups is to first and foremost define the word itself: “STARTUP“.

While researching, I realized there is no “official” definition, and that some “thinkers” can have different point of views.

Olivier Ezratty : “A startup is a company that is experiencing a significant growth”.
Likewise, Paul Graham states “A startup is a company designed to grow fast. […]the only essential thing is growth. Everything else we associate with startups follows from growth.” Graham and Olivier Ezratty both base their startup definition on growth, but Graham sees a growth potential, whereas Olivier Ezratty a significant growth : forecasting to experience a significant growth is not actually experiencing growth. Their point of views differ on the development stage ; Olivier Ezratty considers startups once they actually experience growth, whereas Graham considers startups through their future potential.

Eric Ries has a different approach, as he considers that any project of product creation is, by nature, a startup. He states “A startup is a human institution designed to deliver a new product or service under conditions of extreme uncertainty”. By “extreme uncertainty”, he means all the factors that make future uncertain, such as competitors, consumers behavior, financing… Eric Ries takes in account the newness of the startup’s products or services in the startup definition, and encompasses all types of value creation. Innovation is hence in the core definition of a startup according to Ries, but we need to understand innovation broadly, as it can apply to the product itselft, a new kind of service, or an innovative kind of business model.

Stokes & Wilson (Stokes and Wilson 2006) consider a startup as “a new enterprise, starting up and trading in its chosen field”. Therefore, they consider startups as newly lauched enterprises that have just started up, unlike Graham, who states “being newly founded does not in itself make a company a startup.” And yet, the Oxford dictionnary states a startup means “the action or process of setting something in motion”, and gives this example : “the start-up of marketing in Europe” (Oxford Dictionnary Online nd) Setting something in motion means starting action, moving from nothing to something that is actually moving, as the antonym of motion is stagnation. We can hence understand how most of people, especially French people, can understand the word startup as newly launched enterprises, because the litteral translation gives “entamer, démarrer, commencer, initier, créer, lancer, fonder” (Word reference online nd), all words that actually mean starting something up.

Graham considers that startups differ from enterprises, newly launched or not : they have a different DNA from the beginning of their creation. He considers that if one is willing to create a startup with high growth potential, he/she will not launch “a barbershop”. In other words, one potentially knows if the type a enterprise he/she is launching is likely to be a startup or a small business. Graham says that in order to potentially experience high growth, you need to “make something lots of people want, and reach and serve all those people.” He explains that everyone needs a barbershop but no one will travel to reach this specific barbershop, and that is why it will not be a startup. A barbershop is not scalable. Startups are companies fundamentally designed to scale, according to Graham. He sees startups growth as an S curve with 3 phases : first period of slow or no growth (but having a high potential), second period of growth while the startup figures out what to sell and how, and the startup will eventually become successful in a third period where its growth will slow because the company “is starting to bump up against the limits of the markets it serves.”

Steve Blank proposes a definition comparable with Graham’s but still different from Eric Ries’ : “a startup is an organization formed to search for a repeatable and scalable business model”. A “scalable startup” takes an innovative idea and is constantly seeking a repeatable and scalable business model that will drive it to profitability and growth. This meets Graham ideas quoted above, who said that the barbershop is not scalable unlike startups. Eric Ries’ also integrates innovation in his definition, but does not evokes the notion of a scalable business model. Oussama Ammar agrees with Blank, and even states that a startup is not an enterprise. There is a fundamental difference between both, as a startup is a social entity that seeks for a business model, as an entreprise operates, exploits and optimize an existing business model. Ammar insists on the fact that a startup is not a “miniature” of an entreprise, the size does not impact the startup definition. Oussama Ammar recalls that a SME is not always a startup, and he agrees with Graham saying that a recently launched enterprise does not has to be a startup. Therefore, they disagree with Stokes and Wilson stated above.

The chart below has been designed by Steve Blank, and emphazises his vision of the definition of a startup.


On this chart, we understand Blank’s definition of a startup. It illustrates how a startup could become a company, through a transition from searching a business model to actually execute it.


About me

Capture d’écran 2013-04-28 à 13.35.42


I am Charlotte Rambeaud, a French International Business master student at Grenoble Ecole de Management. As part of my master degree, I am required to write a final management research paper. Highly captivated by the Internet sector, I am intern at Criteo in FP&A, and chose to write my research paper about the core reasons of web startups failure.

My aim is not to write theory about startups failure, because of lack of experience and wisdom, but I just want to focus on about 20 web startups that did not succeed, understand why, and to be aware of those entrepreneurs’ current state of mind. I am passionate about entrepreneurship and would like to learn from those who have already been brave enough to try.