How to develop “Successful Software”

I am sure different people have different perspectives, criteria, and ways to do it. Here are some:

Is it one that generates the highest revenues? But it heavily depends on marketing and sales. The one that makes the best profit? Same dependencies plus cost management. Has the highest number of users? Well… is government software successful? The fastest time to adoption? We could argue about the initial Obamacare marketplace. Is the leader in the market? Not if it has extremely limited or no competition. Has the most positive impact in society by improving people's lives? Then we go to the realm of ethics or philosophy.

I’ve been working in software engineering, as a developer or and also as a user, for over 30 years, most of them for the Oil &Gas industry, employed by the best software and services providers. Due to my experience and background, I could be biased to define a “Successful Software” as one that users want to use and are not forced to use, one that evolves without having to make but minor changes to the original platform and provides a platform that allows adding the most diverse functionality even to user groups that were not the original target. Granted, changes in technology force in some cases radical changes, but bar that, I stick to my definition.

I designed and developed my first system, all by myself, for a civil construction company. It was developed when DOS was still king and the only change I made to it in 10 years of uninterrupted service was enhancing the currency fields because the inflation in that country spiked in those years, it was not until the customer needed to upgrade the computer and didn’t want to pay my fees to reprogram it for Windows. The users adopted it immediately and the system was fit like a glove for them, saving time and improving building projects monitoring, which minimized conflicts with their customers and improved their margins.

More recently, in 2014 I lead the release of directional drilling operations data management system (I named it 3DMS) for my employer at that time. I designed the entire system (UI/UX, data model, security, business rules, functionality), managed the tender, lead the project, and was the SME. It was developed offshore and was completed on fixed price and time and was adopted within a month in the entire company (at the end of the month the incumbent software was completely decommissioned) and has worked uninterrupted since. The platform has allowed expanding its functionality to several other divisions and user groups, including finance and human resources.

I have also witnessed failed development projects where I was only an end user. As a Product Manager, I was able to learn some of the criteria that prevent some systems to achieve greatness. In the next posts, I will explain my software development principles. The ones that I try to instill in the developers that worked for me that allowed us to achieve that success.

I’m convinced that the criteria I use to define “Successful Software” not only works for the O&G industry with its very particular dynamics, but for any software. In fact, watch Google, Facebook, LinkedIn, WhatsApp, Twitter, and their brethren. Those, which in fact define success by revenue, can only achieve it if they pile up users by the millions who do it only willingly and not because they are forced to. I think we can learn some lessons from those highly successful companies.

Stay Tuned …