GTD's Capture Step: The Phase Everyone Gets Wrong (and How to Fix It)
I've been doing GTD wrong for years.
Not because I failed to review. Not because my projects weren't defined. Not because I missed the weekly review.
I was losing the war at the very first step.
David Allen's methodology begins with a single, non-negotiable premise: everything that has your attention must go into an external system. Not most things. Not the things you remember. Everything.
The Capture step sounds simple. It isn't. And the failure mode is almost universal — while being almost universally ignored.
What David Allen Actually Said About Capture
Go back to the original text of Getting Things Done. Allen spends less time on Capture than on any other phase. He treats it as obvious: you write it down, you record it, you get it out of your head.
What he couldn't have anticipated in 2001 was that "writing it down" would eventually mean navigating a touchscreen computer in your pocket during a conversation — a scenario that introduces 7-15 seconds of friction into a moment where the thought has at most 30 seconds before it starts to fade.
Allen's original capture tools were low-tech: index cards, a small notebook, a portable voice recorder. Friction was minimal. You grabbed a card, wrote the thought, moved on.
Today, the average GTD practitioner uses their iPhone as their primary capture device. They reach for it, unlock it, navigate to an app, type, categorize, and file. Sometimes they use Siri. Sometimes they use Drafts or Obsidian or Bear.
All of these tools are improvements over nothing. None of them are as fast as a 3x5 index card.
This is the Capture Gap — and it's quietly destroying GTD systems that are otherwise brilliant.
The Hidden Assumption That's Breaking Your GTD System
There's an assumption baked into most productivity advice that I want to challenge directly:
"Your phone is always available and can always be used."
This assumption is wrong in the moments that matter most.
Think about where you have your most important thoughts, ideas, and insights:
- In meetings: Pulling out your phone signals inattention. Typing notes is permissible; app-switching to your task manager is conspicuous.
- During 1:1s: Same problem, amplified. The other person knows when you're not listening.
- While commuting: If you're driving, your phone shouldn't be in your hand. If you're on a bike, same. If you're on public transit, it's feasible, but you're already context-switching.
- While exercising: Headphones in, phone in an armband or pocket, sweating. Not an ideal capture interface.
- During the hypnagogic state: The half-asleep, half-awake phase. Ideas emerge. Reaching for your phone wakes you fully and often kills the insight.
- In the shower: Your best thinking location, statistically. Zero capture options.
The GTD practitioner who relies exclusively on their smartphone for capture has, by design, excluded themselves from capturing in the moments where the most valuable thoughts occur.
Why Your GTD System Is Losing Tasks You Don't Know You're Losing
Here's the insidious part: you don't know what you're missing.
When a task is captured, it enters your system and you can see it. When a task is lost, it vanishes without trace. You don't get a "Task Not Captured" notification. You just eventually have a vague sense that something was supposed to happen, or you fail to follow up on something you intended to, without knowing why.
I ran a personal experiment for two weeks. I kept a notebook — not a task list, just a log — of every moment I thought "I should capture this" but didn't. Every time I felt the friction of capture and chose not to engage with it, I made a quick hash mark.
In the first week: 47 hash marks. In the second: 39 marks (I was getting better at noticing).
86 uncaptured potential tasks in two weeks. Not all of them were important. But some were. I still wonder what I lost.
The problem with unmeasured loss is that you don't feel it as a system failure. You feel it as forgetting, which gets attributed to memory, which gets attributed to character. "I'm just not a detail person." "I'm bad at follow-through." "I need to be more disciplined."
You don't need more discipline. You need less friction.
The GTD Capture Ladder
Through conversations with hundreds of productivity practitioners, I've mapped what I call the GTD Capture Ladder — a ranked assessment of capture methods by their real-world effectiveness:
Rung 1: Memory
Don't do this. This is the anti-capture. David Allen is explicit: your mind is not a filing system.
Rung 2: Deferred capture ("I'll write it later")
Almost as bad. Works for items in your short-term context (writing in a meeting at your laptop). Fails catastrophically for everything else.
Rung 3: General-purpose notes app (Apple Notes, Google Keep)
Available, fast, always on your phone. Requires processing and migration into your GTD system — creating a two-step capture that rarely gets completed. Notes accumulate; tasks don't.
Rung 4: Your GTD task manager on your phone
Better. Captures directly into your system. Still requires: retrieval, unlock, navigation, input, categorization. In ideal conditions, 7-10 seconds. In the field, often 15+.
Rung 5: Your GTD task manager on your watch
Faster. Drafts on Apple Watch, for example, can capture with 2 taps and voice. Still requires wrist-raising, lock-screen navigation, and screen-checking. Good for simple captures; challenging for nuanced ones.
Rung 6: Dedicated voice capture (always accessible)
The theoretical ideal: a device whose only purpose is receiving captures. No navigation, no authentication, no distractions. Press, speak, release. The capture goes directly to your GTD inbox, pre-processed by AI for context, priority, and categorization.
Most GTD practitioners live on Rungs 3-4. The goal is Rung 6.
The Economics of Imperfect Capture
Let's put actual numbers on the cost of capture failure.
If you run a GTD system, you probably have 100-300 projects, 500-2,000 tasks, and an active mind that generates 15-20 capturable items per day.
With a Rung 4 capture system (phone-based), realistically capturing 70% of those items, you're losing 4-6 tasks per day.
Over a year, at 250 working days: 1,000-1,500 tasks lost.
What's the value of a task? Obviously variable. A low-stakes task (buy more coffee) lost = minor inconvenience. A high-stakes task (follow up with the investor by Thursday) lost = potentially thousands of dollars, a relationship damaged.
If even 1% of your lost tasks represent significant lost opportunities — let's say 10-15 per year — and each of those is worth a conservative $1,000... you're leaving $10,000-$15,000 on the table every year because your capture system operates at 70% efficiency instead of 99%.
The cost of a perfect capture system, amortized over a year, is almost certainly less than the value of the tasks it prevents you from losing.
What Actually Improves Capture Rates
Here's what the research (and our user testing) tells us:
1. Physical immediacy wins. Tools that are already in your hand, or accessible in one motion from your pocket, get used. Tools that require retrieval, unlock, and navigation get skipped when the context is difficult.
2. Social acceptability matters. A tool you can use without it being visible or audible gets used in meetings. A tool that requires you to speak loudly to Siri, or that lights up a screen in a dark conference room, gets abandoned.
3. Zero decisions is better than one decision. Every decision point in the capture flow is a place where friction accumulates. "Where should this go?" is a decision. "What project is this in?" is a decision. The ideal capture tool handles categorization automatically, so you never have to choose.
4. Feedback loops create habits. Tools that provide immediate confirmation ("captured") build the capture habit. Tools that require you to check later whether it worked create uncertainty and abandonment.
5. Failure mode matters. A tool that fails gracefully (partial capture > no capture) is better than one that fails hard (app crash, connectivity loss). Your capture system should be designed assuming that it will sometimes fail.
Building a Zero-Failure Capture System
Here's a practical framework for auditing and improving your GTD capture setup:
Step 1: Map your contexts.
Where do you have thoughts that need capturing? List your top 5. For most people: desk, meetings, commute, gym, home.
Step 2: Assess your coverage.
For each context, what's your current capture tool? Rate its friction (Low/Medium/High) and social acceptability (Easy/Acceptable/Awkward).
Step 3: Find your coverage gaps.
Contexts with Medium or High friction, or Awkward capture methods, are where you're losing tasks. These are your capture failure zones.
Step 4: Design for the failure zone.
For each failure zone, identify a lower-friction alternative. You may need a different tool for different contexts — and that's okay. The goal is coverage, not simplicity.
Step 5: Build the processing habit.
Even a perfect capture system creates a backlog that needs processing. Schedule 10 minutes at the start and end of each day to clear the inbox that your capture tools feed into.
The Capture Device Question
For the past year, we've been building a solution specifically for the hardest capture failure zones: meetings, physical activity, and any context where your phone isn't accessible.
Todovo is a 12-gram button that lives in your pocket. You press it, speak your thought, and release. It sends a processed task — with context, priority, and suggested project already assigned — directly to your GTD inbox of choice, in under three seconds.
No screen. No authentication. No navigation. No social disruption.
It's designed specifically for GTD practitioners who have solved the processing and organization sides of the methodology, but still lose tasks in the capture phase.
If your current GTD capture system captures 70-80% of your tasks, the 20-30% you're losing is worth examining. Not because the tools are bad, but because the contexts they fail in are predictable — and solvable.
We're launching on Kickstarter on July 1, 2026. If you're a GTD practitioner interested in eliminating capture failure, we'd love to have you in the community we're building before launch.
Early access and Early Bird pricing at www.todovo.ai.
The Bottom Line
The Capture step isn't failing because GTD practitioners lack discipline. It's failing because:
- The capture tools we use were designed for desk environments
- The moments that most need capture happen away from desks
- The friction of phone-based capture causes us to skip it in difficult contexts
- We don't track what we lose, so we attribute the loss to personal failure
The fix isn't working harder. It's reducing the activation energy for capture to near zero — in every context, not just the easy ones.
A trusted system is only as trustworthy as its capture layer. And capture fails silently.
Don't let it.
Takahiro Torii is the founder of GRAFFITI, building Todovo: a dedicated capture device for knowledge workers and GTD practitioners. He practiced GTD for 8 years before building hardware to solve the problem that kept breaking his system.
Related: The Hidden Cost of Task Capture