← Blog

Your Essential Eight maturity level does not carry over

ASD's Essentials replaces the maturity level with an effectiveness rating per action and domain, and says the one does not convert into the other. What that changes about what a rating claims.

On 2026-10-15 the Australian Signals Directorate published the ASD Essentials, which it describes as “an evolution of the Essential Eight”. Most of what is new in it is structure: principles, actions and domains, and a chapter each for enterprise IT, cloud and operational technology.

The more important change is in how it is measured. ASD’s frequently asked questions put it in one sentence: “Maturity level ratings under the Essential Eight maturity model do not automatically convert into effectiveness outcomes under the Essentials series.”

What the two models measure

The same answer says why. “The Essential Eight maturity model measured implementation of specific controls, while the Essentials series measures whether outcome-based actions are effective for the assets within each specific domain.”

A maturity level was a statement that a set of controls had been implemented across the environment, to a level chosen against the tradecraft of the adversary you expected. Under the Essentials, organisations no longer select a target level. Each action is rated in each domain it applies to, on a scale that asks whether the action achieves its intended outcome for the assets in that domain.

ASD’s guidance on transitioning from the Essential Eight is direct about what that means for work already done. Organisations that reached Maturity Level Two can expect many of their controls to contribute to actions in the early priority phases. But they “should not assume that previous achievement of Maturity Level Two means an action will be fully effective in every applicable domain”. It gives a similar caution for Maturity Level Three.

None of this makes an Essential Eight programme wasted. ASD says plainly that the investment remains valuable and that organisations should continue their uplift programmes while they assess alignment. What does not carry over is the rating. The controls are still there. The claim about them has to be made again, in the new terms.

What a rating now has to say

The assessment process guide sets out what those terms are. Each applicable action is examined on four dimensions: coverage, consistency, currency, and the quality of the evidence. Currency is defined as “whether the evidence is recent enough to reflect the current implementation and operating intent of the action”.

Three cautions in the guide are worth reading slowly, because each is a common way a rating ends up claiming more than it can show.

  • “Deployment is not effectiveness.” Evidence that a control is configured or enabled does not prove it achieves its outcome.
  • “Evidence must be current.” Older evidence may no longer reflect the environment, and the assessor is told to seek more recent evidence where needed.
  • “Controls may only partially satisfy an action.” A control mapped to an action may cover only part of its outcome.

The guide also says what a mapping from an existing framework is for. It “should be used as a starting point for implementation and assessment activities”, not as evidence that the outcome has been achieved.

And it is specific about the bottom of the scale. Where the evidence is not enough to decide, the action is rated not assessed and the limitation is recorded. Not assessed “must remain distinguishable from not effective in all reporting”. Partial activity, “such as work commencing or a plan existing, is not partial effectiveness”. Within a domain, a principle takes the rating of its weakest action.

Taken together, a rating under the Essentials is a claim about a scope, at a time, resting on evidence of a stated quality. When the evidence ages, the rating should be looked at again. Read that way, a rating shows how much of it rests on what is happening now, and how much on what was true when someone last looked.

Why the maturity level was the weaker claim

A maturity level was never meant to be a permanent property of an organisation. It was a description of implementation at the point it was assessed. The difficulty is that a single number is easy to repeat. It can go into a board paper or a supplier questionnaire and keep being repeated after the assessment that produced it has stopped describing the environment.

That is the same gap we described in the eleven months your certificate isn’t watching: an assessment is accurate on its date, and the statement made from it outlives it. The Essentials does not remove that gap. Any assessment is still taken at a time. What it does is make the gap harder to hide, because currency and evidence quality are part of the rating rather than something a reader has to ask about separately.

What this means for the transition

ASD has not yet set dates. Its FAQ says it has been talking to regulators and “will advise on expectations for transitioning to the new framework in due course”. In the meantime, organisations should keep meeting whatever Essential Eight obligations already apply to them, and follow any direction their regulator gives. ASD also advises against starting work solely to reach a superseded Essential Eight target, unless a regulator or other obligation still requires it or it addresses a material risk.

That leaves a practical first step, and ASD names it: a transition baseline. Record your current posture, informed by your Essential Eight maturity, your existing controls, your risk assessments and your key business functions. Then scope the assets and functions in the first priority phase’s domains.

The useful discipline in that step is to keep two things apart. One is what your Essential Eight work suggests about an action. The other is what you can show about it today, in that domain. Where you cannot yet show anything, the honest rating is not assessed, and ASD’s guide asks that it be reported that way.

What Frank does with the Essentials

Frank carries the enterprise IT chapter beside the Essential Eight: its 58 actions under 28 principles, keyed on the action identifiers ASD publishes. It replaces nothing. While ASD maintains both frameworks and regulators have not moved, both sit side by side, and both are free on every plan. The cloud and operational technology chapters will follow once ASD publishes their action identifiers. Frank does not make up identifiers that ASD has not published.

Frank reads ASD’s own assessment workbook as your statement, one rating per action and domain. A rating is read as your claim and earns no credit by itself. The reasoning and evidence beside it are tested the way Frank tests any justification. Not assessed is kept as its own status, never folded into not effective, as the guide requires. A rating in a domain where ASD does not apply that action is reported, not quietly accepted.

ASD’s workbook also lists, under each action, the Essential Eight controls that count towards it. Frank carries that mapping and offers it as a suggestion on your existing work, action by action. Nothing is linked automatically, and a suggestion counts for nothing in your coverage until you accept it. That is the guide’s own rule: a mapping is a starting point, not evidence.

You can see what your own ratings rest on. Start by seeing where you stand. It shows each claim with the evidence beside it, and gives no credit to a claim with nothing behind it.