Before most major releases, news services survey economists and publish the median forecast. Prices already reflect that expectation before the figure comes out.
When the figure is published, markets move on the difference between actual and expected. A strong jobs number that was fully expected can leave markets unmoved; a modestly strong one that beats a weak consensus can move them a lot.
Why it matters
This is why the same piece of news can move markets in opposite directions on different days. It also explains why markets sometimes rally on bad news: if the news was less bad than expected, it is a positive surprise.
On FinVolio
Consensus data is licensed, so FinVolio shows its own estimates instead, clearly labelled, with their track record.
General information, not advice.