Performance with multiple FLATTENs

[Matt]

I have JSON data with with a nested list and am using FLATTEN to extract 
two of three list elements as:

~~~
SELECT id, FLATTEN(data)[0] AS dttm, FLATTEN(data)[1] AS result FROM ...
~~~
This works, but each FLATTEN seems to slow the query down dramatically, 
3x slower with the second flatten.

Is there a better approach to extracting list elements?

~~~
[
  {
    "id": 16,
    "data": [
      [
        "2016-07-13 00:00",
        509,
        "OK"
      ],
      [
        "2016-07-13 00:01",
        461,
        "OK"
      ],
      [
        "2016-07-13 00:02",
        508,
        "OK"
      ],
~~~