Sure, but
- Twitter highlights impressions when it's a particularly useless metric for their platform. Why not expose uniques?
- Why include a user's own activity in their analytics data? Does anyone really want that?
- A tweet doesn't even have to appear on the screen to acquire an "impression". It could be far down the timeline and out of view, but as long as it's included in a batch in an API response, the counter goes up.
- This number is exposed to everyone, not just ad industry insiders, and is subtitled "number of times people saw this tweet" - but that's not what's counted.
The implementation may be technically correct, but it is, I think, misleadingly biased toward higher numbers. I can post a tweet and it can get 10 impressions in less than a minute, just from my clicking around. If I expand the tweet and click "reply" to attach another 140 characters, that's two "engagements". If I click a link to make sure it works, that's three. Honest values for those numbers would probably be zero.