Counts show all available guides. Filters combine to narrow your results.
Create work & incomeGuide
Break a large idea into milestones someone can actually use
Plan useful end-to-end increments before assigning dates or estimating effort.
Illustrative resourceNot tool-testedAdults working with non-sensitive usable milestones information
What you’re working towards
A small sequence of independently useful deliverables with exclusions and review evidence, before dates or effort are estimated. This is an authored practice example; no consumer AI run has been recorded.
Your choice also saves this guide’s current worked-example checkmarks. Only this browser; no account or sync. View saved guides
Saving needs JavaScript. Reading and downloading this guide does not save it.
Make a start
One step at a time.
4 steps
Start with the synthetic worked example or other non-sensitive inputs. Run one prompt at a time in your chosen AI tool. Replace any bracketed placeholders with the requested material, check each reply, and supply the source or earlier outputs that step needs. Paste them again if the tool lacks that context.
Start with the synthetic example, then adapt each prompt to your task.
01
Try the complete example
Open a new chat in a text AI assistant you already have permission to use. Paste all of Input from this guide, then the prompt below, in one message. Do not paste Output: that is the illustrative answer for comparison. Expect a written result that will plan useful end-to-end increments before assigning dates or estimating effort. Different wording is fine; the source facts and rules must agree.
Your starting prompt
Propose the three requested usable increments. For each give the complete reader task it supports, included scope, explicit exclusions, dependencies and concrete acceptance evidence to collect. Preserve all 12 resource IDs by reference; do not invent entry content. Keep tests proposed, flag missing search/build decisions and do not estimate dates or build anything.
Use only the complete synthetic input pasted above. Return the finished written result. Keep missing facts unknown. Do not send, publish, order or change another service.
02
Test whether each milestone stands on its own
Read each milestone without borrowing a later milestone’s feature. Ask whether a reader can both find a resource and read its contact route at that point. If an increment only delivers an internal component, ask for a usable reader-facing increment instead. No actual reader test has occurred in this guide.
Your starting prompt
Review the three proposed milestones from the synthetic reader task. For each, state how a reader can find a resource and read its approved contact route without later features. Flag any hidden dependency on a future milestone. Return corrected scope and proposed review evidence without inventing a build or test result.
03
Trace the answer back to the source
Compare the answer with Input yourself. M1, M2 and M3 each support finding a resource and reading its approved contact route. All milestones preserve the same 12 entries; M1’s three categories have four IDs each. No dates, effort, test success, actual directory entry or publishing permission is invented. Search rules remain an explicit dependency of M3 rather than assumed behavior. If a check fails, paste: “This part is wrong: [quote it]. My source says: [quote the exact line]. Return the complete corrected result and explain the change.” Recheck before saving; if the source itself is incomplete, keep that part unresolved.
Your starting prompt
Audit the result against the source pasted in this chat. Check these specific points:
- M1, M2 and M3 each support finding a resource and reading its approved contact route.
- All milestones preserve the same 12 entries; M1’s three categories have four IDs each.
- No dates, effort, test success, actual directory entry or publishing permission is invented.
- Search rules remain an explicit dependency of M3 rather than assumed behavior.
Show each mismatch and return a complete corrected replacement. Distinguish your text audit from checks that only a person or the actual application can perform.
04
Use your own input
For your own version, open a new chat. Paste the prompt below after replacing every bracketed field. Replace related IDs, quantities, dates, rules and limits together; none of the practice facts carries over. Expect the same named deliverable using your source. Review it yourself: walk through the plan on paper. For each milestone, identify the reader task, the materials it would need, and what would show that it works. Ask for a revision if a milestone is only a disconnected piece with no usable path. Run real reader tests only after the relevant deliverable exists; this guide creates the plan. If a check fails, paste: “This part is wrong: [quote it]. My source says: [quote the exact line]. Return the complete corrected result and explain the change.” Recheck before saving; if the source itself is incomplete, keep that part unresolved.
Your starting prompt
Split my project into a small sequence of independently useful deliverables before estimating effort or dates. For each, state the end-to-end user outcome, included and excluded scope, prerequisites and concrete acceptance evidence to collect. Preserve my constraints and existing material. Reject milestones that supply only an internal component with no usable path; mark unknown decisions and all unperformed tests clearly.
MY INPUT
Product or service goal: [replace with your own complete non-sensitive details, or state unknown]
Intended user task: [replace with your own complete non-sensitive details, or state unknown]
Must-have end-to-end behavior: [replace with your own complete non-sensitive details, or state unknown]
Known restrictions: [replace with your own complete non-sensitive details, or state unknown]
What is already available: [replace with your own complete non-sensitive details, or state unknown]
Out-of-scope work: [replace with your own complete non-sensitive details, or state unknown]
Use only MY INPUT for facts and constraints. Ask for an essential missing fact or conflicting rule before deciding; do not import practice-example names or numbers. Return the complete result plus unresolved questions and a source-based check list. Do not send, publish, buy, schedule or change another service.
Optional: improve the result
Reduce the first delivery without losing usefulness
Explore a smaller approved scope as a proposal rather than quietly dropping records. This optional extension uses only the synthetic example. Return to the sample chat, or paste its full Input into a new chat first. Do not transfer the changed sample values to your own input.
Advanced prompt
For this synthetic project, the owner asks to consider an initial Drawing-only directory with R01–R04. Compare that option with the original complete M1. State what readers can do, what is deliberately missing, and what explicit approval would be needed before changing the 12-resource first-delivery scope. Do not treat the reduced scope as adopted.
Show, then try
See the shape of the result.
This input is synthetic. The output is a written illustration of the intended result, not a verified AI-tool run.
01 / The starting point
Synthetic input
Synthetic practice input. All names, products, records and policies below are invented.
Fictional adult club wants a public directory of 12 approved workshop resources. Reader task: find a relevant resource and read its approved contact route. Existing source is one reviewed text file with IDs R01–R12; each has a name, short description, one approved category and an approved contact route. Categories: Drawing R01–R04, Paper craft R05–R08, Printmaking R09–R12. The source content itself is not needed or supplied for this planning task; do not write resource entries.
Owner wants three proposed usable increments: categorized text first, browsable page next, then search. No accounts, member submissions, payments, analytics or new contact details. No dates, duration estimates, hosting choice or publication authorization supplied. Need outcome, scope, exclusions, dependency and review evidence for each increment. All tests described must remain proposed until performed.
Copies text only. For image or file tasks, use the source files named in the steps too.
Illustrative answer authored for this guide; not a recorded AI response.
Proposed milestone sequence; no build, test or publication has occurred.
M1: A complete categorized text directory
Reader outcome: browse a category, find a resource and read its approved contact route without needing a website.
Scope: one readable text document containing R01–R12 under the three supplied categories, preserving the reviewed names, descriptions and routes.
Dependency: obtain the actual reviewed source file before drafting entries.
Proposed acceptance evidence: compare all 12 entries with source; confirm each ID appears once, with four entries in each category. Ask a reader to find one item in each category and locate its contact route using only this document. Record actual results separately.
Excluded: webpage, search, invented entries or publication.
M2: A browsable page containing the same directory
Reader outcome: reach a category and read the relevant resource and contact route in a page preview without search.
Scope: implement the approved M1 content with clear headings and navigation to the three categories.
Dependency: reviewed M1 content and a separately chosen page-building/preview approach. No platform is assumed here.
Proposed acceptance evidence: reconcile all 12 entries with M1; try category navigation, reading order, keyboard navigation and contact-route presentation in the actual preview. A reader must complete the task using this milestone alone. Record problems before publication review.
Excluded: accounts, member submissions, payments, analytics, search and deployment authorization.
M3: Add search with a useful no-match state
Reader outcome: search within the approved directory, open a matching entry and read its contact route, or understand that no match was found and return to browsing.
Scope: agree searchable fields and matching rules, add search across the same 12 approved entries, preserve the M2 browsing route and provide a clear/reset action.
Dependency: reviewed M2 page and owner-approved search rules; those rules are not yet supplied.
Proposed acceptance evidence: test a known matching query, a no-match query, an empty query and clearing a query under the approved rules. Recheck all contacts and ordinary browsing, including the actual keyboard flow.
Excluded: external search results, inferred contact details, accounts, submissions, payments or analytics.
Before estimating or scheduling
Supply the reviewed file, choose how to build/preview the page, agree search semantics, assign real reviewers and obtain any later publishing approval. Dates and effort remain unknown. Each increment provides the full find-and-read task at its own level; “database done” or “design done” alone would not satisfy the reader outcome.
These checks use the worked example and its particular facts. With your own material, check your actual names, numbers, sources and rules; do not reuse the sample’s values as the answer.
Unsaved checks stay in this tab and reset when you reload. If you choose Save guide, its current example checkmarks are kept in this browser when storage is available.
!
A few things to watch for.
A milestone named after an internal layer may deliver nothing a reader can use. Describe the complete user task.
Acceptance criteria are questions to test, not evidence that a test passed. Keep actual observation separate.