4 ms·

If you choose to run a software development agency: - Shoot for between 3 and 10 clients. Any less and you'll be very stressed about losing a client. Any mor

If you choose to run a software development agency: - Shoot for between 3 and 10 clients. Any less and you'll be very stressed about losing a client. Any more and you'll be overwhelmed with juggling too many balls. - Fire bad clients. They aren't worth the stress, frustration, and opportunity cost. - Work on your process. Doing an hour of client works earns you one hour of revenue. Improving your agency processes can earn you a large multiple of that. - Don't be a <insert software technology> shop. Be a solver of a specific business problem for a specific type of business. Example: "We streamline backend processes for multi-million dollar trucking companies." The most valuable contracts are for solving expensive problems.

akulbe 9y ago
Can you elaborate on what you mean by improving your agency process?
- jrs235 9y ago
  > Can you elaborate on what you mean by improving your agency process? I think he means what this person said: https://news.ycombinator.com/item?id=14147079 https://news.ycombinator.com/item?id=14147079
- lprubin 9y ago
  There are certain core things that your agency will have to do in order to succeed. Some high level examples: - Aquire Clients - Do client work - Manage projects - Bill clients - Do pricing/timeline and estimates - Hire employees or freelancers Within each of these high level tasks are numerous subtasks. You do not want to re-invent how you do those things each time you need to do them. You want a tried a true process that you keep evolving and improving based on a historical success and experimentation. This will allow you to scale and handle larger and larger projects. It will also allow you to onboard new team members quickly and when they leave, the things they learned won't leave with them because you'll have documented the learnings. This is the book that inspired this line of thinking in me. https://www.amazon.com/dp/B000RO9VJK/ref=dp-kindle-redirect?_encoding=UTF8&btkr=1 https://www.amazon.com/dp/B000RO9VJK/ref=dp-kindle-redirect?...
wonderwonder 9y ago
Fire bad clients. This is a very important piece of advice. Bad clients will kill your energy, drive and time.
- thecrazyone 9y ago
  Define bad clients. Is it: 1) clients not worth the money for the effort (but good learning), OR 2) clients who are unprofessional i.e. don't pay up etc (I'm sure there are multiple such delineations on what a bad client can be)
Tenhundfeld 9y ago
>Be a solver of a specific business problem for a specific type of business. Totally agree in principle, but I've found it's very hard to do this in practice. Unless you have a priori knowledge of an industry, before your first consulting gig in it, it's hard to just step into a business domain and know what specific problems an industry has that can be improved with better software and which of those specific problems are generic enough to be shared by many companies in that industry – while not being served "well enough" by an existing product. And you have to figure out if it's a niche that's big enough to support a company, while balancing other desires, e.g., not spending weeks traveling around the country to sell and work with companies in this sector. I'm not saying it can't be done. It can, obviously; there are companies doing it. But it's not easy to just go do it. From my limited exposure to consultancies taking this approach, many kind of kept their eyes open and fell sideways into it, i.e., they worked with one client in a sector and realized the client's competitors had similar problems.
- lprubin 9y ago
  I agree that it is challenging for all the reasons you stated and more. This book really helped me overcome many those issues: https://philipmorganconsulting.com/the-positioning-manual-for-technical-firms/ https://philipmorganconsulting.com/the-positioning-manual-fo...

akulbe 9y ago

Can you elaborate on what you mean by improving your agency process?

jrs235 9y ago
> Can you elaborate on what you mean by improving your agency process? I think he means what this person said: https://news.ycombinator.com/item?id=14147079 https://news.ycombinator.com/item?id=14147079
lprubin 9y ago
There are certain core things that your agency will have to do in order to succeed. Some high level examples: - Aquire Clients - Do client work - Manage projects - Bill clients - Do pricing/timeline and estimates - Hire employees or freelancers Within each of these high level tasks are numerous subtasks. You do not want to re-invent how you do those things each time you need to do them. You want a tried a true process that you keep evolving and improving based on a historical success and experimentation. This will allow you to scale and handle larger and larger projects. It will also allow you to onboard new team members quickly and when they leave, the things they learned won't leave with them because you'll have documented the learnings. This is the book that inspired this line of thinking in me. https://www.amazon.com/dp/B000RO9VJK/ref=dp-kindle-redirect?_encoding=UTF8&btkr=1 https://www.amazon.com/dp/B000RO9VJK/ref=dp-kindle-redirect?...

wonderwonder 9y ago

Fire bad clients. This is a very important piece of advice. Bad clients will kill your energy, drive and time.

thecrazyone 9y ago
Define bad clients. Is it: 1) clients not worth the money for the effort (but good learning), OR 2) clients who are unprofessional i.e. don't pay up etc (I'm sure there are multiple such delineations on what a bad client can be)

Tenhundfeld 9y ago

>Be a solver of a specific business problem for a specific type of business. Totally agree in principle, but I've found it's very hard to do this in practice. Unless you have a priori knowledge of an industry, before your first consulting gig in it, it's hard to just step into a business domain and know what specific problems an industry has that can be improved with better software and which of those specific problems are generic enough to be shared by many companies in that industry – while not being served "well enough" by an existing product. And you have to figure out if it's a niche that's big enough to support a company, while balancing other desires, e.g., not spending weeks traveling around the country to sell and work with companies in this sector. I'm not saying it can't be done. It can, obviously; there are companies doing it. But it's not easy to just go do it. From my limited exposure to consultancies taking this approach, many kind of kept their eyes open and fell sideways into it, i.e., they worked with one client in a sector and realized the client's competitors had similar problems.

lprubin 9y ago
I agree that it is challenging for all the reasons you stated and more. This book really helped me overcome many those issues: https://philipmorganconsulting.com/the-positioning-manual-for-technical-firms/ https://philipmorganconsulting.com/the-positioning-manual-fo...