Choice of alternate needs to be related to the risk of requiring it. Lots of work has been done to develop risk based alternate selection processes.
The principles of risk based selection are:
1. Safety
2. Regulatory compliance
3. Commercial requirements
The risk evaluation uses a process flow chart methodology to assess weather factors; celing, visibility, wind, precipitation, runway surface condtions, severe weather events (CB, ZR) in conjunction with aircraft performance, runway length and approach aids.
If the risk of requiring the alternate is low or very low, use the No Alternate option or the closest legal alternate (you are not going there).
As the risk of requiring the alternate increases more care is required in risk evaluation and the third prinicple is required - commercial requirements. Commercial requirements evaluation can be challenging and is dependent on the aircraft type, the airline commercial agreements with handling agents, partner airlines, airport facilities, crew duty day and replacement ability.
Putting this together for an individual airline requires a little bit of work and analysis to determine the ranking of alternate airports with respect to handling capability, airport approach aids etc. for each destination.
The objective should be to balance the correct alternate choice against the distance to the alternate. A closer alternate is more fuel efficient, but may not be suitable from a weather or aircraft handling capability. However, care must be taken to remember that the focus on airport facilities needs to be tempered with the assessment of likelihood of going to the alternate. In practice in many parts of the world, for many flights, the risk of going to the alternate is low and the alternate airport merely satisfies a regulatory requirement.
Every flight we operate always has a risk of diversion, but we must temper our pilot paranoia over a diversion, against running the business. You cannot eliminate all diversions. Diversions are a cost of doing business, but we cannot pile on the fuel "just in case", believing we will avoid diversions. Yes I know diversions have associated costs, but in a reasonable size airline the fuel carriage costs can easily be 25-100 times higher than all the diversion costs.