You can derive at differing intelligence with the same basic information, it depends on the mindset.
Intelligence based operations are in my opinion a buzzword for the old OODA loop idea with improved data collection. Like every OODA loop there's no need to run faster if you can hack it.
The faster an OODA-loop runs, the less new information will be inserted and the less analyses can be conducted, making the change of status and derived action increasingly predictable parameters. Predictability allows you to jump ahead in the cycle and win despit having a slower OODA yourself.
A suitable option are extensive information and mindset campaigns, again focusing on predictability, by creating a mispresentation for outsiders. Language barriers are very convenient filters to enable such tricks, like the traditional lack of Arab speakers in the Western intelligence community.
The Brazilian success is more due to the novelity and unpreparedness, but intelligence outside the usual OODA loops will gain an increasing role in OODA-loop hacking and protection, requiring more networking of frontline forces with scientists.