“Biblical” is doing a lot of work in worship conversations and rarely gets defined. It does not mean a song is stitched together from verses — you can quote Scripture constantly and still teach something Scripture does not say. Here is what I actually check for, and why each one matters more than it sounds.
1. Is the claim about God true?
This is the whole thing, and everything else is downstream. A congregation singing a line is asserting it. If the line says God will do something He has not promised, the song has put words in His mouth on behalf of a room full of people.
The most common failure is not heresy. It is over-promising — songs that guarantee breakthrough, healing, or a lifted circumstance as though these were contractual. Someone in that room has been asking for exactly that for eleven years. The song has just told them, in front of everybody, that the arrangement is not working for them.
God’s steadfast love is a promise. A specific outcome by a specific date is not. Both can be sung; only one can be sung as certainty.
2. Who is the subject?
Count the pronouns. Not to enforce a ratio — the Psalms are full of “I” — but to see what the song is about.
Psalm 40 has more first-person in it than most modern songs, and it is unmistakably about God, because every “I” is doing something in response to something He did. “He drew me up from the pit… He put a new song in my mouth” (40:2-3). The self is present as a witness, not as the subject.
The test is not how often “I” appears. It is whether removing God from the song would leave anything standing. If the song still works as a general statement about feeling hopeful, it was never about Him.
3. Does it have any content?
Some songs are entirely adjectives. God is described as great, good, worthy, mighty — all true, none of it saying what He did. Biblical praise is almost always narrative: it recounts an act and praises the actor.
“Oh sing to the LORD a new song, for he has done marvelous things” (Psalm 98:1). The for is the content. Psalm 136 repeats one line 26 times, and between each repetition names a specific historical act. That is the shape.
A useful check: after the song, could someone say what God has done, or only how the room felt about Him?
4. Can it be sung on a bad day?
Most worship songs are written for a good room. But they get used in hospital car parks, in kitchens at 3am, at funerals.
A song that only works at volume, on a Sunday, with everyone feeling it, has a very small useful life. The Psalms survive because they were built for the whole range — including the days that do not resolve. If a lyric would be unbearable to sing while grieving, it is worth asking whether it is true or just buoyant.
5. Is it singable by people who are not you?
This is craft, not theology, but it is a spiritual issue in disguise. A melody in a range only the leader can reach, a lyric too specific to one person’s experience, a phrase nobody can grab on the second time — all of these turn congregational worship into a performance with an audience.
Psalm 33:3 says “play skillfully”. Skill in this context includes making room for other people’s voices.
6. Does it end somewhere real?
Not every song has to resolve — Psalm 88 does not. But a song should know what it is doing. Ending on unearned triumph is dishonest; ending nowhere at all is just unfinished. The laments that turn, turn after the complaint, and the turn is what makes the complaint bearable rather than what erases it.
What this looks like in practice
I write lyric-first, which is the practical form of all six. If the words have to hold up before a chord progression is allowed to make them feel important, most of the failures above get caught early — the over-promise reads as an over-promise on the page, and the song that is only adjectives is obviously empty when spoken aloud.
Redemption & Faith is being written to this list, and it is slower for it. The singles so far are here, and this is where the standard comes from.