Skip to content

Our approach

Questions before explanations

None of what follows is new. It is roughly how a good teacher has always worked, and roughly the opposite of how an answer machine works. The difference shows up about three weeks later, when you find out what you actually kept.

Now these were more noble than those in Thessalonica, in that they received the word with all readiness of mind, examining the Scriptures daily to see whether these things were so.

Acts 17:11

Inside a session

What twenty minutes actually looks like

A session is short on purpose. Six moves, in this order, every time.

  1. You read the passage first

    Trinity does not summarise it for you. The text goes up, in full, and you sit with it before anything else happens. A few verses, not a chapter: enough to hold in your head.

  2. It asks what you noticed

    Not a quiz question with a right answer waiting behind it. An open one: what is strange here, what repeats, who is speaking, what would you ask the writer. Whatever you say becomes the material.

  3. It works from your answer, not around it

    If you read something into the passage that is not there, Trinity points back at the words rather than correcting you in the abstract. If you notice something real, it follows your thread instead of its own.

  4. It fills in only what you now need

    Context arrives when it unlocks something: who Paul is arguing with, what a first-century reader heard in that phrase, why the Hebrew repeats a word the English smooths over. Never as a lecture up front.

  5. It asks you to say it back

    In your own words, without scrolling up. This is the part most tools skip, and it is the part that does the work. Saying it back is how you find out whether you have it.

  6. It brings the passage back later

    A week on, the same text returns in a different shape: inside a related passage, or as a question you now have the pieces to answer. Not because you failed. Because that is when returning helps.

Why answer-vending fails

A chatbot and a tutor are not the same tool

Both can produce a true sentence about a passage. Only one of them leaves you able to produce it yourself.

How a general-purpose chatbot and a Bible tutor handle the same six situations
SituationA general-purpose chatbotTrinity
You ask what a verse meansAnswers at once, at length, in a confident voice.Asks what you already think it means, then works from that.
You are wrong about somethingTends to follow the framing of your question and agree with it.Points at the words on the page and asks you to look again.
You want the referenceSometimes cites. Sometimes produces a verse that does not exist.Book, chapter and verse every time, or it says it does not know.
The question is genuinely disputedPicks whichever view your phrasing seems to prefer.Names the main views, says who holds them, and does not arbitrate.
Next weekStarts over. None of last week's work is anywhere.Brings the passage back before you lose it.
What you walk away withA paragraph you will not be able to reconstruct.A passage you can find again and explain to somebody else.

This is a description of a tendency, not of any one product. Some chatbots cite carefully. The structural problem is the same either way: a tool built to end the conversation quickly cannot also be built to keep you in it.

Spaced review

Forgetting is normal. Planning for it is the whole trick.

You will forget most of what you read this week. That is not a character flaw and it is not a sign that you read badly. It is what memory does with anything it has only met once.

So Trinity keeps a light record of what you have worked on and brings it back on a widening schedule: a few days, then a couple of weeks, then longer. The return is never a test. It is usually a new passage that leans on the old one, so recalling the first is the price of understanding the second.

The effect compounds quietly. After a few months you are not recognising verses; you are navigating a book you have some sense of the shape of.

Small parchment squares laid out as a path, some of them marked

The learning science, plainly

Three findings we did not invent and will not overstate

These are broad, well-replicated patterns in how people learn, described here in plain language. We are deliberately not citing a study, a number, or a researcher. The principle is what matters, and a fabricated citation would be worse than none.

  • Recall beats rereading

    Pulling something out of your own memory strengthens it far more than reading it again does. Rereading feels productive because it feels easy, which is exactly the problem: the ease is the signal that nothing is being built.

  • Spacing beats cramming

    The same total amount of study, spread across days, holds better than the same study done in one sitting. Trinity plans returns for the point where a passage is starting to fade rather than the point where it is still fresh.

  • Generating beats receiving

    An answer you produced yourself, even a wrong one you then corrected, sticks in a way a supplied answer does not. That is why Trinity makes you commit to a reading before it offers one.

Yes, if you call out for discernment, and lift up your voice for understanding; if you seek her as silver, and search for her as for hidden treasures; then you will understand the fear of Yahweh, and find the knowledge of God.

Proverbs 2:3-5

Try one passage and see

Pick something short: Psalm 1, or the first paragraph of John. Twenty minutes is enough to tell whether this way of reading suits you.

Join the waitlist
Two chairs facing one another, a question rising between them