When the media reports on jobs numbers, they almost always refer to the "seasonally adjusted" figure which smoothes out predictable fluctuations in employment driven by the annual calendar. The seasonal adjustment usually isn't that big a deal in practice. The big exception is the jobs report that comes out in February, because in January the un-adjusted jobs report is invariably a disaster:
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/3379010/fredgraph_1024.0.png)
This is the macroeconomic miracle of Christmas running in reverse, and it's kind of ugly. When you hear that jobs were created in January, all that means is that a few hundred thousand fewer people got laid off than get laid off every January. And every January, you have over 2 million people losing their jobs.
That doesn't change the fact that seasonally adjusted data is the right thing to use for trying to make economic policy. But it does mean that even a "good" January for the labor market isn't going to feel very good to the millions of people left unemployed by the post-Christmas bust.
