snowfalcon2, the initial search was a long, narrow, rectangular box from the last reported position almost to Tasil, favoring the right of the expected track. A second search box was flown that day in a more square-shaped rectangle from the last known position to about half-way to Tasil.
Searches on the days following were made to the right of the track, and south of the last known position, even south of the St Peter & St Paul islands. Various sightings of possible debris led in part to searches in these areas. I believe no debris from AF 447 was actually found in the searches from June 1 through June 5.
Once a search was again done that covered the area that was left of the track, the first bodies were discovered on June 6, approximately halfway between the last reported position and Tasil. These bodies were left of the track. If the bodies had been in that location on June 1, I believe that they would have been within the boundaries of the two search boxes of June 1.
By being outside the boundaries of the initial search, they were missed.
Hindsight is always 20/20, but it is unfortunate that the search on subsequent days was devoted to areas to the right of the track. If the search boxes of day 2, which were run basically perpendicular to the right of the track (as if the aircraft had deviated right) had been mirrored by searches to the left of the track, its quite likely bodies and floating wreckage from AF447 would have been found on June 2nd or 3rd. That would have fixed the impact position a lot more precisely.
Having limited resources, a long distance to fly to the search region, the Brazilian military made the judgment of where, given the currents and meteorology, the impact was more likely to be, and focused their efforts there. In a more perfect world, they would have looked as earnestly at the area to the left of the track as they did on the right.