Why do notes you take while walking never get reviewed?
Because the capture is the easy half, and it is the half every tool optimises. Recording a thought while walking already takes seconds. What fails is everything after: the note arrives without the thing that prompted it, in a format that is expensive to skim, in a place you have no reason to open, at a moment when you are done thinking about it.
None of that is a discipline problem, which is why resolving to review more does not work. Four separate mechanisms are involved and they have different fixes.
1. A voice memo costs as much to read as it did to record
Text can be skimmed. Audio cannot — to find out what is in a two-minute recording you play it for two minutes, and there is no way to glance at it. So a folder of twelve memos represents about half an hour of listening to find out whether any of it mattered, and that price is paid before you know if it was worth paying.
This is why voice memos accumulate rather than get processed. The fix is transcription, which turns the recording into something you can scan in seconds and search later. Any tool that records your voice without also giving you text is handing you the problem in the shape that guarantees it stays unsolved.
2. The note is a fragment and the context is somewhere else
Thoughts on a walk are almost always about something — an article, a paper, a chapter, an email. What you say out loud is the delta: "this is wrong", "same as the other thing", "the bit about incentives is the interesting part". That is a complete thought at the time because the material is loaded in your head. A week later it is a sentence with no subject.
Reconstructing it means finding the source, finding the place in the source, and rebuilding the state you were in. That is more work than having the thought again, so people have it again, or more often do not. The fix is not a better note — it is keeping the note attached to the exact point in the material that produced it, at capture time, because that link cannot be recovered afterwards.
3. Review needs a moment, and walking notes do not arrive in one
Notes that do get reviewed are usually the ones that land somewhere you already return to for another reason — the document you are writing in, the task list you check every morning, the thread you are already reading. Notes made on a walk land in a general-purpose bucket you open only when you deliberately decide to go through your notes, and that decision competes with everything else.
The fix is arrival, not storage: the note should turn up attached to the thing you will open next anyway. A capture system with no destination is a capture system whose output is optional.
4. The moment you would act on it has already passed
A lot of walking thoughts have a short useful life — a question worth asking while the argument is fresh, a connection worth chasing before you forget what it connected to. By the evening the question is answerable only by re-reading the source, which is the work you were trying to avoid.
This one has no storage fix at all. Either something can be done with the thought close to when you had it, or the thought needed to be more self-contained than a walking thought usually is.
What actually helps, in order
- Transcribe everything you speak. The single highest-value change, and the cheapest — it converts an unskimmable archive into a searchable one.
- Capture against the material, not beside it. If you are listening to something, use a tool that records where you were when you spoke. Anything that files the note separately has thrown away the part you cannot rebuild.
- Give the notes one destination you already visit. Fewer buckets beats better buckets.
- Do the cheap follow-up immediately. If a thought is a question, ask it while you are still walking rather than filing it to ask later.
- Expect to discard most of it. A capture system that assumes everything gets reviewed sets a standard nobody meets and most people then abandon. Most walking thoughts are thinking, not material.
Where Dubs fits, and where it does not
Dubs is a native iPhone app built around the second item on that list. It is in active development and not yet on the App Store.
It plays written material — articles, PDFs, documents, scans, notes you brought, and anything you forward to the personal email address it gives you, which is how most people get newsletters in without thinking about it. A thought captured while it plays is anchored to the exact moment you were hearing, so the note keeps its subject. The reply can be spoken rather than typed, and it can be sent on to an AI agent to continue the conversation about that same material, which is the fourth item: asking the question while it is still live rather than filing it.
It does not help with the third item if your existing notes live somewhere else; it is another destination, not a merge. It does not record meetings or conversations, and it does not subscribe to podcast feeds. And if your walking notes are not about anything you were reading or listening to — if they arrive out of nowhere — then the anchoring this is built on has nothing to anchor to, and a plain voice recorder with good transcription will serve you better.
Questions this does not answer
If you are weighing Dubs against something specific, say what you are trying to do and you will get a straight answer about whether it fits, including when it does not.