After the first change, unless there is some indication of fluid degradation (via UOA or dark, smelly oil), or a documented history of issues with the fluid, IMO 30K is beyond obsessive. 50K is merely pedantic. ( : < )
If a trans has a history of design "issues" more frequent fluid changes generally won't help unless they are documented to be related to the oil (some ATs do work their oil pretty hard). And then there are contamination issues (see more below), which means that a particular trans prefers to operate with cleaner oil. Diligent research should turn up detailed answers so a person would know how to proceed. I guess it's all moot since the car is going bye-bye, but these things can be evaluated along the way.
As I said, the first change is the most important. Eleftherakis & Kahlil, two engineers who literally wrote the book on auto trans contamination thru decades of research, found that the average automatic generates 75 percent of it's lifetime of contaminants in the first 5K miles. That's the junk built in from manufacturing and from break in. The amount built in is variable according to how good or bad the manufacturing process was. The Japanese and Europeans have been better at clean AT manufacturing over the past 25 years or so, but the Yanks have caught up lately... by necessity, I think. Break in is still an issue.
After break in, the rate of contamination levels off at a greatly reduced rate and the contamination level in the oil increases at a steady rate from normal wear. Contamination rates are often related to the use of the vehicle. Easy use = low rate and vice versa.
The main goal then is to monitor fluid condition and change the oil when either, a) the contamination level rises past about 25 mg/l or, b) the fluid has oxidized or is depleted in some way. In most cases, the contamination level rises past an optimal level before the fluid is depleted (this is an across-the-board, generic statement; there are individual exceptions).
For long term service, some addition trans filtration (beyond the pad in the pan), which will hold the contamination in check so the fluid can live a natural life. If the contamination level is maintained below 25mg/l (the optimal level is below 10 mg/l) then in an easily used trans with a premium fluid, ATF life can indeed be virtually considered "lifetime" in many cases. Certainly 100K.
Determining all these conditions does require extra effort on the part of the owner. For the most part, IMO, if you have a good supplemental filtration system (an inline filter is fine), you could pretty much tell by fluid color but I am pedantic enough to want to do a periodic UOA (every 30K or so) long enough to determine some working guidelines as to the life of the oil. Once I have a basic pattern, then monitoring is unnecessary IMO, until some part of the equation changes (operationally or the oil).
