I have downloaded an ALOS PALSAR PLR SLC L1.1 data scene from ESAEARTHNET database.
I load the data into SNAP using File ==> Import ==> SAR sensors ==> ALOS PALSAR CEOS
First of all in World view pane in left side of software I found that the overlay area of this dataset is not located truly and it does not show even the rectangle of the data border (it shows its location by a point in a wrong location).
When I show the bands of the dataset, it shows a constant coordinate for all pixels equal to 0,0.
In addition it seems that bands are rotated 180 degrees. All image bands are longer than the real length of image border.
I used the following tutorial for preprocessing the dataset: http://sentinel1.s3.amazonaws.com/docs/ALOS%20PALSAR%20Orthorectification%20Tutorial.pdf
But when do step by step, In Terrain Correction an error appears with a message about the SNAP which is not supporting this datatype!
I redownloaded the dataset and it does not fixed my problem. Also I applied all steps on another data scene and get the same error.
The position of the rectangle in the World View pane is not very correct. It is just a coarse estimate on where to locate the product.
The reason for the 180 degree rotation is that SNAP alwas shows the scene according to its acquisition geometry (ascending or descending) and north is not automatically up. That is completely normal and will be corrected after projection or terrain correction.
The reason why your image looks distorted lies in the nature of Level 1.1 products. They are displayed in slant geometry, where some pixels are still closer to the sensor than others which leads to irregularly shaped pixels. Once you performed multi-looking the image looks better. Please also see here: GTC (Geocoded terrain corrected) data
If you are generally new to SAR, I can recommend the SAR EDU materials:
I had the exact same problem. You can try the SNAP beta version. With that one I could open the data from ESA. Although I could not georeference it but I could process it in other ways. Maybe you have more luck with the georeferencing.