clearstream
(He, Him)
A valuable guide is DMG 237 - "Only call for a roll if there is a meaningful consequence for failure". The advice on subsequent pages - Resolution and Consequences - offers some benefits; taking a different approach offers others.It's in discussions like these where it's easy to see which DMs use progress combined with a setback on certain failed checks and which ones don't.
A positive here is that even on a failure the characters can still proceed through the game world, because the meaningful consequence is framed in other terms (unwanted attention). However, when the setback is instead that the characters can't proceed through the game world on their intended route, that too can propel the narrative along interesting lines.Failed the Intelligence (Investigation) check to figure out how the secret door works? You figure it out, but make noise in the doing and have drawn unwanted attention to yourself. Now we have a new interesting situation to resolve.
In play, a major difference might often be who the work to develop the narrative falls upon. Progress with a setback puts the work on the DM to decide what the setback might be. Stymieing progress puts the work on players, to think of another approach.
This is standard in many published adventures. For example, the city of Omu in ToA and the Darklake in OOTA both offer time-based encounters. The choice here overlays choices about how to handle the meaningful consequences of failure. A DM might well use both.One can also assign a resolution time to common tasks in adventuring locations (say, 10 minutes), then put a wandering monster check at particular intervals (maybe every 10, 30, or 60 minutes). At that point, every failed check becomes wasted time (a meaningful consequence in this context) which pushes the PCs closer to having monsters show up to bother them. This also works with any sort of time pressure e.g. save the prince before midnight or he is sacrificed by the cultists.
I feel a DM should just be clear on what some of the useful modes are, and why they would - or would not - use them. A clear benefit of progress with a setback is that characters still get to proceed down their planned route. It also provides an excellent chance to introduce new factors into the narrative. On the other side, I enjoy putting the burden directly onto players rather than introducing further elements of my own devising. I also notice that efforts to come up with setbacks sometimes feel like a bit of a reach.
It is worth noting that a major cost of allowing characters to flatly fail - where the meaningful consequence is (at least in part) blocked progress - is the effort to flexibly expand the game world out in directions off the planned path. As players are forced to invent other approaches.