If you want to give them class features instead of spells, why not?
Or donyou want to add a lot of features and retain all their spellcasting?
Giving ranger class features instead of spells was the suggestion I answered to.
Giving class features instead of spells works, but ignores the point. Looking only at a table ignores the context of the rest of the features of a class to create an inaccurate assessment of that class.
I'm a fan of using multiclassing to create variations for the character concept. Like in the OP, the easiest way to reduce a bard's spellcasting, like all spellcasters, is through the multiclassing method.
If I want a character that's more like a 2e bard then 1st level as a fighter with the entertainer background, 2nd level as a wizard, 3rd level as a rogue, 4th level as a wizard, 5th level as a wizard, 6th level as wizard, 7th level as a rogue, 8th level as a wizard, 9th level as a wizard, 10th level as a rogue, etc ending it with fighter 1 / rogue 8 / wizard 11. Use the thief subclass and the illusionist subclass. Expertise should include history and arcana.
That completely ignores the 2e caster level advantage bards had from the experience progression and bonus XP awards but that seems to be what some people want. It's basically how 2e bards worked, however, and the varied spell list can be included by swapping wizard out for land druid.
That's literally all it takes for an older style "bard" character.
Alternatively, there's nothing wrong with saying "5e bards don't fit my campaign style" and just banning the class at that table. Players can adapt.
Or ban the class at that table adding it as a background.
Bard
Ability Scores
: Wisdom, Intelligence, Charisma
Feat: Musician
Skill Proficiencies: Arcana, History
Tool Proficiency:
Choose one kind of Musical Instrument
The ability scores match what we would expect from poets and philosophers, musician grants an inspiration ability, and those skills are the classic bardic knowledge / lore abilities. Also very easy to implement.
That's not what I see happening here. The fixation on taking a class that functions fine as is and recreating it in a format weaker than it's ever been (it's been 17 years since 3e ended, and the arguments ignore benefits like 2e caster level or 3e songs and similar spell level progressions to clerics or druids and relative comparisons to rangers or paladins) by focusing on old charts while ignoring context relating to those charts is just punching down on the class.
There's no gain in changing bards into weaker spellcasters because it's easily possible to make a character like that. There's only the loss of existing archetypes in expecting others to change bards too. Not recognizing all the benefits bards actually had in those systems seems like they didn't play or understand the bard mechanics under those systems and their expectation is that the only bard that should exist is a weak bard. That's a weird expectation. But another easy solution is to play those older systems or a retro clone.
The 5e bard works and has worked for a decade. The 2024 rules did adjust their spellcasting down by restricting some spell access and reducing the rate at which they gained high level spells known.
Now if it's a setting concern that's different. Bards or rangers or any other class should be reasonably capable for players. That doesn't mean reducing spellcasting for the sake of reducing spellcasting (which is what some posters seem to be doing on these boards) and the warlock chassis is a good way to go.
A bard spell list that uses rebranded pact magic, magical secrets that's rebranded bard spells magic arcanum, and powerful bard songs that replace invocations could work very well. The spells part is easy to maintain a thematic bard while further limiting access to high level spells; the work is in designing the songs. A person could call it "bardic training" and include songs plus other benefits.
Either way, the current bard also works well. ;-)
