I'm curious to know, when designing new RPGs, once the basic idea is decided upon, what is a useful design priority to follow.
Here's what I have been trying as an example:
1. Attributes and Skills, or equivilent
2. Derived Attributes and How To Calculate Them
3. Some, not all, Special Systems
4. How to build a character
5. ...
At the moment, I'm stuck trying to figure out if I'm missing something by step 4, or if I need to fully plan step 5 and onwards. (Hopefully what gets posted here can help other designers as well as me.)
For me the order of design is this...
1) Concept (you've said you've got this worked out).
2) Design a core task resolution system: when a character tries to do something, how do you determine success or failure. dice total vs. difficulty? Opposed rolls? Percentage checks? Success counting dice pool? card draws? trait bidding system?
It doesn't have to be the same for everything, but having the right system can either work with (ex. using playing cards for a Wild West game can be quite thematic) or against (ex. rolling on a table for a LARP-based game) your concept.
Also important at this stage is figuring out odds of success using your system for an average unskilled person, a skilled expert and the best-of-the best as these will help you gauge the values you need for attributes, skills, difficulties, etc.
3) What does a character in your setting need to make checks for?
Make a list of what you think most characters will need to make checks for and, if there's a bunch, rank them in terms of how common those checks are (numeric, stars, whatever works).
Then take your list and start grouping tasks together into similar categories with an eye towards creating three to eight "piles." These are going to the attributes your system needs with individual tasks becoming either skills (for a highly detailed system), parts of a skill (for a less detailed system) or a general task (for a system that doesn't want the detail of skills at all or where you think skill should never apply).
Pay attention to particularly small piles and/or those with many low-ranked tasks and for piles that contain lots of high-ranked tasks. Those could end up as universal "dump stats" or "super stats" respectively. Try moving some tasks around until the piles are fairly even if you want your attributes to be about equally useful.
4) Make your attribute and skill lists and figure out any other traits you're going to need (ex. hit points and damage to measure the success of combat task checks) and how to determine those (ex. From one or more attributes or skills, randomly rolled, set value).
5) Figure out how a character is built. If you've got superstats or dump stats you should look at either random generation (ex. old school D&D) or different costs using build points (ex. HERO System). If reasonably balanced with each other, then a more general point-buy or attribute array system might be where you look (or go completely random since you'll still theoretically get something playable on average).
A Lifepath system can be used in relation to any of those.
Your concept will help a lot here. If campaigns are intended to be short or characters fairly disposable, then random generation can speed up creation by removing a lot of choices from the player. Games where players will expect to be using the same character for a long haul then giving them more control via point buys or arrays will make them more invested in the resulting character.
6) Now start filling in all the supporting details: opponents, allies, terrain, equipment, special abilities, etc. For these sub-systems you just need to run through steps 1-5 within a more focused area (spellcasting for example) to figure out what's needed and then figure out how that attaches to the main system.
At least this is the approach I've followed.