One of the oldest, and often most contested, debates in the world of entrepreneurship research concerns the most basic definition of the field: what, exactly, is meant by the term entrepreneur? In its original use by Jean-Baptiste Say, it was someone who undertook economic activities and capitalized on arbitrage opportunities. Joseph Schumpeter ushered in the modern way in which people typically use the term by equating it with newness--new products, services, combinations, business models, etc. Israel Kirzner saw entrepreneurs as those who targeted and eliminated disequilibria in the economy (for Schumpeter, entrepreneurs created those disequilibria).
By now, however, the word has come to be so overused as to potentially lose a great deal of meaning. Those who navigate the world of new technology companies that receive venture capital or angel money prefer "startup" to entrepreneur, or sometimes simply "venture." This often connotes a company that is aiming for scale or growth from day one. You can find "Startup [Anything]" these days, it seems, all over the world. A related term that has gained currency is "founder," used to distinguish individuals/teams and their ideas or businesses. And, of course, there is the constant tension between those who use "entrepreneur" to refer to small businesses that stay small and those who refer to small businesses that grow quickly or to larger sizes. On these points, Erik Hurst and Ben Pugsley delivered a useful discussion and taxonomy in their paper on the heterogeneity of the world of small business.
So, does it matter? Our daily language is likely full of various words and phrases that are used by people to express different meanings. This is unavoidable and, in most cases, inconsequential. That's just how language is. With regard to entrepreneurship, there are probably some poor outcomes from linguistic slipperiness. Policymakers at all levels probably have different things in mind when they consider something like the JOBS Act or a "small business tax cut." The consequences of confusion (or willful ignorance) in those cases could be undesirable.
I've been playing with a draft of a paper for several months in which I try to put numbers to the question, "what is a new business?" There is a lot of what we would expect: construction, health care (doctor's offices), restaurants, retail, business services (law, accountancy, etc), and so forth. The distribution of sectors has shifted a bit over the past three decades, but not dramatically, and we predictably see sectoral bubbles. There were lots of real estate and construction companies started during the mid-2000s, for instance. This isn't really surprising. The needs for shelter and food and clothing are not going away anytime soon, so it's not really a shock that we see lots of new businesses starting (and failing) in those sectors. There are likely shifts in the types of restaurants and retail stores, of course, but those remain within the broad sectoral classifications.
Many of these types of businesses remain small (construction is, on average, one of the smaller sectors by employment per firm), but we do see a good number of very fast-growing firms among these sectors as well. This is one of the hallmarks of the Inc. magazine annual list of the 500/5,000 fastest-growing companies; they are not just concentrated in technology sectors per se but come from all areas of the economy. Many seek to apply technological disruption to sectors in need of it. Businesses in some sectors, moreover, such as restaurants and retail, often need to start with a minimum number of employees. A web-based startup can start with two or four people--and get to just over a dozen before being acquired for $1 billion--but a restaurant or a retail store, depending on the type and maybe the geographic area, have minimum employee needs.
(I suspect that this sectoral distribution of new and young firms partially accounts for our finding that new and young companies constitute such a large share of net job creation. I also suspect that accounting for sector (and geography) will explain a lot of what we think we know about entrepreneurship, but I hate to take it to an extreme and say that X explains everything.)
The most useful way I have found to think about all of this is as a spectrum, most recently, and eloquently, expounded to me over dinner by Philip Delves Broughton. (And Jared is correct: immediately add The Art of the Sale to your reading list.) On one end you have the self-employed and non-employer businesses; on the other end you have the "startups" and "founders" with "frighteningly ambitious" ideas. The space in between is occupied by small, local businesses; fast-growing companies that nonetheless serve only regional markets; franchises; $100 million revenue businesses that occupy small but important and profitable economic niches; innovative web and mobile applications that will remain small; and so forth.
For better or worse, the word "entrepreneurship" has become an umbrella term for all of this. And it seems silly to argue over which type is better or more economically important or more virtuous than the other kinds because, of course, the economy needs all kinds. That is what makes it the economy.