This one isn’t even a House Rule as much as an observation that any attack that would have hit but for the AC bonus armor grants presumably smacked the target but failed to penetrate the armor. I like to emphasize this when describing the result of the attack. In DCC this is dead easy because the dividing line between a clean miss and at least hitting the armor is just 10, give or take agility modifiers.
I like this both because it makes the description more vivid at almost no cost and because otherwise some folk get hung up on the notion that armor is making you dodgier and want to introduce some sort of damage reduction instead. Best nip that in the bud.
I think it’s a toss-up whether you want to extend that observation to true 3e-style “touch attacks” where spells like Chill Touch only need to beat the AC before armor bonus in order to take effect, or you stick with beating the AC. On the one hand, if the caster has to get into melee and beat the AC that makes all the levels of Chill Touch where it only lasts one round pretty damn punk. A spell that probably has less than 50% chance of affecting a target even if you succeed in casting it (up to a check of 17), and then only if you put your unarmored wizard in the thick of things, and always costs a point of Spell Burn to cast? What a rip-off. On the other hand, if you allow that kind of touch you actually need to worry about how much of a foe’s AC is armor vs agility bonus, and that’s the kind of finicky drag that I play DCC to avoid. How much of a griffon’s AC 17 is tough hide vs. speed? Obviously you can just make a ruling and go, but it’s one more thing to remember when it comes up. That said, I’m inclined to allow it, if only because of how much my one player who got stuck with the spell hates it when he’s reduced to trying it.
TL;DR Make note of when an attack roll beats 10 but not the AC and use that to inform your description of the attack bouncing off the armor.