Absolutely true, Moondog...a lottery it is. But that isn't necessarily the method itself. It is the way the method is filtered down the ranks...with the end result being something completely different from that which was taught by the expert to the 'select few'.
An example, of what I considered to be excellent practice:
A Higher Level Teaching Assistant was talking to parents about Sensory Processing and how they work on sensory needs at DD1's school. She gave the illustration of a little boy I will call 'Mark'. He has severe and complex needs, as do many of the children at DD1's school.
'Mark' had been observed over a period of time. Detailed notes as to his response to stimuli had been kept. The HLTA had looked at his data and realised that he had gradually decreased the use of his left arm. Because they had such detailed notes, they could see that he had been using his left arm more, and they could track the reduction in use. So they then looked at strategies to increase it.
Now, that school doesn't use ABA, and doesn't use SMART targetting (yet). But they are looking at the children as individuals, seeing, noting and tracking trends in their development, and responding to it.
At the end of the day, whether it is called 'ABA' or 'SMART' or anything else, doesn't matter. So perhaps it would be best to start from the weakest 'edge' of the problem. Start by getting good tracking, or even good individualisation...whatever.
DD1's school started to get a focus on data last year. It was a key part of their School Development Plan. This year they've said 'do you know...we aren't challenging our children enough', and they are looking at how to ensure they have challenging expectations right at the start of the school year for each child. It is baby steps, but by having baby steps, you don't get a patchy result.