Ask HN: Any good essays/books/advice about software sales?

Do you know what you’re gonna build, or is the agency just building to client spec?

Who’s the client?

How will you find them? Where are they? How will you communicate with them? How will you get them to use your agency?

That’s the trick

Edit: spelling

This is so great! Thanks for the reply, going through it rn.

Sales is a matter of talking to people you know. There is no shortcut. The work is:

  1. Meet people.
  2. Get to know them.
  3. Make a pitch if they have a problem they will pay you for.
  4. Get paid.

That their problem inclines them to pay you is the only important feature of the problem and that you get paid is the only important feature of your service.

Meeting people is literally meeting people.

Getting to know them means time and effort spent fashioning a relationship. B2B relationships are long term. You can assume that people who regularly use agencies already have agencies they work with and will continue to work barring change to their business or the agency's business. These happen but not on a high pressure timeline. Consulting is a long con. Good luck.

Have you used much software yourself? If you know the difference between good and bad software (ahem, Electron-based apps are inherently bad software), then you have everything within you to write good software.

The principles of selling software successfully are thus:

* Charge a low price. A low price gets many customers, and a high price gets fewer, but the money earned equals about the same. Better to have more customers, because that's more eyeballs and mouths seeing and talking about it.

* Make the software available for as many platforms as possible, with GNU/Linux being a first-class citizen. Although most of your customers will use macOS and Windows, having GNU/Linux support signals robustness and longevity, earning trust.

* Use a generous license. Best to AGPLv3+ -- competitors can't beat it. If others share your program gratis, it just leads to even more official customers. Any changes can be reincorporated into the official software, so the first-mover advantage is everything.

* The software must be GOOD. It's got to save people time in their otherwise busy lives, and it has to be robust -- it has to work every time. The software has to know when things won't work, and fail gracefully. This is what sets hobby-ware apart from professional-ware.

* Update the software regularly, and keep in contact with customers in a visible public place -- even if it's just a static, one-directional web page. Let people -- and search engines -- know the project is chugging along. Give customers something to look forwards to.

* Fill a niche, and give the software a broad appeal. A tool with "something for everyone" -- features that not everyone uses, but everyone uses SOME features -- is important to have.

* Write GREAT documentation, and typeset it with LaTeX. This is important to convey quality. Hobby-ware has a Readme.txt -- professional-ware has a PDF manual that is so well-written, it could be printed out, put in a box with the software on a CD, and shipped.

* Record the project into history. Be everywhere on relevant forums, and push the product when it's relevant. When someone has a problem that your software fixes, they will see those comments -- even years from now -- and that helpful, relevant advice is genuine marketing that stays posted forever. Bought ads don't come anywhere near that kind of value.

And, as far as the actual software is concerned, write it in a popular language with a popular graphics toolkit. Python 3 + Qt5 or 6 -- using PyInstaller to generate single-file executables -- is dang near perfect. Stick to conventional user interface guidelines. Build software that you, yourself, use everyday. Don't work on software that you don't use personally.

Familiarize yourself with macOS software of the 1990s (most software of this era was good, most software today is bad), and this article -- How To Design Software Good. https://www.haiku-os.org/docs/HIG/index.xml

Small channel but Craig Hewitt is legit: https://www.youtube.com/@thecraighewitt