What Is Progressive Summarization?
You highlighted the whole article, so six months later you open the note, see 400 words of yellow, and learn nothing. Highlighting everything is the same as highlighting nothing.
Progressive summarization is a method for compressing a note in layers — bolding the best lines, then highlighting the best of those, then writing a one-line summary — so a future read takes seconds, not minutes. It was named by Tiago Forte, and the core idea is simple: distill a note over time, only when you actually revisit it, instead of all at once when you first save it.
The five layers
Each layer is a smaller, denser subset of the one before it. You do not create them in one sitting — you add a layer each time the note proves useful enough to reopen.
| Layer | What you do | When |
|---|---|---|
| 1 | Save the source text (article, quote, transcript) | On capture |
| 2 | Bold the passages that carry the meaning | First real read |
| 3 | Highlight the best of the bolded passages | When you reuse the note |
| 4 | Write a one- or two-line executive summary in your own words | When the note earns it |
| 5 | Remix: turn the note into something new (an outline, a post, a decision) | Rare — only the best notes |
The discipline is in the ratios. A good target: roughly 10–20% of layer 1 gets bolded, and 10–20% of that gets highlighted. If you bold half the note, you have not summarized — you have recopied.
Why layers beat one big summary
Two reasons, both about cost.
- You summarize on demand, not on capture. Most saved notes are never reopened. Summarizing every note up front spends effort on notes that will never pay it back. Progressive summarization defers the work to the notes that prove their value by being reread.
- You preserve the original. A summary written today answers today's question. Layering keeps the full source underneath, so when a different question arrives next year, you can re-distill from the raw text instead of being stuck with a summary aimed at the wrong target.
The result is a note you can opportunistically scan: read the highlights in three seconds, drop to the bold for context, fall back to the full text only when you need it.
A concrete example
You save a 1,200-word piece on interest rates. On first read, you bold four sentences that state the actual mechanism (layer 2). A month later you cite it in your own notes, so you highlight the single sentence that mattered most (layer 3) and add a one-line summary at the top: "Rate cuts lag the economy by ~12 months — don't trade the announcement" (layer 4). Now the note is a five-second read with a 1,200-word backstop. You never wrote a formal summary; the note compressed itself through use.
Where it matters
Progressive summarization is most valuable for a second brain — a personal knowledge base you expect to search and reuse for years. (A second brain is an external store of notes that holds what your memory won't.) The payoff compounds: the notes you reopen most get the most distillation, so your library self-organizes around what is actually useful, with zero up-front filing decisions.
It is a poor fit for two cases: notes you will act on within a day (just act), and reference data like phone numbers or commands (no meaning to distill). Use it where understanding is the goal, not retrieval of a single fact.
Try this
Open one note you saved weeks ago and never reread. Do only layer 2: bold the five sentences that carry the meaning, delete nothing. That single pass usually cuts the reread cost by more than half — and you will know within thirty seconds whether the note was worth keeping at all.
In JustJot.ai, bold and highlight live in the editor and stay searchable, so a distilled note surfaces in semantic search by what it means, not just the words you bolded — the layers make your best thinking the easiest thing to find again.