The Problem: You Can't Find the Good Stuff in Your Notes
You save articles. You highlight passages in books. You take notes in lectures and meetings. Then, months later, you return to that note and face a wall of text — everything highlighted feels equally important. You can't skim to find the insight that made you save it in the first place.
The fundamental problem: you processed information at one level of compression and now need it at a different level. The note you saved was right for the moment of capture; it's wrong for the moment of use.
Progressive summarization is Tiago Forte's solution: a layered compression technique that makes notes more useful every time you return to them, distilling from "everything saved" to "the most important parts" to "the essential insight" — without losing the layers underneath.
Where Progressive Summarization Comes From
Progressive summarization was developed by Tiago Forte and published as a multi-part series on his Forte Labs blog starting in 2017. The technique became a central component of his Building a Second Brain course and was incorporated into his 2022 book.
Forte's motivation came from a fundamental problem with common note-taking approaches: notes optimized for capture (comprehensive, contextual) aren't optimized for retrieval (scannable, essential). The inverse is also true: notes written for retrieval (lean, summarized) lose the context needed when you want to verify, expand, or share them.
The key insight: Don't choose between capture and compression. Do both, progressively. Capture fully first; compress in layers as you revisit notes, so the level of compression you need is always available.
The "progressive" element is important: you don't summarize everything immediately. You summarize a note the next time it becomes relevant to your work — just-in-time compression rather than just-in-case compression. This ensures that the additional processing effort is spent on notes that have demonstrated usefulness.
The Progressive Summarization Method, Step by Step
Layer 0: Save the Original Content
Capture the full source material — the article, the book chapter excerpt, the meeting notes. No processing at this stage.
Tools for Layer 0:
- Web clipper (WebSnips, Pocket) for articles
- Kindle highlights export for books
- Voice-to-text for meetings or ideas
- PDF annotations
The discipline: Don't filter too aggressively at capture. The point of Layer 0 is to preserve material you might want to access at full fidelity. If in doubt, save it.
Layer 1: Bold the Most Important Passages
On first reading (or first meaningful return), read the note and bold the passages that stand out as most important, surprising, or useful.
The standard: Would you share this passage if someone asked about this topic? Is this the part that made you save it?
Typical ratio: 10-20% of the original content should be bolded. If you're bolding more than 25%, you're applying the layer too liberally; if less than 5%, you may not be engaging critically.
When to do Layer 1: Ideally when you first save something, if the content is clearly important. Alternatively, the next time you return to the note — which ensures Layer 1 compression effort is only invested in notes worth revisiting.
Layer 2: Highlight the Most Important Bold Passages
Return to the note again (on a separate occasion — days or weeks later, typically prompted by relevance to a current project). Of the bolded passages, highlight the most essential ones.
The standard: If someone could only read one line of this note, which line would best convey the essential insight?
Typical ratio: 10-20% of the bolded passages — which means approximately 1-4% of the total original content.
After Layer 2, you should be able to scan just the highlighted passages and quickly understand the note's essential point — even if you've never seen the surrounding content before.
Layer 3: Write an Executive Summary (Optional)
For the most important notes — ones you return to frequently or that are central to active projects — write a 3-5 sentence summary in your own words at the top of the note.
The discipline: Write the summary in your own words, not copied from the bolded/highlighted passages. This tests whether you actually understand the material and creates a new synthesis.
Layer 3 is optional and should only be applied to the most valuable notes — perhaps 5-10% of your total note collection. Applying it more broadly defeats the "just-in-time" principle.
Layer 4: Remix (Express)
Layer 4 isn't a compression step — it's application. When you use the processed note in creative work (writing a draft, building a presentation, making a decision), you're expressing the captured and distilled knowledge. This is where the investment in the earlier layers pays off.
A Worked Example
A product manager saves an article about "Jobs-to-be-Done theory in product design." Here's how progressive summarization works across multiple returns:
Layer 0 (Save):
Full 2,500-word article saved to note-taking app. "Jobs to be Done" folder. No processing.
Layer 1 (Two weeks later — working on user research):
PM reads the note while researching. Bolds:
- The core definition of JTBD: "People don't buy products; they 'hire' products to accomplish a job they need to get done."
- The milkshake example (Clayton Christensen's classic illustration)
- The distinction between functional, social, and emotional jobs
- The interview technique recommendation for uncovering jobs
Approximately 15% of the original content is bolded.
Layer 2 (One month later — using JTBD in a presentation):
PM returns to the note for a product strategy presentation. Highlights:
- The core definition sentence
- The functional/social/emotional job distinction
Two passages from 2,500 words — the irreducible essence.
Layer 3 (After the presentation proves effective):
PM writes a 4-sentence summary at the top: "JTBD reframes product decisions around the job customers are trying to accomplish rather than the features they request. The key insight is that the same 'job' may be accomplished by very different products (a milkshake 'hired' for a morning commute). Jobs have functional, social, and emotional dimensions — missing any dimension misses part of why customers 'hire' or 'fire' your product. Interview technique: ask what customers were doing before they found your product, not what features they want."
When the PM needs JTBD insights six months later:
They scan the top of the note (Layer 3 summary: 4 sentences), confirm relevance, and if needed read the highlighted passages (Layer 2). The full content is available beneath, but they rarely need it. The note is genuinely useful, not just filed.
Where Progressive Summarization Shines
For active knowledge workers with large note collections: If you accumulate hundreds of notes regularly, progressive summarization prevents all of them from becoming equally useless. It ensures the most valuable notes are the most processed.
For people who revisit notes: The technique rewards revisiting — each return adds a layer of processing, making the note more useful. If you never return to a note, progressive summarization doesn't help (and that note probably should have been saved).
For long-form content (articles, books, reports): Layer 1 and 2 processing is most valuable for content dense enough that skimming is necessary. Short, already-essential notes don't benefit as much.
Where Progressive Summarization Breaks Down
For people who never revisit notes: The just-in-time model requires returning to notes. If your workflow never prompts revisiting — if you save and never look back — the layers never get applied and the value is never realized.
Over-summarization risk: Applying Layer 3 to too many notes creates a second problem: your executive summaries become the new wall of text. Reserve deep distillation for the most valuable notes.
Tool dependency: Progressive summarization works best in tools that support both bold and highlight formatting (Notion, Obsidian, Bear, Evernote). It's awkward in plain-text systems without formatting.
Doesn't replace understanding: The technique processes existing knowledge; it doesn't create it. If you saved a note you don't understand, progressive summarization of your confusion doesn't help.
Tools That Support Progressive Summarization
| Tool | Bold support | Highlight support | Notes |
|---|
| Notion | Yes | Yes | Good for BASB workflows |
| Obsidian | Yes (Markdown bold) | Yes (with callouts or ==highlight==) | Great for linked notes |
| Evernote | Yes | Yes | Long-time BASB tool of choice |
| Bear | Yes | Yes | Clean, Apple-native |
| Readwise | Yes | Yes | Designed for book/article highlights; syncs across apps |
| WebSnips | Via initial clip selection | Focused clipping (select only what matters) | Best at Layer 0 and focused capture |
WebSnips and progressive summarization: WebSnips handles an important variant of Layer 0 — instead of saving the full article and highlighting later, you clip only the specific passage that matters at save time. This means Layer 0 capture in WebSnips is already more focused than typical article saves, reducing the work needed at Layer 1.
Common Progressive Summarization Mistakes
Mistake 1: Summarizing everything at once.
The just-in-time principle is essential. Summarizing notes immediately after saving them — before you know which ones will be revisited — is wasted effort. Let the next time you need a note trigger its first layer of processing.
Mistake 2: Bolding too much (Layer 1 inflation).
If 60% of a note is bolded, the bold adds no signal — everything is equally important. Force yourself to the 10-20% ratio. Bolding less than you want to feels uncomfortable at first; it's the right constraint.
Mistake 3: Skipping Layer 2.
Layer 2 (highlight) is where the real distillation happens. Many people apply Layer 1 consistently but never return for Layer 2, resulting in notes with many bold passages but no clear essential insight. Schedule a Layer 2 pass on your most frequently returned-to notes.
Mistake 4: Using progressive summarization as an end in itself.
The whole point is Layer 4 — expression, producing work from your notes. Progressive summarization that never feeds into creative output is just a tidier way to collect information you don't use.
Frequently Asked Questions
How long should each layer take?
Layer 1 (bolding while reading): 5-15% added time vs. reading alone for long-form content. Layer 2 (highlighting): 5 minutes for a long note. Layer 3 (summary): 10-20 minutes for a complex note. These are estimates — the actual time depends on content density and how deeply you're engaging.
Should I apply progressive summarization to all my notes?
No. Layer 0 and Layer 1 are applied to notes you've judged worth keeping. Layer 2 is applied when you return to a note a second time. Layer 3 is applied only to the most important notes. The technique is a filter that differentiates note quality over time, not a uniform processing requirement.
Can progressive summarization work with physical notes?
Yes, with adaptation. Use a pencil for Layer 1 (underline or bracket passages), a highlighter for Layer 2, and write a summary on a sticky note at the front. The layering principle works; the tools differ.
Key Takeaways
- Progressive summarization is Tiago Forte's layered note-distillation technique: save fully (Layer 0), bold the best passages (Layer 1), highlight the best of those (Layer 2), write a summary (Layer 3 — optional), then use them (Layer 4).
- Just-in-time, not just-in-case: process notes when you return to them, not immediately at capture — this focuses effort on notes that have demonstrated usefulness.
- 10-20% ratios: Layer 1 should bold 10-20% of the original; Layer 2 should highlight 10-20% of the bold — the compression ratio forces selective judgment.
- Layer 3 is for the most important notes only — perhaps 5-10% of your collection.
- Tool support matters: needs bold and highlight formatting; works best in Notion, Obsidian, Evernote, Bear.
- The goal is Layer 4 (expression) — if your processed notes never feed into creative work, the earlier layers haven't paid off.
Conclusion
Progressive summarization is one of the most practical note distillation techniques for knowledge workers who accumulate more information than they can immediately process. By adding compression incrementally — each return to a note adding a layer of distillation — the technique makes notes more useful over time without requiring a time-intensive upfront processing pass. The insight is elegant: your future self, returning to a note under time pressure, needs a different level of detail than your past self, saving it with full attention. Progressive summarization serves both.
Try WebSnips free — your web clipping is your Layer 0 capture; by clipping specific relevant passages rather than full articles, you start the progressive summarization process with more focused raw material.