My mentor once told me that writing enterprise software is a viable option to pursue as a startup. We both tried to identify good enterprise problems to solve. My choices, since, have brought me to Singapore and our paths have significantly diverged today.
Neither of us is running a startup today. I work for a pre-seed bootstrapped startup in Singapore. Technically, I am building the product while the founder is building the pipeline. We are reaching out to small and medium enterprises as potential customers. In the process, we reached out to one large enterprise as well. We heard back from them, much to our surprise. But the response was a complete letdown.
I have worked with large enterprises in the past. I have some understanding of how things work there. Policies guide the business. An IT policy guides the software purchases. It took me a while to appreciate the policy idea. I acknowledge it, but I don’t revere it. In my opinion, the policies are just guides. We should follow them in spirit and make our decisions. Normally, an informed decision will align well with the policy. When the policy becomes the bible, then the human becomes unnecessary.
The IT policy for most large enterprises would explicitly require the use of Microsoft products and solutions. I’ve never understood this requirement. A former boss of mine, much later, would explain to me that if they went with a non-Microsoft product, they would not know who to approach when something breaks. He used to joke about how every engineering manager in his day would blindly recommend IBM products and solutions.
I can imagine a few scenarios why an enterprise would prefer to use a Microsoft stack as opposed to, say, a Linux stack.
- Contractors run almost all of the IT operations. It is easier to find a Microsoft certified administrator than to find a good quality Linux administrator. Besides, most Linux administrators I’ve met were very opinionated. Opinions do not get along with policy very well.
- It is difficult to ensure access to high-quality administrators or engineering managers to make sure the IT operations remain smooth all the time. It is easier to hope that Microsoft support can train and maintain a high-quality support team that you can commission as needed.
In theory, the same argument applies to all open source software. I will expect an IT policy to warn against the use of open source software. This is where the human factor should prevail. But when you get an email from an IT person categorizing open source software as vulnerable as opposed to others, it is a letdown. I am no security or vulnerability expert. Security, as far as I understand, is an exact science. Every single software is just as insecure as any other. It takes a lot of consideration and effort to make any software secure enough to remain useful.
In my opinion, this person has already rendered himself redundant.
On the other hand, I seriously wonder if my mentor missed some vital point. I am sure he mentioned building enterprise software. But building software for a large enterprise is an uphill battle or a capital intensive operation. While startups have enough problems as is, should they also signup to educate and evangelize enterprises?