Every single point he made is crucial for a startup to succeed. I know - I spent 14 years bootstrapping my software company (exited last year). I think this art

Every single point he made is crucial for a startup to succeed. I know - I spent 14 years bootstrapping my software company (exited last year). I think this article is excellent. While ALL his points are valid, I want to differentiate between two specific points he made. 1. "If you’re not talking to your users, then you’re simply guessing at what they need and want from your business. This is slow suicide." 2. The section called "8 — Building Too Much Stuff" When you talk to your users, you WILL get that feature list one-mile long. If build that list, you'll end up with a product that doesn't have an opinion. Your product won't be targeted at anyone specific. Why should a customer pick you over competitor X? So, to know what to build, you must talk to your customers and know them. But you must know yourself as a company - why do you exist? What kind of customers do you want (size - small/med/enterprise, industry, location if applicable). For example, define your company as "Performance review software for medium sized US cities, counties and law enforcement" versus simply saying "We make performance review software". This helps you determine a focus and know which customers to target and which features help them. Further to the "building too much" section, you only need so much to close a sale. After a while, the new features are great for reducing support or customer retention (both of which you should be measuring), but build too many and you risk having a complex product with training costs that are too high and even more importantly, investing in features that don't help you sell the next customer. At some point, move on to the next product. Also a tip, we stopped building lesser-used admin features that required a UI. UI is hard to maintain and train people (and hard to kill that feature). Sometimes it's possible to simply have a script you can run that will make changes to the data model that customers would like to have. When customer call asking for something, you can offer that solution and have your dev team run the script.