Skip to content

Rung 4 · Automation mode · Paid Assessment

One real task. One Claude-native workflow. Forty-eight hours.

A$1,200. Credited in full against any follow-on engagement with us within 60 days.

Concurrency capped at 2 per week. Next available slot confirmed within 4 business hours.

What we deliver in 48 hours

No slide deck. No strategy document. A runnable thing, on one of your real tasks, with numbers attached.

A working Claude-native workflow

Not a slide deck. A runnable workflow on one of your real tasks, in your house style, with documented setup so your team can run it end-to-end without us in the room.

A blind accuracy score

We compare the draft output against a signed-off comparable you already have. You get the percentage, the distribution of differences, and the edges we found.

A time-saved calculation

Hours saved per task × your annual volume. The business case in numbers — not in feelings.

A one-page ROI summary

Board-ready. Covers the opportunity, the risk, the 4D Framework classification of this workflow, and the three things we'd do next.

How it runs

  1. 00

    You book + pay — A$1,200

    Stripe checkout. Receipt is emailed immediately with a calendar link to book your kickoff call within 24 hours.

  2. 01

    45-minute kickoff call

    Willie + Paul on our side. You + one person who actually knows the workflow. We agree the one real task, collect the raw inputs and signed-off comparable, and sign NDAs if needed. 48-hour clock starts here.

  3. 02

    48 hours of build

    We build the Claude-native workflow. We score it blind against your signed-off comparable. We document the setup. We calculate the business case.

  4. 03

    Delivery + 30-minute debrief

    You get the workflow + score + ROI summary at the 48-hour mark. We debrief by day 7 on what you found. From there: Rung 5 (productised build) or we part ways cleanly.

Who this is for

A good fit if

  • — You have a recurring task that takes 5+ hours each time
  • — You have examples of what "good" output looks like (signed-off work)
  • — A 5-15% error rate on first draft is recoverable — a human reviews before it ships
  • — You want to see the actual thing before committing to a retainer
  • — You're in a regulated AU industry (insurance, health, finance, government, professional services)

Not a good fit if

  • — You want an autonomous agent that needs no human review (that's Rung 7 — we'd ladder you there through 4→5→6 first)
  • — You want a strategy deck or "AI roadmap" — we don't sell those
  • — The task changes too much to have a signed-off comparable
  • — You're shopping prices without a specific task in mind
  • — You want a different foundation model — we're Claude-exclusive

Questions

What's the A/A/A mode of this?

Automation. We build a specific scripted workflow on precise instructions. If Rung 4 lands, Rung 5 (productised Automation) or Rung 6 (Augmentation pod) is usually the right next step — Rung 7 (Agency managed agents) comes later, once Automation is bulletproof.

Do I need an NDA first?

We'll send one within 4 business hours of your request. Or use yours — we'll sign most standard AU mutual-NDAs without redline.

What if the workflow doesn't score well?

You still get the blind accuracy score, the time-saved calculation, and a clear read on why. That's useful even if we don't proceed — you know whether it's a Description problem (fixable), a Discernment problem (harder), or a fundamental mismatch.

What if I want to use a different foundation model?

We don't. Adaptation AI is Claude-exclusive — that's the boutique specialisation. If a different model is genuinely right for you, we'll say so and refer you. We'd rather refer you away than deliver a mediocre result on the wrong tool.

How is the A$1,200 credited?

Full A$1,200 credits against any follow-on engagement with us within 60 days. No catch — it's priced as a commitment mechanism, not a revenue centre.

Who delivers this?

Willie Prosek personally delivers the first two assessments each week. After that, our engineering team delivers under Willie's review. Concurrency is capped at two per week until the process is bulletproof.

What happens after I pay?

You'll get an email with a kickoff booking link within 5 minutes. We ask you to book your 45-minute kickoff call within 24 hours, and send your one real task (raw inputs + signed-off comparable). 48 hours starts at the kickoff call.

Ready?

Two slots per week. Next available confirmed within 4 business hours of purchase.

Prefer to talk first? Book a 20-minute call — or email hello@adaptation.ai.