Module 2 of 9~30 minutes

Planning & Knowledge Management

Most people do not have a memory problem; they have a retrieval problem. Give each kind of thing one home, and give AI enough context to help without handing over more than the task needs.

Objectives

Tell a capture problem apart from a retrieval problem in your own week.
Choose one place where a given kind of information will live, and say why.
Write a note that your future self can actually act on.
Turn a messy collection of material into a plan with an order to it.
Judge which parts of your own information a task genuinely needs.
Remove unnecessary detail from what you share without removing the meaning.
Check an AI-produced plan against the reality it will meet.

Introduction

Module 1 ended on a problem it could not solve: you found one improvement worth making, and then your own information got in the way. The note was in one place, the decision in another, and the thing you wanted to reuse was inside a message you could not find.

This module is about the shape of your own information, not about an app. Most people do not have a memory problem; they have a retrieval problem. The thing was written down. It just was not written down anywhere a future version of them would look.

The skill here is deciding, before you save anything, where the future you will go looking — and giving AI enough of that material to be genuinely useful without handing over more than the task needs.

Lesson 1

Capture Is Not the Problem

Almost everyone captures. Far fewer retrieve. The loss does not happen at the moment of writing something down — it happens weeks later, at the moment of looking, when the place you look and the place it lives turn out to be different.

That is why buying a better capture tool so rarely helps. The tool was never the failure.

A capture problem sounds like: I never wrote it down.
A retrieval problem sounds like: I know I wrote it down somewhere.
The second is far more common, and the fix is different.

Lesson 2

One Home Per Kind of Thing

The instinct is to look for one system that holds everything. That instinct is why so many systems get abandoned in week three — everything is a lot, and a home for everything is a home for nothing.

The working version is smaller. For each kind of thing you handle repeatedly, decide one place it lives, and go there without thinking. Three homes, not thirty. A learner with a paper notebook and a phone has enough.

Something to do — where will I look when I ask "what now?"
Something to remember — where will I look when I ask "what did we decide?"
Something to reuse — where will I look when I ask "how did I do this last time?"

The test is not tidiness; it is whether you can answer each question without searching.

Lesson 3

Notes Your Future Self Can Use

A note titled "meeting notes" fails for a specific reason: it records that something happened, not what you are now supposed to do about it. Six weeks later it is archaeology.

The repair is small. Write the decision and the next action, in the words you would use if someone asked you about it out loud.

Write what was decided, not what was discussed.
Write the next action as something you could start without re-reading anything.
Write it for a version of you who has forgotten the context entirely — because you will have.

Lesson 4

From Pile to Plan

A pile becomes a plan in two separate steps, and it matters that they are separate. Sorting is grouping like with like, removing duplicates, and noticing that two items are the same task described differently. Ordering is deciding what comes first.

AI is genuinely good at the first and genuinely poor at the second, because ordering depends on what your week can absorb, what you owe other people, and what you are willing to let slip. None of that was in the pile.

Sort first: group, deduplicate, merge the items that are secretly the same task.
Then order: first, next, later — with a reason attached to the first item only.
If you cannot say why the first item is first, it is not first.

Lesson 5

Minimum Necessary Context

AI gets better as you give it more of the situation. That is true, and it is where people quietly overshare — because the easiest way to give context is to paste everything.

Four questions, applied in order before anything leaves your hands: Information → Is it needed? → Is it sensitive? → What is the minimum that carries the meaning?

Read that as four questions, not a diagram to memorize. "Needed" removes what is merely nearby. "Sensitive" flags anything about your health, money, employer, or another person. "Minimum" is the interesting one: it usually turns out you can keep the shape of the situation and drop the identifying detail entirely.

Needed — does the task actually require this, or is it just nearby?
Sensitive — does it concern health, money, an employer, or another person?
Minimum — what is the least that still carries the meaning?

"Reschedule the Thursday appointment for my mother, who has a cardiology follow-up at St. —" becomes "Help me draft a short, polite message rescheduling a medical appointment to the following week." The second version gets you the same draft. It also does not put a relative's health information — which they never agreed to share — into a general AI tool. Never paste passwords, keys, account or card numbers, medical or financial records, or confidential work material.

Lesson 6

Checking a Plan Against Reality

A plan that ignores your actual week is a wish. The check takes one question: does the first item survive Tuesday?

If your plan assumes a week you do not have, revise the plan, not the week. This is the same honesty Module 1 asked for about overhead, applied to sequencing instead.

Read the plan against a real week, including the parts you do not control.
Revise the plan when it collapses — that is the plan working, not failing.
A tidy, sensible plan that dies on Wednesday was missing your constraints, which were never supplied and could not be inferred.

Practical exercise

≈6 min

One pile, one plan. Pick one genuine pile — a backlog of messages, a half-finished project, a set of notes from something you attended, a list of things you keep meaning to do. Real, not tidy.

Name the three homes the material belongs to — to do, to remember, to reuse — and sort each item into one. Items that fit none usually turn out to be items you can drop.

Apply the context filter to what remains: needed, sensitive, minimum. Write down what you would remove before sharing any of it.

Produce an ordered plan from the "to do" pile — first, next, later — with a reason attached to the first item only. Then check it against your actual week: does the first item survive Tuesday? Revise the plan if not.

You keep one sorted pile, one context-filtered version, one ordered plan with a stated reason, and one revision made after the reality check. Nothing is submitted or stored. If you would rather not use your own material, or cannot right now, use the supplied pile below. Most people find that ordering was the hard part and sorting was the easy part — and that AI is far better at the sorting.

Key takeaways

The common failure is retrieval, not capture — the note existed, just not where you looked.
Give each kind of thing one home, chosen by the question you will ask: what now, what did we decide, how did I do this last time.
Write the decision and the next action, for a version of you who has forgotten the context.
AI sorts well and orders badly, because ordering needs facts about your life that were never in the pile.
Filter before you share: needed, sensitive, minimum — the shape of a situation is usually enough.

Practice in Prompt Lab — optional

Optional

Practise turning a messy list into something with an order.

Paste a deliberately messy, invented list into Prompt Lab — a dozen half-sentences of the kind you would actually write to yourself — and ask for it to be grouped. Read the grouping, then do the part it cannot do: put the groups in the order your week demands, and write one sentence saying why the first one is first. Use invented material, not a real backlog; the skill transfers and the personal details never leave your hands.

Optional, and never required. Prompt Lab is not GlobSynk Lab™. Opening it is not part of module completion, the checkpoint, course progress, the Final Assessment, the certificate or Reward Points, and a learner who never opens it completes this course exactly as normal. Academy never depends on Prompt Lab being reachable.

Practice Prompting in the Real World

Opens in a new tab. Optional practice — never required for this module, the checkpoint, your progress, the Final Assessment, the certificate or Reward Points.

Before you move on

You can now find your own material and turn a pile into a plan with an order you can defend.

The next thing you will notice is that some of those items are not really separate tasks at all. They are the same sequence of steps, arriving again — every Monday, every invoice, every time a particular kind of message lands. Planning them individually is itself busywork.

Recognizing a repeatable process, and preparing it once instead of every time, is Module 3.