And there are thousands of such niches. Millions, really. They're all around, everywhere. Find any office, anywhere, in any business. You will find some repetitive-but-business-critical proces being performed by someone following a yellow photocopied guide taped to the wall above the fax machine.
but, scaling up is difficult.
start in the wrong country and you have a big barrier. you need to start in the US, no way around it. single biggest homogenic market (language, culture, legal). your 3 sales people can cover a lot of ground. start in France and great, once you have France where do you scale out to? Germany? The US? Your French sales people are likely worthless outside of France.
You see this pattern in enterprise companies. Way fewer in central Europe than there should be. Germany's SAP made it, Software AG follows. Cegedim in France. UK has advantage of US-ties. Eastern Europe? Kill yourself (outliers: Skype, Kasperksy). Qliktech started in Sweden and nearly died. Only after it was re-located to the US it kicked off.
It's easier to move into the US market from Australia, for example, than it would be from France. Common language, lots of shared history, ease of travel, all that stuff.
I'm working on a niche business tool at the moment and it's at the back of my mind that upon moving into the middle tier, I'll need to think about having someone in the USA simply to act as level 1 support.
That said, Patrick McKenzie of patio11 fame pointed out that he's using a virtual assistant service to fill that role. That might be my first port of call.
/sorta-kinda-maserati-problem
I guess it would be the same with support. The only job I've worked in a tech support role in, only had enterprise customers in Europe.
Another point, is that usually each company have diferent processes, and it's quite difficult to make a one-size-fits-all (as in facebook/twitter/google). Intranets, and biz software usually need people to make custom modifications on each company. Usually the people in charge of this changes are not as competent as the ones who write mainstream software and you end with a lot of crap, which is almost impossible to stop due to the fact that there are very few good programmers around to make custom changes to every company in the world.
To sell to corporates you should find close friends among those decision makers.
In enterprise, the customer is the customer and pays for something. Contracts, SLAs are in play. Who is the customer for Facebook?
In enterprise, the purchasing decision is not made by the end-user. You choose to use Facebook or Dropbox. Some middle manager, sometimes a C-level one made the decision to use Sharepoint for you.
In enterprise, support is expected. 2nd level, 3rd level, something. No support means no contract.
In enterprise, country borders are real. You want to roll out in Spain? You need licenses that cover this, need to define support, need a way to roll out there (requirements, implementation, etc.) and very often you need localization. And, if applicable, you have local data security laws in play.
Essentially, in enterprise:
1., You need sales people. Highly paid sales people that bring in their networks. Your website is worth nothing. 80% of your success in enterprise is sales. Your product is simply not that important.
2., You need to know how to build "product" for your initial target group - the people making purchasing decisions. Which can be diametral to what the end users actually need. You need shiny demo features which do not ruin your product in live use.
3., You need configurability. Nobody, ever, in enterprise uses something out of the box. The beast with many names (Accenture, CapGemini, Deloitte, etc.) needs to be fed with the blood, sweat and tears of hapless victims caught in workshops and UATs. The iPad has disrupted this cycle so profoundly it is amazing to watch. No pure software solution has achieved this so far though.
Dirty little secret: A team of 5-8 top-people can implement and roll out any global system. Reality? Hundreds of various consultants running around creating nothing but confusion. But why does this irrationality exist? Because global companies hire their consultants to be middle managers or even be C-Level IT officers. The business model of the big consultancies is practically HIV - penetrate the customer, infect the host, lower the defenses, repeat. The various alumni orgs of those firms form strong bonds. Look into the LinkedIn history of a IT manager and you know which consultancy is being hired, no matter how bad projects turn out to be.
As stupid as it sounds - enterprise is for grown ups. Garage start ups created by college drop outs have a competitive disadvantage as it is not the product that counts, but the right, expensive sales people. Seniority, experience, personal networks are critical.