There is often some confusion at my table between the use of the Perception stat and the Observation skill . By and large the I say that the distinction is as follows: Perception is used to notice things passively with one’s senses and mental acuity, while Observation is used to actively look for something (especially things that are deliberately hidden or actively hiding).
To aid this distinction I note that Observation tests are almost always versus tests, while Perception tests for noticing things are often graduated.
Graduated tests are explained on BWG pg 26, but are barely ever mentioned again and I’ve always found this odd. As such, I’ve never really used them that much. So far in my games, routine tests have been the hardest type to come by; frequently my players are stuck with B2s that they opened but didn’t advance during character creation for dozens of sessions while their B3s and B4s slowly get to advance up to B5s and B6s. In fact I’d say that the B2 has even gotten a bit of a reputation at my table for this reason, having a B2 is like having a bit of a curse that you can only break out from by practising it in downtime. Now, maybe this is intentional design, but a thought occurred to me recently that maybe it’s just because I’m not utilising graduated tests properly?
What follows then, is my assessment of graduated tests, and how I plan to use them going forwards; in particular for Perception and Wises when noticing things. I’d like to hear your kind thoughts on it all, and hopefully you’ll find this mindset useful in your own games too.
Either the GM or player can call for a graduated Perception or a Wises test at any time—usually while dungeon crawling, but anywhere really—to see if the PCs notice or remember anything important when they enter a new room.
Most graduated tests count as base Ob 0 (routine), unless there are additional penalties from things like poor light; but we’re really more interested in seeing how many marginal successes you get. The number of successes are then either judged against the GMs private list of information to give out, or compared to the table below for more general information to give out. The wises obstacles and the absolute obstacles list (on pg 15) have been combined here for a good list of success comparisons:
| 1s | Common knowledge / requires little thought / cursory information | 5s | Rare goods / awfully convenient plot detail / requires expertise |
|---|---|---|---|
| 2s | Interesting facts / basic information | 6s | Bizarre or obscure facts / requires a heroic effort |
| 3s | Details / convenient plot detail / requires concentration | 7s | Freaky details / improbable knowledge / complex secrets |
| 4s | Uncommon knowledge / simple secrets | 8s | Forbidden knowledge / requires almost preternatural insight |
Getting help on a test like this just means two minds are better than one, so to speak, and the characters should roleplay trying to remember stuff together after seeing how many successes they got.
Edit 1: Furthermore, if the GM only planned for some information around 5s, for example, but the party get a miraculous 10 successes: the GM probably shouldn’t make stuff up on the fly if it doesn’t make sense for the characters to be able to glean anything else from what they see. So keep realism in mind.
Edit 2: Also it’s reasonable to assert that an opposite of ‘Slowest and Loudest’ should be employed for these types of rolls: where the ‘Sharpest and Wisest’ should always lead.
On further consideration, I think this might also be why I haven’t had much fun with wises recently either. The example of a graduated test being used on pg 26 is really quite insightful though: in it, it shows that a quick Perception or Wises tests can be used like spot checks in other TTRPGs. It seems to me that intent and task become a little less important to announce before a graduated test, and perhaps this improves pacing?
I really like putting on the breaks and laying out the consequences of every roll, I feel that it keeps us on the right track. But that also means I typically end up saying “yes” a lot to more trivial (routine) things. And maybe this is ultimately why B2s have gotten that bad rep.
If routine tests are similarly difficult to come by in your own games, consider trying this technique. I know I will be 
Happy gaming!