A lesson from my shop: always verify DPF sensor readings before assuming clog
My customer's rig kept failing regen cycles. I hooked up my scanner and found the upstream temperature sensor was giving false low readings. Swapping that fifty-dollar part solved it, but I almost ordered a new filter. Always verify live data before you start pulling components.
Yikes, I read a similar story on a mechanic subreddit. Dude replaced his entire exhaust system over a weekend thinking his DPF was gone. His scan tool had live data the whole time showing a stuck pressure sensor. Cost him a grand in parts he didn't need. That's why your point is so key. Always look at the data stream first, it tells you the real story.
Honestly used to think throwing parts at a problem was just how it's done, maybe it's just me but I learned the hard way. A buddy had this rough idle for weeks, kept cleaning throttle bodies and stuff. Finally hooked up a decent scan tool and watched the fuel trim data, led us right to a cracked hose on the intake. Felt pretty dumb for not just looking at the live data first, could have saved so much time. Now I always tell people to watch the numbers move before they buy a single part.
Ever think to check the wiring to the sensor itself? In my experience, a little corrosion or a chafed wire can mess with those signals just as bad as a bad sensor.