Because position is not traffic. A page moves from nine to six, the report shows green, and clicks do not change because the query has a featured snippet taking most of them.
Meanwhile a page that fell from three to five on a query nobody searches gets attention it does not deserve.
Tracked queries carry position, impressions and clicks together, with position taken from the search data rather than scraped. A movement is shown against what it did to traffic, not on its own.
Queries are tracked by cluster rather than individually, so the report is at the level decisions are made.
The report stops celebrating movements that changed nothing. A cluster whose average position improved while clicks fell is visible as a problem rather than a success.
It also makes the value of a rank realistic: this cluster at position four earns roughly this many clicks, from your own history rather than an industry curve.
It will not track personalised or local variations you have not configured. Positions vary by device, location and person, and a single number is an average that hides that.
It also does not promise daily precision. Position data is sampled, and Growth reports it as a trend rather than pretending to a decimal place.
Nothing here is an integration. Each line is two products reading the same record from different sides.
Tracked queries carry position, impressions and clicks together, with position taken from the search data rather than scraped. A movement is shown against what it did to traffic, not on its own.
The report stops celebrating movements that changed nothing. A cluster whose average position improved while clicks fell is visible as a problem rather than a success.
It will not track personalised or local variations you have not configured. Positions vary by device, location and person, and a single number is an average that hides that.
Fourteen days, every module, no card. Or half an hour with someone who will run it on your own records and tell you where it does not help.