Prepare a decision

Decide your next move with more clarity.

Choose a decision. Add your context. ImpactOS measures your readiness and gives you the next action.

No sign up. Nothing is sent by default.

1

What decision are you preparing?

Everything after this step adapts to the decision you pick, so nothing irrelevant is asked.

This step is required for this scenario.Not started

One decision at a time.

See the full analysis

Optional sections. Open only what is useful for your current decision.

Full command centerEvery area at once, including the ones this decision does not need.
Full diagnosticThe longer guided questionnaire and its detailed result.

Run your own diagnostic, no account needed

Pick a path, answer a short set of fixed questions and get a personalised readiness view straight away. This is a structured decision model, not an AI system. Everything stays in your browser.

1. Choose your path

2. Answer the questions

Decide which role to aim for and what proof is missing.

  • 1.Target clarity

    How precisely can you name the role, level and type of team you are aiming for?

  • 2.Existing proof strength

    What can a reviewer open today that shows you can already do the work?

  • 3.Skill gap severity

    How large is the distance between what you can do now and what the target demands?

  • 4.Network leverage

    How many people already doing this work would review or refer your output?

  • 5.Weekly time capacity

    How many protected hours a week can you actually spend on this?

  • 6.Execution consistency

    Over the last three months, how reliably did you finish what you started?

  • 7.Market alignment

    How well does what you are building match what target postings actually ask for?

Values are self reported. Nothing is saved or sent.
Add your contextNotes kept in this browser only, never uploaded.

Add your context

This beta helps you structure your context. Deep document analysis will be enabled progressively.

Your document

Your file stays on your device. Nothing is uploaded and nothing is stored.

Document type

Accepted formats: .pdf, .docx, .txt, .md, .csv, .xlsx, .json, .yaml, .yml

This text stays in your browser. It is never saved and never sent.

Your context summary

Computed in your browser, with no analysis service.

Describe your situation above to see your context summary here.

Proof portfolioArtefact catalogue and a copyable brief once a diagnostic exists.

Proof Portfolio

Beta

A score is only a snapshot. Proof is what changes how other people judge you. Turn each diagnostic and each sprint into one artefact someone else can open.

No proof brief yet

The Proof Portfolio becomes personalised after a diagnostic.

Run the diagnostic first. The artefacts below stay visible either way, so you can see what strong evidence looks like before you start.

Build my Proof Portfolio
  • Case study

    What it proves:
    That you have already moved a real situation from a starting point to an observable outcome.
    Best for:
    Career, freelance consultant and branding paths
    Expected output:
    One page with context, constraint, decision, action and measured outcome.
    First action:
    Write the outcome sentence first, then work backwards to the decision that produced it.
  • Product brief

    What it proves:
    That you can frame a problem, choose a scope and state what you will not build.
    Best for:
    AI Product Manager and Data Product Manager templates
    Expected output:
    A short brief with problem, user, success criteria, non goals and open questions.
    First action:
    Write the non goals section before the feature list.
  • Market test

    What it proves:
    That demand was observed rather than assumed.
    Best for:
    Founder SaaS and side project paths
    Expected output:
    A test record with audience, offer, channel, sample size, result and stop criteria.
    First action:
    Write the stop criteria before running anything.
  • AI evaluation plan

    What it proves:
    That you can judge quality of an AI feature instead of trusting a demo.
    Best for:
    AI Product Manager and career switch to AI
    Expected output:
    A plan with tasks, test set, failure modes, thresholds and who reviews the results.
    First action:
    List the three failure modes that would be unacceptable, then design a check for each.
  • Data metric definition

    What it proves:
    That you can make a number trustworthy and owned.
    Best for:
    Data Product Manager and analytics oriented roles
    Expected output:
    A definition with owner, source, formula, refresh rule, known limits and consumers.
    First action:
    Name one metric people argue about and write its single definition.
  • Founder validation memo

    What it proves:
    That an idea was tested against the assumption most likely to kill it.
    Best for:
    Founder SaaS and idea validation
    Expected output:
    A memo with riskiest assumption, test run, evidence collected and the kill, pivot, test or build call.
    First action:
    State the assumption in one sentence and the evidence that would disprove it.
  • LinkedIn proof post

    What it proves:
    That the work exists in public and can be found by the people who hire or buy.
    Best for:
    Branding, freelance and job search paths
    Expected output:
    A short public post that links to one artefact and states the outcome, not the ambition.
    First action:
    Publish the artefact first, then write the post around what it shows.
  • Portfolio project page

    What it proves:
    That someone can review your judgement without a meeting.
    Best for:
    Every path once at least one artefact exists
    Expected output:
    One page per project with the problem, the decision, the artefacts and the result.
    First action:
    Create the page with a single project, even an incomplete one, and keep the same structure.
Path templatesStart from a ready made path and jump straight into the diagnostic.

Path templates

What good evidence looks like for five common directions.

  • AI Product Manager

    Template

    Be credible as the person who decides what an AI feature should do, what it must not do and how it is evaluated.

    Required proof
    • One written product spec for an AI feature with explicit failure modes
    • One evaluation plan showing how quality would be measured
    • One shipped or prototyped feature a reviewer can open
    Typical gaps
    • Talking about models instead of user outcomes
    • No evaluation or guardrail thinking
    • No artefact that shows judgement under uncertainty
    Risk signals
    • Every claim is a course certificate rather than a decision you made
    • No opinion on what should not be automated
    • Cannot describe a case where the model was wrong and what you did

    Seven day sprint focus: Publish one AI feature spec with success criteria, failure modes and an evaluation plan.

    Recommended Decision Lens: The sceptical operator

  • Data Product Manager

    Template

    Be trusted to own a data product where correctness, lineage and adoption matter more than volume.

    Required proof
    • One data contract or metric definition document
    • One before and after example of a decision improved by better data
    • One quality dashboard or quality rule set you designed
    Typical gaps
    • No clear metric ownership story
    • Confusing reporting work with product work
    • No evidence of stakeholder alignment on definitions
    Risk signals
    • Cannot name who consumes the data and what breaks when it is wrong
    • Quality is described as a tooling problem only
    • No adoption measure for anything you shipped

    Seven day sprint focus: Write one metric definition with owner, source, refresh rule and known limits.

    Recommended Decision Lens: The accountable owner

  • Founder SaaS

    Template

    Decide whether a SaaS idea deserves the next quarter, using demand evidence rather than enthusiasm.

    Required proof
    • Ten scripted conversations with people who have the problem
    • One live page with a real pricing question and observed responses
    • One manual delivery of the outcome for a single user
    Typical gaps
    • Audience described too broadly to reach
    • No stop criteria written before testing
    • Building before the outcome was delivered manually once
    Risk signals
    • All evidence comes from people who like you
    • No answer to how the first ten users would find it
    • The problem cannot be expressed as a frequency and a cost

    Seven day sprint focus: Run the cheapest test that could prove the idea wrong this week.

    Recommended Decision Lens: The cost of being wrong

  • Freelance consultant

    Template

    Be findable and verifiable for one specific problem, so qualified enquiries arrive without cold outreach.

    Required proof
    • One public claim expressed as a problem you solve
    • One case write up with context, action and observable outcome
    • One reusable artefact a prospect can open before talking to you
    Typical gaps
    • A generalist claim nobody searches for
    • Claims with no attached artefact
    • No repeatable way of reaching the audience
    Risk signals
    • Pipeline depends entirely on one referral source
    • Cannot state a price with a reason behind it
    • Nothing published in the language the buyer uses

    Seven day sprint focus: Reduce the profile to one claim and one proof that supports it.

    Recommended Decision Lens: The impatient buyer

  • Career switch to AI

    Template

    Move into an AI adjacent role by transferring existing strengths instead of restarting from zero.

    Required proof
    • One artefact that applies AI to the domain you already know
    • One written translation of your current experience into the target role language
    • One conversation with someone doing the target role today
    Typical gaps
    • Learning without producing anything openable
    • No bridge between the previous domain and the target role
    • Target role named too vaguely to prepare for
    Risk signals
    • Runway is shorter than the time the switch usually needs
    • Every proof is a tutorial reproduction
    • No one in the target field has reviewed the work

    Seven day sprint focus: Ship one artefact that combines the domain you already know with the direction you want.

    Recommended Decision Lens: The hiring manager with ten minutes

Decision historySaved results and how your readiness moves over time.
Example scenarios, synthetic personasRead only reference examples. Both personas are fictional.

Career Impact Map

Maya, 29, business analyst moving toward AI Product Management

Synthetic persona. No real person, employer or project is described.

Goal: Become credible for AI Product Manager roles within six months

Career Readiness Score

55

of 100, Needs more proof

Evidence Strength Score

42

of 100, Needs more proof

Execution Readiness Score

40

of 100, Needs more proof

Why this score? Career readiness
  • Target clarity4 of 5, weight 3Narrow the target to one role family and one industry.
  • Proof strength2 of 5, weight 3Publish one observable artefact a reviewer can open.
  • Skill gap severity2 of 5, weight 2Close the evaluation and failure analysis gap first.
  • Network leverage2 of 5, weight 2Add two practitioners who can critique the work in public.
  • Time capacity2 of 5, weight 2Protect a fixed weekly block instead of spare time.
  • Execution consistency3 of 5, weight 3Ship something small every week rather than one large plan.
  • Market alignment4 of 5, weight 2Match the proof to what the target postings actually ask for.

What would improve this score

  • Raise skill gap severity: Close the evaluation and failure analysis gap first.
  • Raise network leverage: Add two practitioners who can critique the work in public.
  • Raise time capacity: Protect a fixed weekly block instead of spare time.

Limits

  • Inputs are self reported, so the score reflects what you declared, not verified facts.
  • This is decision support, not career, legal, financial, psychological, medical or investment advice.
  • A high score does not guarantee an outcome. It means the evidence you hold is more consistent.
Why this score? Evidence strength
  • Public artefacts1 of 5, weight 3One published case study would move this the most.
  • Measured outcomes2 of 5, weight 3Attach a number and a method to one shipped result.
  • Third party validation2 of 5, weight 2Get a practitioner critique on the record.
  • Recency of evidence4 of 5, weight 2Keep the newest proof less than three months old.

What would improve this score

  • Raise public artefacts: One published case study would move this the most.
  • Raise third party validation: Get a practitioner critique on the record.
  • Raise measured outcomes: Attach a number and a method to one shipped result.

Limits

  • Inputs are self reported, so the score reflects what you declared, not verified facts.
  • This is decision support, not career, legal, financial, psychological, medical or investment advice.
  • A high score does not guarantee an outcome. It means the evidence you hold is more consistent.
Why this score? Execution readiness
  • Time pressure2 of 5, weight 3Reduce scope per week so the plan survives a busy period.
  • Scope spread2 of 5, weight 3Drop two of the three target roles for now.
  • Proof debt1 of 5, weight 2Convert learning into one artefact each cycle.
  • Feedback loop delay3 of 5, weight 2Ask for critique weekly instead of at the end.

What would improve this score

  • Raise proof debt: Convert learning into one artefact each cycle.
  • Raise feedback loop delay: Ask for critique weekly instead of at the end.
  • Raise time pressure: Reduce scope per week so the plan survives a busy period.

Limits

  • Inputs are self reported, so the score reflects what you declared, not verified facts.
  • This is decision support, not career, legal, financial, psychological, medical or investment advice.
  • A high score does not guarantee an outcome. It means the evidence you hold is more consistent.

Inputs

Current strengths

  • Five years of analysis work on data heavy products
  • Comfortable writing specifications and acceptance criteria
  • Has shipped two internal reporting tools end to end

Missing proofs

  • No public artefact that shows AI product judgement
  • No documented discovery interview set
  • No measurable outcome attached to a shipped AI feature

Constraints

  • Around six hours a week outside current work
  • Prefers a move without relocation
  • Wants to avoid a step back in seniority

Target roles

  • AI Product Manager
  • Product Manager, data platform
  • Technical Product Manager

Next best move

Publish one AI product case study that shows a problem, the option set, the trade off and the measured result.

Top skill gaps

  • Model behaviour judgement, evaluation and failure modeshigh

    Run a small evaluation on an open dataset and write the failure analysis.

  • Pricing and unit economics of AI featuresmedium

    Model the cost per request of one feature and publish the sensitivity table.

  • Stakeholder narrative for technical audienceslow

    Record a five minute walkthrough of one decision and the reasoning behind it.

Proof of work recommendations

  • One written case study with the decision, the alternatives and the measured outcome
  • One working demo that a hiring manager can open in under a minute
  • Six discovery interview notes with a synthesised problem statement
  • One benchmark comparing two approaches on the same task

Personal Decision Pack preview

The one page summary you would take into a decision conversation.

Decision context

Move toward AI Product Management

Maya, 29, business analyst moving toward AI Product Management

Goal: Become credible for AI Product Manager roles within six months

Career Readiness Score
55 of 100, Needs more proof
Evidence Strength Score
42 of 100, Needs more proof
Execution Readiness Score
40 of 100, Needs more proof

Evidence available

  • Two shipped internal tools, self reported
  • Three public job descriptions used to derive the required proof
  • No third party validation of the shipped outcomes yet

Unknowns

  • How target employers weigh public proof against internal experience
  • Whether six hours a week is enough to publish one proof a month

Risks

  • Learning content is consumed but no observable proof is produced
  • The target role definition stays vague, so effort spreads across three profiles

Next best action

Publish one AI product case study that shows a problem, the option set, the trade off and the measured result.

Seven day sprint

  • Day 1Write the target role definition and the proof it demandsOutput: One page role brief
  • Day 2Pick one problem worth a public case studyOutput: Problem statement
  • Day 3Run two discovery conversationsOutput: Two interview notes
  • Day 4Build the smallest demo that shows judgementOutput: Working demo
  • Day 5Measure one outcome and record the methodOutput: Result table
  • Day 6Write the case study and the trade off sectionOutput: Published case study
  • Day 7Ask two practitioners for a critiqueOutput: Two critiques logged

Decision support only. ImpactOS Personal structures your own inputs into an explainable decision view. It is not career, legal, financial, psychological, medical or investment advice, it does not predict outcomes, and every important decision stays yours.

Do not just claim ambition. Build proofs.The proof types that make a next move credible.
  • Public case study

    Medium effort

    Shows judgement, trade offs and a measured result in one artefact.

    What it proves:
    That you can frame a problem, choose between options and report a measured result.
    Best for:
    Career and career switch paths
    First step:
    Write one page: context, options, decision, outcome, what you would change.
  • Portfolio page

    Low effort

    Gives a reviewer one place to verify the claim quickly.

    What it proves:
    That your claims are consistent and easy to check in one place.
    Best for:
    Personal branding
    First step:
    Publish a single page listing three artefacts with one sentence each.
  • Demo project

    Medium effort

    Turns a claim into something a stranger can open and use.

    What it proves:
    That you can ship something a stranger can open and use.
    Best for:
    Career switch to AI and side project
    First step:
    Scope one narrow use case that works end to end, nothing else.
  • Customer interview notes

    Low effort

    Evidence that the problem exists outside your own head.

    What it proves:
    That the problem exists outside your own head.
    Best for:
    Founder and idea paths
    First step:
    Run five short conversations and record the exact words used to describe the pain.
  • Benchmark analysis

    Medium effort

    Shows you can compare options on the same measurable basis.

    What it proves:
    That you can compare options on the same measurable basis.
    Best for:
    Career and interview preparation
    First step:
    Pick three options, define two measures, publish the comparison table.
  • Certification evidence

    Low effort

    Baseline signal only. It supports a proof, it does not replace one.

    What it proves:
    A baseline of vocabulary and fundamentals, nothing more.
    Best for:
    Career switch, as a support signal only
    First step:
    Add it next to one artefact that applies what it covers.
  • Prototype walkthrough

    High effort

    Explains the reasoning behind the build, not only the result.

    What it proves:
    That you can explain the reasoning behind a build, not only the result.
    Best for:
    Interview preparation and founder paths
    First step:
    Record a short walkthrough covering one trade off you made and why.
Decision LensesFixed critical questions, so a decision is stress tested from more than one angle.
Preview concept
  • CEO lens

    Career: If you only get one bet this year, why is this role the highest return use of it?

    Founder: What has to be true for this to be worth two years of your life?

  • Product lens

    Career: Which single proof would remove the most doubt for a hiring manager?

    Founder: What is the smallest version that still creates the outcome people want?

  • CTO lens

    Career: Can you explain the failure modes of the systems you claim to understand?

    Founder: What part of this is genuinely hard, and are you solving that part first?

  • Recruiter lens

    Career: In thirty seconds of scanning, what proves you can already do the job?

    Founder: Who would join you, and what would convince them this is real?

  • Customer lens

    Career: Whose problem do you make smaller, and can they describe it themselves?

    Founder: What are they doing today instead, and why is that tolerable?

Premium betaNo payment is taken. This is an interest signal only.

Premium beta

An early access offer, not a paid product yet.

9 to 19 € per month

Beta pricing intent

Premium is how ImpactOS Personal will be funded. During the beta there is no payment, no card and no subscription. You join a list, you use the features and you tell us what is worth paying for.

What premium unlocks in beta

  • Saved decision history
  • Evidence Graph
  • Sprint tracking
  • Score progression
  • Shareable Decision Packs
  • Advanced path templates

No payment is taken anywhere in this beta. The free diagnostic stays free.

Cohort pilotsIllustrative preview for schools, bootcamps and career programmes.

Cohort pilots

For schools, bootcamps, incubators and freelance communities.

Help a whole cohort make better career and founder decisions with the same evidence model: a structured diagnostic per person, a seven day sprint they can actually track and a Decision Pack a mentor can read in two minutes.

  • A shared readiness language across the cohort
  • Evidence and gaps instead of confidence and opinion
  • Sprint progress a coach can follow without chasing people
  • Decision Packs a mentor can review before a session
Request a cohort pilot

No cohort is running yet. Pilots are being scoped during the beta.

Coach dashboard

Layout preview only, not connected to any data source.

Preview data, illustrative only

Completed diagnostics

48

Sprint progress

41%

Packs shared

12

Cohort readiness distribution

  • Ready to move18%
  • Close, evidence missing34%
  • Direction unclear31%
  • Not started17%

Top gaps

  • Existing proof strength
  • Target clarity
  • Network leverage

Top risks

  • Evidence is self reported and unreviewed
  • Target role named too broadly to prepare for
  • No time reserved for the seven day sprint

These figures are placeholders that show the intended layout. No cohort data exists and no customer is represented here.

Launch readinessHow this beta is being tested before wider release.

Launch readiness

Roadmap targets, not achieved metrics.

What this beta is trying to prove next, in the open. None of these objectives is reached yet, and none of them is presented as a result.

  1. Next 7 days

    Target

    Reach 30 targeted testers.

    Measured by diagnostic completion, not by sign ups.

  2. Next 30 days

    Target

    Validate repeat usage and premium beta intent.

    Measured by saved decisions, returning sessions and qualitative feedback.

  3. Next 60 days

    Target

    Validate Proof Portfolio creation.

    Measured by proof briefs copied and packs shared.

  4. Next 90 days

    Target

    Open one cohort pilot conversation.

    Measured by a pilot intent conversation, not by revenue.

Decision support only. ImpactOS Personal structures your own inputs into an explainable decision view. It is not career, legal, financial, psychological, medical or investment advice, it does not predict outcomes, and every important decision stays yours.

Give feedback on ImpactOS Personal

This is a public beta concept. Tell us whether this would change a real decision for you.

Give feedback or request private beta access

Two minutes. It shapes what gets built next, and nothing you send here is used as demo content.

How clear is the product, 1 to 5

How useful would it be, 1 to 5

Only what you type here is stored. No browser details, no location and no tracking across other sites.

Deciding for a team, not only for yourself?

ImpactOS Enterprise applies the same evidence and readiness model to complex software changes.