Notifiable Data Breach scheme: assessing a suspected website or app breach
For owners, managers and technical leads whose website or web application may have exposed personal information.
This guide sets out Part IIIC of the Privacy Act 1988 from its own text: who it covers, what an eligible data breach is, the assessment of a suspected breach under section 26WH, when notification is required and what the statement contains. It then gives you a worksheet for the assessment period, with separate handoffs for a forensic responder and a post-remediation tester. Whether an incident must be notified is a decision for your business and its own adviser.
In this guide
Suspected website breach worksheet for the section 26WH assessment
Seventeen rows for the period between a suspicion that a website or web application has exposed personal information and the review afterwards, quoting the Privacy Act 1988 as Compilation No. 104 (compilation date 4 June 2026) prints it, and the OAIC's guidance as read on 15 November 2026. Each row reads left to right: the point in the assessment, the record to make or question to answer, what the Act or the OAIC says, who answers it, what the record can and cannot show, and the handling limits for that row.
How to use itHide the instructions
Start at row 1, which records the day the rest of the sheet counts from, and make the row 3 copies before any row that reads logs, a database or file storage. Rows 4 and 16 send requests only to systems the business owns or has written authority to test, and carry their own limits; every other row makes a record, reads documents or reads a preserved copy. Rows 15 and 16 are the two handoffs: one to a forensic or incident response provider, one to a tester after the fix. No row decides whether there has been an eligible data breach, whether serious harm is likely or whether an exception applies, and no row tells anyone to contact individuals or the OAIC. Those decisions are the entity's, made with its own legal adviser.
17 rows
Row and point in the assessment
Record to make or question to answer
1. Day 0: when the business became aware
- Record to make or question to answer
- The date and time the first information arrived suggesting that personal information held through the website may have been accessed, disclosed or lost; what it was, such as a customer email, a host alert, a tester's report or a message about an open file; who received it; and who else was told, and when. Beside it, the date 30 days on, marked provisional until the adviser confirms the starting day.
- What the Privacy Act or the OAIC says
- Section 26WH(1): the assessment duty applies if an entity is aware that there are reasonable grounds to suspect that there may have been an eligible data breach of the entity, and is not aware that there are reasonable grounds to believe that the relevant circumstances amount to one. Section 26WH(2)(b): take all reasonable steps to ensure the assessment is completed within 30 days after the entity becomes aware as mentioned in paragraph (1)(a). The OAIC: whether an entity is aware is a factual matter in each case, and an assessment should not be unreasonably delayed, for instance by waiting until the CEO or board is aware.
- Who answers it
- The entity, with its legal adviser, decides when awareness began and how the period is counted.
- What the record can show, and what it cannot
- A dated account of what was known, by whom and when. It cannot show whether those were reasonable grounds to suspect an eligible data breach.
- Handling limits
- A record only. Touches no system.
2. Day 0: who owns the assessment
- Record to make or question to answer
- The person or group responsible for completing the assessment, the legal adviser, the host and developer contacts, any forensic or incident response provider, and the date each was engaged. If the business holds cyber insurance, what the policy asks it to do before engaging anyone, as read with the broker. Note any other entity that holds the same information (row 6) and which of them is assessing.
- What the Privacy Act or the OAIC says
- The OAIC suggests an assessment could be a three-stage process, starting with initiate: decide whether an assessment is necessary and identify which person or group will be responsible for completing it. The Commissioner recommends that entities document the assessment process and outcome. Section 26WJ: if an entity complies with section 26WH in relation to its eligible data breach, and the same access, disclosure or loss is an eligible data breach of other entities, that section does not apply in relation to their eligible data breaches.
- Who answers it
- The entity.
- What the record can show, and what it cannot
- Who held each part of the work, and from when. It cannot show that the assessment was reasonable and expeditious.
- Handling limits
- A record only.
3. Day 0, before anything changes: preserve the evidence
- Record to make or question to answer
- Ask the host and developer for dated copies of what the assessment may need: web server and application logs, the database and file storage, form entries, the hosting control panel's login history and the deployed code, stored away from the live server. Record who made each copy, when, where it is kept and who can open it. Ask how long each log is kept, whether anything rotates or deletes on a schedule, and for that to be paused.
- What the Privacy Act or the OAIC says
- The OAIC's Part 3: during the preliminary containment stage, be careful not to destroy evidence that may be valuable in identifying the cause of the breach.
- Who answers it
- The host, the developer or the forensic responder makes the copies. Nobody on the business side analyses them.
- What the record can show, and what it cannot
- A record of what was kept and where it came from. A copy made after changes cannot show what the system looked like before them.
- Handling limits
- Copy, do not change. Restore no backup over the live site and delete nothing yet. Copies stay off everyday computers, and nobody opens a configuration file or screen that shows a password, key or secret. For a site that is compromised now, the hacked website guide's first-hour checklist comes first.
4. Day 0 onward: containment
- Record to make or question to answer
- Each containment step, when and by whom: an exposed file or route taken offline, a compromised account disabled, a leaked key revoked and replaced, an integration paused. Check each step against row 3 before it is taken. Then record one confirmation that the exposed address no longer returns the information.
- What the Privacy Act or the OAIC says
- The OAIC's Step 1: contain the data breach to prevent any further compromise of personal information. Its Part 3: if it is not practical to shut down the system, or if it would result in loss of evidence, revoke or change computer access privileges or address weaknesses in physical or electronic security. Its Part 4: at any time, including during an assessment, an entity can, and should, take steps to reduce any potential harm to individuals.
- Who answers it
- The host, developer or forensic responder takes the steps. The entity records them.
- What the record can show, and what it cannot
- When the exposure stopped, at the address checked. It cannot show who obtained the information before then.
- Handling limits
- Sends one request, once the row 3 copies exist and the responder knows the time: from a signed-out browser, with no account and nothing submitted, to the exact address on the business's own site that exposed the information. This one check goes to the live address because that is where the exposure was. Record the status code and time only, and copy nothing from the response. If the address is a link that works for whoever holds it, such as a shared or signed file link, describe it in the worksheet rather than pasting it. If personal information still comes back, stop, close the page, note when the request was sent, and tell whoever is running containment. No other requests, no load, and no form or message that could reach a real person.
5. Days 1 to 30: was there unauthorised access, disclosure or loss?
- Record to make or question to answer
- What was exposed and how: the route, file, account or setting; the first and last dates it could have exposed information; and what the preserved logs show inside that window, such as requests to the exposed address from outside the business. Keep exposure (the information could be reached) apart from access (the logs show it was reached).
- What the Privacy Act or the OAIC says
- Section 26WE(2)(a)(i): unauthorised access to, or unauthorised disclosure of, the information. Section 26WE(2)(b)(i): the information is lost in circumstances where unauthorised access to, or unauthorised disclosure of, it is likely to occur. The serious harm condition in (a)(ii) and (b)(ii) is row 10. The OAIC: the Privacy Act does not define these terms, and its guide draws on their ordinary meaning; unauthorised access includes access by an employee, an independent contractor or an external third party, such as by hacking.
- Who answers it
- The forensic responder establishes what happened from the preserved copies. The entity and its adviser apply the Act's terms to those facts.
- What the record can show, and what it cannot
- A dated window and what the logs record within it. It cannot show access the logs never captured, or what happened to a copy once it left the system.
- Handling limits
- Reads the preserved copy, not the live system, and only the routes, dates and records the question needs. Record counts and request times in the worksheet, never names or other personal details.
6. Days 1 to 30: does the entity hold the information, and who else does?
- Record to make or question to answer
- Each system the information sat in: the website database, file storage, the form tool, the customer database, the mailing platform, the host's backups and any service outside Australia. For each, who has possession, who has control under the contract, and the vendor's own incident contact.
- What the Privacy Act or the OAIC says
- Section 6(1): an entity holds personal information if it has possession or control of a record that contains it. Section 26WC(1): where information was disclosed to an overseas recipient, APP 8.1 applied and the recipient holds it, Part IIIC has effect as if the APP entity held it and were required under section 15 not to breach APP 11.1 in relation to it. The OAIC: where records are stored with a cloud service provider that has possession and can access them, and the entity keeps control under its contract, both hold the information; whether an entity holds information depends on the particular circumstances of the arrangement.
- Who answers it
- The entity and its adviser, with each vendor.
- What the record can show, and what it cannot
- A map of who holds what and who has agreed to take which step. It cannot settle which entity notifies: the OAIC says the scheme leaves that to the entities.
- Handling limits
- Reads contracts, vendor documentation and account settings pages that display no password, key, token or shareable link. No vendor platform is tested; where a question about one comes up, check what that vendor's terms allow and raise it with the vendor in writing if they do not say.
7. Days 1 to 30: kinds of information involved
- Record to make or question to answer
- For each system in row 6, the fields exposed and the number of individuals' records, from the schema, the form definition or a count. Mark separately: fields in a category of sensitive information as section 6(1) defines it, identity document details, financial information, tax file numbers, passwords or other sign-in details, and fields that together say more about a person than any one does.
- What the Privacy Act or the OAIC says
- Section 26WG(c) and (d): the kind or kinds of information, and its sensitivity. Section 26WK(3)(c): a statement must set out the particular kind or kinds of information concerned. The OAIC: kinds that may increase the risk of serious harm include sensitive information such as health information, documents commonly used for identity fraud (including Medicare card, driver licence and passport details), financial information and a combination of types; a statement should name a driver's licence or passport rather than using generic terms like ID documents.
- Who answers it
- The developer lists fields and counts. The entity marks the categories, with its adviser where the definition is unclear.
- What the record can show, and what it cannot
- The kinds and number of records involved. It cannot show whether serious harm is likely.
- Handling limits
- Field names and counts only, from the preserved copy. Copy no values, open no record beyond what a count needs, and never copy a password hash, key or token.
8. Days 1 to 30: security measures, and whether they could be overcome
- Record to make or question to answer
- Whether the exposed information was protected, for example passwords stored as hashes, fields or files encrypted, or downloads behind signed links, and whether whatever unlocks each measure, such as an encryption key or signing secret, was stored or exposed in the same place.
- What the Privacy Act or the OAIC says
- Section 26WG(e), (f) and (h): whether the information is protected by one or more security measures, the likelihood that any of them could be overcome, and, where a security technology or methodology designed to make it unintelligible or meaningless to unauthorised persons was used, the likelihood that persons who have obtained or could obtain the information, and have or are likely to have the intention of causing harm to any of the individuals, have obtained or could obtain the information or knowledge required to circumvent it. The note to section 26WG gives an encryption key as an example of that information. The OAIC: an entity should consider whether its method of encryption is an industry-recognised secure standard at the time it is assessing the likelihood of risk, and, for example, if an attacker holds both encrypted data and the encryption key, the entity should not assume the data is secure.
- Who answers it
- The developer or forensic responder states the facts. The entity weighs them.
- What the record can show, and what it cannot
- What protected the information, as the people who built the system describe it. It cannot show that a measure held.
- Handling limits
- Describe where each key or secret is kept. Never open, copy or paste one into the worksheet.
9. Days 1 to 30: who has obtained, or could obtain, the information
- Record to make or question to answer
- Whether the exposure was open to anyone on the internet, to signed-in customers, to staff or to one mistaken recipient; whether search engines or caches may hold copies; any contact from a person claiming to hold the data; and how long the information was accessible (row 5).
- What the Privacy Act or the OAIC says
- Section 26WG(g): the persons, or the kinds of persons, who have obtained, or who could obtain, the information. The OAIC: information publicly accessible for a significant period before the entity is aware may be more likely to have been accessed in ways that will result in serious harm, and a third party that appears to target the personal information of a particular individual or group may increase the risk of serious harm.
- Who answers it
- The forensic responder, from the preserved copies. The entity and its adviser weigh it.
- What the record can show, and what it cannot
- Who could reach the information, and for how long. It cannot show what a person who obtained it will do.
- Handling limits
- Reads the preserved copies only, for the window in row 5. Checking search engines or caches for copies, and searching for or downloading copies from places an attacker may have posted them, is work for the forensic responder. Record any contact from a person claiming to hold the data and pass it to the adviser; nobody replies to it without advice.
10. By day 30: is serious harm likely?
- Record to make or question to answer
- The entity's reasoning against each section 26WG matter, rows 7 to 9 included, for each group of individuals; the kinds of harm considered; who formed the view, and on what date.
- What the Privacy Act or the OAIC says
- Section 26WE(2): a reasonable person would conclude that the access or disclosure would be likely to result in serious harm to any of the individuals to whom the information relates. The OAIC: the reasonable person is a person in the entity's position who is properly informed; likely means more probable than not, rather than possible; serious harm may include serious physical, psychological, emotional, financial, or reputational harm; entities should assess the risk of serious harm holistically.
- Who answers it
- The entity, with its legal adviser. This worksheet does not answer it.
- What the record can show, and what it cannot
- The reasoning the entity relied on, and when. The other rows feed it; none of them decides it.
- Handling limits
- A record only.
11. Any time: remedial action and its result
- Record to make or question to answer
- Each action taken to prevent serious harm, when, and its result, such as a mistaken recipient's written confirmation of deletion, a key revoked before any use was seen, or an exposed copy removed from a cache. Note which individuals each action protects.
- What the Privacy Act or the OAIC says
- Section 26WF(1) and (4): where the entity takes action before the access or disclosure results in serious harm to any of the individuals and, as a result, a reasonable person would conclude that it would not be likely to result in serious harm to any of them, it is not, and is taken never to have been, an eligible data breach. Section 26WF(3): for a loss, the same where action taken before any unauthorised access or disclosure means there is none. Subsections 26WF(2) and (5): where the action has that result for a particular individual, the Part does not require the entity or any other entity to take steps to notify that individual of the contents of a statement.
- Who answers it
- The entity and its adviser decide whether the action had that effect.
- What the record can show, and what it cannot
- What was done and what it achieved, on the evidence available. A retest can show a route is now closed; it cannot show what happened to information already taken.
- Handling limits
- A record only. Clearing a cache or deleting a copy removes evidence too, so row 3 comes first.
12. Any time: does an exception apply?
- Record to make or question to answer
- The facts an adviser needs for each exception: whether the breach was, or must be, notified under section 75 of the My Health Records Act 2012; whether another entity holding the information has assessed or notified; whether the entity is an enforcement body; whether a Commonwealth secrecy provision covers the information; whether the Commissioner has been asked for, or has made, a declaration; whether the Commissioner has given a direction under section 26WR.
- What the Privacy Act or the OAIC says
- Section 26WD (My Health Records Act 2012), sections 26WJ and 26WM (other entities), section 26WN (enforcement related activities), section 26WP (secrecy provisions) and section 26WQ (declaration by the Commissioner); for a direction under section 26WR, sections 26WS (enforcement related activities) and 26WT (secrecy provisions). The OAIC says the Commissioner expects declarations to be limited to exceptional cases.
- Who answers it
- The entity's legal adviser.
- What the record can show, and what it cannot
- The facts each exception turns on. It cannot show that one applies.
- Handling limits
- A record only.
13. By day 30: the outcome of the assessment
- Record to make or question to answer
- Whether the entity is aware that there are reasonable grounds to believe there has been an eligible data breach, the date it reached that view, who reached it and why. If the assessment could not reasonably be completed within 30 days, the steps taken and the reasons.
- What the Privacy Act or the OAIC says
- Section 26WH(2)(a): a reasonable and expeditious assessment of whether there are reasonable grounds to believe that the relevant circumstances amount to an eligible data breach. Section 26WK(1): the statement duty applies if the entity is aware of such grounds. The Commissioner recommends that entities document the assessment process and outcome and, where an assessment cannot reasonably be completed within 30 days, document this so that they can demonstrate that all reasonable steps were taken to complete it within 30 days, the reasons for the delay and that the assessment was reasonable and expeditious.
- Who answers it
- The entity, with its legal adviser.
- What the record can show, and what it cannot
- A dated record of the decision and its reasons. It cannot show that the decision was correct.
- Handling limits
- A record only.
14. If the entity decides to notify: the statement and who receives it
- Record to make or question to answer
- The facts the statement draws on, from rows 5, 7, 9 and 11: the dates, the kinds of information, who obtained or could obtain it, and what was done. Which option under section 26WL(2) the entity chose, and why. The date the statement was given to the Commissioner, and the dates individuals were notified or a notice was published.
- What the Privacy Act or the OAIC says
- Section 26WK(2) and (3): a statement setting out the entity's identity and contact details, a description of the eligible data breach, the particular kind or kinds of information concerned and recommended steps for individuals, given to the Commissioner as soon as practicable. Section 26WL(2) and (3): take such steps as are reasonable in the circumstances to notify the contents to each individual to whom the information relates, or to each individual at risk, where that is practicable, or otherwise publish the statement on the entity's website (if any) and take reasonable steps to publicise it, as soon as practicable after preparing the statement. The OAIC: an entity should use its online Notifiable Data Breach form, and should attach a copy of its template notification to affected individuals when completing it.
- Who answers it
- The entity and its legal adviser decide and draft. Nothing in this worksheet contacts anyone.
- What the record can show, and what it cannot
- A record of what was sent, to whom and when. It cannot show that the statement is adequate.
- Handling limits
- A record only, with no individual's details copied into it. Where the statement is published under paragraph 26WL(2)(c), the OAIC says entities should take care that the online notice does not contain any personal information or make an affected individual reasonably identifiable.
15. Handoff 1: to a forensic or incident response provider
- Record to make or question to answer
- What goes to the responder: rows 1 to 9, the row 3 copies and their custody record, the row 6 system list, and the questions the assessment needs answered: how the information was reached, when it started and stopped, what was taken, and whether anyone still has access. What comes back: the responder's written findings, and what they could not establish.
- What the Privacy Act or the OAIC says
- The OAIC's Part 2: in some cases an entity may need to include external experts in its response team, for example legal advice, data forensics, or media management. Its Part 4: the Commissioner will not be able to provide detailed advice about the application of the scheme to specific data breaches, and entities should seek their own legal and technical advice.
- Who answers it
- An incident response or forensic provider the entity chooses, or one its insurer's process names where the policy sets one.
- What the record can show, and what it cannot
- An account of what happened, as far as the evidence reaches. A forensic report is an input to the assessment, not the assessment.
- Handling limits
- Handled under the responder's own terms. The copies pass through the row 3 custody record; passwords, keys and links that work for whoever holds them are not passed in the worksheet, and the responder arranges its own access with the host. Clever Ops does not provide this work.
16. Handoff 2: to a post-remediation tester
- Record to make or question to answer
- Once containment and the responder's work allow it: the weakness to retest, the build it was fixed in, the responder's findings, test accounts for each role, and written authorisation. What comes back: whether the weakness reproduces on the build tested, whether similar routes share it, and each finding's retest status.
- What the Privacy Act or the OAIC says
- The OAIC's Part 3 lists, among what a review after a breach might involve, a security review including a root cause analysis of the data breach, and a prevention plan to prevent similar incidents in future. The text of Part IIIC does not mention security testing.
- Who answers it
- A tester the entity authorises in writing, with the start agreed with the responder.
- What the record can show, and what it cannot
- How the fixed weakness and similar routes behaved on the build and accounts tested, and each finding marked fixed, partially fixed, still reproducible or not retested. It cannot show who reached information before the fix, whether serious harm is likely, or that no other weakness exists.
- Handling limits
- Sends requests, after the responder agrees the start. Targets are limited to systems the entity owns or has written authority to have tested, using a staging copy if there is one, test accounts and made-up records, with no real customer's account or record touched. Each route gets the small, fixed number of requests the statement of work sets, test-account sign-ins counted among them, and nothing that adds load. Any email or text a test sets off goes to an inbox or number the business controls. Passwords, keys, session tokens and links that work for whoever holds them, such as reset or shared file links, are used only for the made-up records and never copied into the evidence. A vendor's platform is touched only as far as that vendor's terms allow: read the terms, and ask the vendor in writing where they say nothing. If a response shows a real person's information, the tester stops, records the request and what came back, opens nothing further and tells the person running the assessment.
17. After the assessment: review
- Record to make or question to answer
- What would have made the assessment faster or surer: how long logs were kept, whether a field list existed, which vendor contacts were missing, how long it took to find the right people. Who owns each change, and when it is due.
- What the Privacy Act or the OAIC says
- The OAIC's Step 4: review the incident and consider what actions can be taken to prevent future breaches. Its Part 2: all entities should have a data breach response plan, and the plan should be in writing.
- Who answers it
- The entity.
- What the record can show, and what it cannot
- A list of changes and owners. It cannot show the changes were made; a later review or retest can.
- Handling limits
- A record only. Keep the worksheet and its copies for as long as the adviser or insurer advises.
What the Notifiable Data Breaches scheme is, and where its text sits
The Notifiable Data Breaches scheme, which the OAIC also writes as the Notifiable Data Breach (NDB) scheme, is Part IIIC of the Privacy Act 1988 (Cth).
Section 26WA, headed Guide to this Part, says the Part sets up a scheme for notification of eligible data breaches; that an eligible data breach happens if there is unauthorised access to, unauthorised disclosure of, or loss of, personal information held by an entity, and the access, disclosure or loss is likely to result in serious harm to any of the individuals to whom the information relates; and that an entity must give a notification if it has reasonable grounds to believe that an eligible data breach has happened, or if it is directed to do so by the Commissioner.
That guide is an outline: the tests themselves, with their conditions, are in the sections that follow it, and those are what this page quotes.
Quotations from the Act on this page come from Compilation No. 104, as the Federal Register of Legislation publishes it. The compilation describes itself as the text of the law as amended and in force on 4 June 2026, which it calls the compilation date.
When the Register's text page for the Act was read on 15 November 2026, it listed the Act as in force and Compilation No. 104 as its latest version. The Office of the Australian Information Commissioner (OAIC) says the requirements of the scheme apply to breaches that occur on or after 22 February 2018.
Its page on when to report adds that some data breaches occur over a period of time, and suggests that where a system may have been compromised before that date but data may have been accessed after it, the organisation or agency should assume the breach is subject to the scheme, while the circumstances will need to be assessed.
The OAIC's page on the scheme points to its Data breach preparation and response guide for more information on complying with it. The guide's index page says it was updated in June 2024, and its Part 4, on the scheme, shows an update date of February 2025.
The index page also says the OAIC is updating the guide to reflect changes made by the Privacy and Other Legislation Amendment Act 2024, including a new Division 5 of Part IIIC, which commenced on 11 December 2024 and, in the OAIC's description, sets up a framework where the Minister may make a declaration that permits the collection, use and disclosure of personal information that would otherwise not be permitted under the APPs, to prevent or reduce the risk of harm to individuals in the event of an eligible data breach.
The OAIC's quick reference guide for responding to data breaches shows a publication date of 29 June 2026.
Where this page quotes OAIC guidance, it keeps the guidance's should and may, and credits it to the OAIC rather than to the Act.
This guide stays with the part of the scheme a website or web application touches. Read it as background for a conversation with your lawyer: it gives no legal advice, audits nothing and does not respond to an incident.
It does not decide whether an incident is an eligible data breach or whether serious harm is likely. That decision belongs to the entity, made with its own legal adviser, and the OAIC says the Commissioner will not be able to provide detailed advice about the application of the scheme to specific data breaches, and that entities should seek their own legal and technical advice.
Who the scheme covers, and what it means to hold information
Section 26WE(1) sets the scope.
The section applies where an APP entity holds personal information relating to one or more individuals and is required under section 15 not to do an act, or engage in a practice, that breaches Australian Privacy Principle 11.1 in relation to it; where a credit reporting body holds credit reporting information and is required to comply with section 20Q in relation to it; where a credit provider holds credit eligibility information and is required to comply with section 21S(1) in relation to it; and where a file number recipient holds tax file number information and is required under section 18 not to do an act, or engage in a practice, that breaches a section 17 rule that relates to it. Section 26WB says that, for Part IIIC, entity includes a person who is a file number recipient.
Part 4 of the OAIC's guide puts the same scope in plainer terms. It says entities that have existing obligations under the Privacy Act to secure personal information must comply with the scheme, and that this includes Australian Government agencies, businesses and not-for-profit organisations that have an annual turnover of more than $3 million, private sector health service providers, credit reporting bodies, credit providers, entities that trade in personal information and tax file number (TFN) recipients.
It says small business operators generally do not have obligations under the APPs unless an exception applies, and that in certain circumstances one must comply with the APPs, and therefore with the scheme, including where it holds health information and provides a health service, trades in personal information, or has opted in to APP coverage under section 6EA.
It also says entities that have Privacy Act security obligations in relation to particular types of information only, for example small businesses that are required to secure tax file number information, do not need to notify about data breaches that affect other types of information outside the scope of their obligations.
For a website, the word to read closely is holds. Section 6(1) says an entity holds personal information if it has possession or control of a record that contains the personal information. The OAIC's guide gives the example of an entity that stores its records with a cloud service provider: the provider holds the information because it has possession of the records and can access the information physically or by use of an electronic device, and the first entity holds it too because it has contractual rights to retain control of the records.
The guide adds that whether an entity holds personal information depends on the particular circumstances of the arrangement.
Under section 26WC(1), where an APP entity has disclosed personal information to an overseas recipient, Australian Privacy Principle 8.1 applied to the disclosure and the recipient holds the information, Part IIIC has effect as if the APP entity held it and were required under section 15 not to do an act, or engage in a practice, that breaches Australian Privacy Principle 11.1 in relation to it.
When one access, disclosure or loss is an eligible data breach of more than one entity, section 26WJ says that if one entity complies with the assessment duty in section 26WH in relation to its eligible data breach, that section does not apply in relation to the eligible data breaches of the other entities, and section 26WM does the same for the statement and notification duties in sections 26WK and 26WL.
The OAIC says the scheme leaves it up to the entities to decide which of them should take those steps, and suggests that, in general, the entity with the most direct relationship with the individuals affected by the data breach should carry out notification.
Whether your business is covered, and for which information, is the first question to settle with your adviser; the Australian Privacy Principles guide sets out the small business exemption from the Act's own text.
What makes a data breach an eligible data breach
Section 26WE(2) sets out the test in paragraphs (a) and (b).
Under paragraph (a), there is unauthorised access to, or unauthorised disclosure of, the information, and a reasonable person would conclude that the access or disclosure would be likely to result in serious harm to any of the individuals to whom the information relates.
Under paragraph (b), the information is lost in circumstances where unauthorised access to, or unauthorised disclosure of, it is likely to occur and, assuming that access or disclosure were to occur, a reasonable person would conclude that it would be likely to result in serious harm to any of those individuals.
Where either paragraph is met, the access, disclosure or loss is an eligible data breach of the APP entity, credit reporting body, credit provider or file number recipient, as the case may be, and an individual covered by the serious harm condition in subparagraph (a)(ii) or (b)(ii) is at risk from the eligible data breach. Section 26WE(3) makes all of this subject to section 26WF.
The Act does not define unauthorised access, unauthorised disclosure or loss, and the OAIC's guide draws on their ordinary meaning. It says unauthorised access includes access by an employee of the entity, or an independent contractor, as well as by an external third party, such as by hacking.
It says unauthorised disclosure occurs when an entity, whether intentionally or unintentionally, makes personal information accessible or visible to others outside the entity and releases it from its effective control in a way that is not permitted by the Privacy Act, and gives as an example an employee accidentally publishing a confidential data file containing personal information on the internet.
Serious harm is not defined in the Privacy Act either. The OAIC says that in the context of a data breach, serious harm to an individual may include serious physical, psychological, emotional, financial, or reputational harm; that likely to occur means the risk of serious harm to an individual is more probable than not, rather than possible; and that the reasonable person is a person in the entity's position, rather than the position of an individual whose personal information was part of the breach or any other person, who is properly informed, based on information immediately available or following reasonable inquiries or an assessment of the data breach.
Section 26WG lists what to have regard to in determining whether a reasonable person would conclude that an access to, or a disclosure of, information would, or would not, be likely to result in serious harm to any of the individuals to whom the information relates: the kind or kinds of information; its sensitivity; whether it is protected by one or more security measures and, if so, the likelihood that any of them could be overcome; the persons, or the kinds of persons, who have obtained, or who could obtain, the information; where a security technology or methodology designed to make the information unintelligible or meaningless to persons not authorised to obtain it was used, the likelihood that the persons, or the kinds of persons, who have obtained or could obtain the information, and who have or are likely to have the intention of causing harm to any of the individuals to whom it relates, have obtained or could obtain the information or knowledge required to circumvent it; the nature of the harm; and any other relevant matters.
The note to the section says that if the technology is encryption, an encryption key is an example of information required to circumvent it.
Not every data breach is an eligible one. Under section 26WF(1), where an access or disclosure is covered by paragraph 26WE(2)(a), the entity takes action in relation to it before it results in serious harm to any of the individuals to whom the information relates, and as a result of the action a reasonable person would conclude that it would not be likely to result in serious harm to any of those individuals, the access or disclosure is not, and is taken never to have been, an eligible data breach of that entity or of any other entity.
For a loss covered by paragraph 26WE(2)(b), section 26WF(3) gives the same result where the action is taken before there is unauthorised access to, or unauthorised disclosure of, the information and, as a result, there is none; section 26WF(4) gives it where the action is taken after the access or disclosure but before it results in serious harm to any of the individuals and, as a result, a reasonable person would conclude that it would not be likely to result in serious harm to any of them.
Subsections 26WF(2) and (5) apply the test to one particular individual: where the action is taken before the access or disclosure results in serious harm to that individual (for a loss, after the access or disclosure has happened) and, as a result, a reasonable person would conclude that it would not be likely to result in serious harm to the individual, the Part does not require the entity, or any other entity, to take steps to notify that individual of the contents of a statement that relates to it.
Among the OAIC's key points is that not all data breaches are eligible, and it gives as an example that if an entity acts quickly to remediate a data breach, and as a result the breach is not likely to result in serious harm, there is no requirement to notify any individuals or the Commissioner.
Assessing a suspected breach: section 26WH and the 30 days
Section 26WH applies if an entity is aware that there are reasonable grounds to suspect that there may have been an eligible data breach of the entity, and is not aware that there are reasonable grounds to believe that the relevant circumstances amount to one.
The entity must then carry out a reasonable and expeditious assessment of whether there are reasonable grounds to believe that the relevant circumstances amount to an eligible data breach, and take all reasonable steps to ensure that the assessment is completed within 30 days after it becomes aware as mentioned in paragraph 26WH(1)(a).
The note to the section points to section 26WK, which applies if an entity is aware that there are reasonable grounds to believe that there has been an eligible data breach of the entity.
The OAIC describes the period as 30 calendar days after the day the entity became aware of the grounds, or information, that caused it to suspect an eligible data breach. It says the Commissioner expects that wherever possible entities treat 30 days as a maximum time limit and endeavour to complete the assessment in a much shorter timeframe, as the risk of serious harm to individuals often increases with time.
Where an entity cannot reasonably complete an assessment within 30 days, the Commissioner recommends that it document this, so that it can demonstrate that all reasonable steps have been taken to complete the assessment within 30 days, the reasons for the delay, and that the assessment was reasonable and expeditious.
When the period starts is itself a question of fact. The OAIC says whether an entity is aware of a suspected breach is a factual matter in each case, having regard to how a reasonable person who is properly informed would be expected to act in the circumstances, and that an entity should not unreasonably delay an assessment, for instance by waiting until its CEO or board is aware of information that would otherwise trigger reasonable suspicion of a breach within the entity.
For a website, the first information can arrive by any route: a customer's email, a host's alert, a tester's report or a stranger's message about an open file.
Row 1 of the worksheet records when it arrived and who received it. Whether that was the moment of awareness is for the entity and its adviser.
The Act does not set out how to assess, and the OAIC says entities may develop their own procedures. It suggests an assessment could be a three-stage process: initiate, deciding whether an assessment is necessary and identifying which person or group will be responsible for completing it; investigate, quickly gathering relevant information about the suspected breach, including, for example, what personal information is affected, who may have had access to it and the likely impacts; and evaluate, making a decision, based on the investigation, about whether the identified breach is an eligible data breach.
The Commissioner, it says, expects the time and effort an entity spends to be proportionate to the likelihood of the breach and its apparent severity, and recommends that entities document the assessment process and outcome.
It also says that at any time, including during an assessment, an entity can, and should, take steps to reduce any potential harm to individuals.
When notification is required, and what the statement must contain
Section 26WK applies if an entity is aware that there are reasonable grounds to believe that there has been an eligible data breach of the entity.
The entity must prepare a statement that complies with subsection 26WK(3) and give a copy of it to the Commissioner, and do so as soon as practicable after it becomes so aware. Subsection (3) says the statement must set out the identity and contact details of the entity; a description of the eligible data breach that the entity has reasonable grounds to believe has happened; the particular kind or kinds of information concerned; and recommendations about the steps that individuals should take in response to that breach.
Under subsection (4), where the entity has reasonable grounds to believe the same access, disclosure or loss is an eligible data breach of other entities, the statement may also set out their identity and contact details.
Section 26WL applies if the entity is aware of those grounds and has prepared a statement that complies with subsection 26WK(3) and relates to the eligible data breach it has reasonable grounds to believe has happened. Section 26WL(2) then says the entity must: under paragraph (a), if it is practicable to notify the contents of the statement to each of the individuals to whom the relevant information relates, take such steps as are reasonable in the circumstances to do that; or, under paragraph (b), if it is practicable to notify the contents to each of the individuals who are at risk from the eligible data breach, take such steps as are reasonable in the circumstances to notify each of them; or, under paragraph (c), if neither paragraph (a) nor (b) applies, publish a copy of the statement on the entity's website (if any) and take reasonable steps to publicise its contents.
Subsection 26WL(3) says the entity must comply with subsection (2) as soon as practicable after the completion of the preparation of the statement.
Subsection (4) says that if the entity normally communicates with a particular individual using a particular method, the notification to that individual under paragraph (2)(a) or (b) may use that method, and that this does not limit those paragraphs.
The OAIC says entities should use its online Notifiable Data Breach form, attach a copy of the template notification to affected individuals when completing it, and that the more an entity tells it about the circumstances of the breach, what it has done to contain it and any remedial action, the more that will help the OAIC respond.
It expects the statement to include sufficient information about the breach to allow affected individuals to properly assess the possible consequences for them and to take protective action, and says a description may include the date, or date range, of the unauthorised access or disclosure; the date the entity detected the breach; the circumstances, such as any known causes; who has obtained or is likely to have obtained access; and the steps the entity has taken to contain or remediate it.
It says the statement should set out whether an individual's driver's licence or passport has been compromised, rather than using more generic terms like ID documents, and its quick reference guide says notifications to affected individuals should be written in plain English.
Parts of the OAIC's guidance on notifying concern a website directly. Its illustration of notifying only the individuals at risk, an example in the guide rather than a reported case, is a retailer whose website carries malicious software that intercepts payment card details: after a comprehensive risk assessment, the retailer considers that customers who made purchases while the software was active are at likely risk of serious harm and that customers who only had basic account details accessed are not, and the guide says the retailer is only required to notify those individuals it considers to be at likely risk of serious harm.
For publication under paragraph 26WL(2)(c), which the guide calls option 3, the OAIC says it is not enough to simply upload a copy of the statement prepared for the Commissioner on any webpage of the entity's website, and that entities must also take proactive steps to publicise the substance of the eligible data breach.
It says the Privacy Act does not specify how long the statement must stay accessible, that the Commissioner would generally expect it to be available for at least 6 months, and that, if option 3 is chosen, entities should take care that the online notice does not contain any personal information, and that any description of the affected group should not identify an affected individual or make one reasonably identifiable.
It also says the scheme does not require the Commissioner to be notified before individuals, so an entity may begin notifying individuals before, or at the same time as, notifying the Commissioner.
Separately, section 26WR lets the Commissioner, where aware that there are reasonable grounds to believe there has been an eligible data breach of an entity, direct the entity by written notice to prepare a statement and give the Commissioner a copy. Under section 26WR(2) the direction must also require the entity to notify the individuals, or publish and publicise the statement, on the same practicability conditions as section 26WL(2), and under section 26WR(3) the Commissioner must first invite the entity to make a submission within the period the invitation specifies.
Section 26WR(10) says the entity must comply with a direction as soon as practicable after it is given.
Section 26WS says an entity is not required to comply with a direction if it is an enforcement body and its chief executive officer believes on reasonable grounds that compliance would be likely to prejudice one or more enforcement related activities conducted by, or on behalf of, the body.
Section 26WT says that where complying with paragraph 26WR(1)(b) or subsection 26WR(2) in relation to a statement would, to any extent, be inconsistent with a secrecy provision, that provision of section 26WR does not apply to the entity in relation to the statement: to the extent of the inconsistency, or, for a prescribed secrecy provision specified in the regulations, without that limit.
The exceptions, and who works out whether one applies
Part IIIC carries its own exceptions, each with conditions.
They are listed here so the worksheet can record the facts each one turns on, not so a reader can apply them alone.
- Section 26WD: if an unauthorised access, disclosure or loss has been, or is required to be, notified under section 75 of the My Health Records Act 2012, Part IIIC does not apply in relation to it. The OAIC says only notifications under section 75 fall within this exception.
- Sections 26WJ and 26WM: where the access, disclosure or loss is an eligible data breach of more than one entity and one of them complies with section 26WH, or with sections 26WK and 26WL, those sections do not apply to the eligible data breaches of the others.
- Section 26WN: where the entity is an enforcement body and its chief executive officer believes on reasonable grounds both that there has been an eligible data breach of the entity and that complying with section 26WL would be likely to prejudice one or more enforcement related activities conducted by, or on behalf of, the body, paragraph 26WK(3)(d) and section 26WL do not apply. The OAIC says the enforcement body must still give the Commissioner a statement, which does not have to include the steps recommended for individuals.
- Section 26WP: a secrecy provision is a provision of another Commonwealth law that prohibits or regulates the use or disclosure of information. Where giving the Commissioner a copy of the statement, or complying with section 26WL, in relation to a statement would to any extent be inconsistent with a secrecy provision other than a prescribed secrecy provision, subsection 26WK(2) or section 26WL, as the case may be, does not apply to the entity, in relation to the statement, to the extent of the inconsistency. Where the inconsistency is with a prescribed secrecy provision, one specified in the regulations, subsection 26WK(2) or section 26WL does not apply to the entity in relation to the statement.
- Section 26WQ: where the Commissioner is aware that there are reasonable grounds to believe that there has been an eligible data breach of an entity, or is informed by the entity that it is aware of such grounds, the Commissioner may, by written notice to the entity, declare that sections 26WK and 26WL do not apply in relation to the eligible data breach, or declare that subsection 26WL(3) has effect as if it required compliance with subsection 26WL(2) before the end of a period the declaration specifies. Section 26WQ(2) says that second power may only be used to extend the time to the end of a period the Commissioner is satisfied is reasonable in the circumstances. Section 26WQ(3) says the Commissioner must not make a declaration unless satisfied it is reasonable in the circumstances, having regard to the public interest, any relevant advice from an enforcement body or the Australian Signals Directorate, and such other matters as the Commissioner considers relevant. Under section 26WQ(9), if an entity applies for a declaration that, to any extent, relates to an eligible data breach of the entity, sections 26WK and 26WL do not apply in relation to that breach until the Commissioner makes a decision in response to the application. The OAIC says the Commissioner expects declarations under section 26WQ to be limited to exceptional cases.
What a website or web application adds to the assessment
For a website or web application, the facts the section 26WG matters ask about can sit in the systems it runs on or sends data to: the application's database and file storage, form entries and notification emails, the services a form passes submissions to, server and application logs, and backups.
That makes the site's developer and host part of the assessment, and it makes their records evidence. The worksheet turns each matter into a record someone can make.
The database schema and the form definitions show which fields an exposed table or form held, and a count shows how many people's records it held. The OAIC lists kinds of information that may increase the risk of serious harm if there is a data breach, including sensitive information such as information about an individual's health; documents commonly used for identity fraud, including Medicare card, driver licence and passport details; financial information; and a combination of types of personal information, rather than a single piece, that allows more to be known about the individual or is more likely to cause the individual serious harm if compromised. Row 7 lists the fields and counts without copying their values.
Whether a password hash, encryption at rest or a signed download link counts as a security measure that could be overcome is part of the entity's assessment, and the facts come from the people who built the system. The OAIC says an entity should consider whether its method of encryption is an industry-recognised secure standard at the time it is assessing the likelihood of risk, and that if an attacker holds both the encrypted data and the encryption key needed to decrypt it, the entity should not assume the data is secure.
Row 8 asks the developer where each key or secret is stored and whether that place was exposed as well, described rather than copied.
An exposed file at a public address, a route that answered any signed-in customer and a database reached through a stolen administrator login are different facts for the section 26WG question of who has obtained, or could obtain, the information. The OAIC says that if personal information is publicly accessible for a significant period before the entity is aware of the breach, it may be more likely to have been accessed in ways that will result in serious harm, and that a third party that appears to target the personal information of a particular individual or group may increase the risk of serious harm.
Row 5 records the first and last dates of possible exposure and what the preserved logs show inside that window, and row 9 records who could reach the information; the worksheet gives both to a forensic responder.
The OAIC says that, in general, a data breach response should follow four key steps: contain, assess, notify and review. Its Part 3 gives, as examples of containment, stopping the unauthorised practice, recovering the records or shutting down the system that was breached, and says that if it is not practical to shut down the system, or if it would result in loss of evidence, the step is to revoke or change computer access privileges or address weaknesses in physical or electronic security.
It also says to be careful during this preliminary stage not to destroy evidence that may be valuable in identifying the cause of the breach.
On a website, deleting an exposed file, redeploying over a changed server or restoring a backup can each remove what a responder needs, so row 3 asks for copies first.
For a site that is compromised now, the hacked website guide covers the first hour, and its steps are not repeated here.
Assessment, notification and testing are separate jobs
A suspected breach of a website draws in different kinds of work that are easy to blur.
The assessment, and any notification, sit with the entity, on which sections 26WH, 26WK and 26WL place those duties, working with its adviser. Working out what happened belongs to a forensic or incident response provider: how the information was reached, when, what left the system, and whether anyone still has access.
Testing belongs after the fix, to check whether the weakness still reproduces and to look for others like it.
The OAIC's guide to preparing a response plan says that in some cases an entity may need to include external experts in its response team, for example legal advice, data forensics, or media management.
A security test supplies a narrow, dated kind of evidence. On a staging copy or the fixed build, with test accounts and invented records, it can show whether the route, file or access rule that exposed the information still does so, whether similar routes share the same weakness, and, after a consolidated retest, which findings are fixed, partially fixed, still reproducible or not retested. A clean result is a negative result for what was tried, on the build and accounts tested.
What a test cannot do matters more during an assessment. It cannot say who reached information in the past, how many records left the system or whether someone still holds a copy, because it looks at the system as it stands rather than at what happened.
It cannot decide whether a reasonable person would conclude that serious harm is likely, whether remedial action prevented it or whether an exception applies. The OAIC's Part 3 lists a security review including a root cause analysis of the data breach among the things a review after a breach might involve.
A test can contribute to that review, but it is not the forensic record, and it does not stand in for one.
Timing needs agreement. Requests sent to a live system while a responder is still reading its logs can add entries to the same logs, and a fix deployed before copies are made can overwrite evidence. Agree with the responder when testing can start, and test a staging copy where one exists.
Keeping the worksheet with a data breach response plan
The OAIC says all entities should have a data breach response plan, describes one as a framework that sets out the roles and responsibilities involved in managing a data breach and the steps the entity will take if one occurs, and says the plan should be in writing.
Its Part 4 says the Commissioner expects an entity's approach to data breach management, including its response plan, to incorporate the scheme's requirements for assessing suspected eligible data breaches.
The quick checklist in its Part 2 includes an approach for conducting assessments, processes for responding to incidents that involve another entity, and a record-keeping policy to ensure that breaches are documented.
For a business whose personal information passes through a website, the worksheet can hold the website part of that plan before anything happens: who the host and developer are, where logs are kept and for how long, which vendors hold copies of what, and who the legal adviser, forensic responder and tester would be.
The OAIC suggests an entity can test its plan by, for example, responding to a hypothetical data breach, and an invented exposure on a staging copy of the site, worked through the worksheet with no real records involved, is one way to do that.
How to use the suspected breach worksheet
The worksheet follows the assessment period from the day of awareness to the review afterwards.
Each row is a record to make or a question to answer, with what the Privacy Act or the OAIC says beside it, who answers it, what the record can and cannot show, and the limits on handling it. Start it as soon as there is a suspicion, and add to it rather than rewriting it, so the dates stay honest.
- Fill in row 1 first. It records the date the rest of the worksheet counts from, and the entity's adviser decides whether that date is the right one.
- Preserve before you look. Row 3 comes before any row that reads logs, a database or file storage, and those rows read the preserved copy, only for the routes, dates and records the question needs.
- Passwords, keys, tokens and secrets stay out of the worksheet, and so do links that work for whoever holds them, such as reset, booking, payment or shared file links. Apart from row 16, where the tester uses passwords, keys, session tokens and links that work for whoever holds them only for the made-up records and never copies them into the evidence, nobody opens a screen or file that displays one; the worksheet notes where each is kept, not its value.
- Rows 1 to 3 and 5 to 15, and row 17, make records, read documents or read preserved copies, and send no requests to the website. Row 4 sends one request to confirm containment, and row 16, the tester's handoff, sends requests under the limits written on that row.
- The only targets for a request are systems the business owns, or has written authority to test, and any email or text a request sets off goes to an inbox or number the business controls. A response that carries a real person's information ends the check: note the request and what came back, open nothing more and pass it to the person running the assessment.
- Systems run by vendors, such as a form tool, booking engine, mailing platform or payment provider, come into the record through their contracts and their own incident contacts. Before any testing reaches a vendor's platform, read what that vendor's terms permit, and put the question to the vendor in writing if the terms say nothing.
- No row tells anyone to notify affected individuals, the OAIC or anyone else about the breach. Rows 13 and 14 record what the Act and the OAIC say about notification, and the entity decides, with its adviser.
Who decides, and what the OAIC can do
The entity decides, with its own legal adviser, whether it is aware of reasonable grounds to believe there has been an eligible data breach, and which option for notification is practicable.
Part 4 of the OAIC's guide says the Commissioner's roles under the scheme include receiving notifications of eligible data breaches; encouraging compliance, including by handling complaints, conducting investigations and taking other regulatory action in response to instances of non-compliance; and offering advice and guidance to regulated entities and information to the community about how the scheme operates.
Section 13(4A) of the Act says that if an entity contravenes subsection 26WH(2), 26WK(2), 26WL(3) or 26WR(10), the contravention is taken to be an act that is an interference with the privacy of an individual. Section 26WU applies where the Commissioner has reason to believe that a person or entity has information or documents, or can answer questions, relevant to an actual or suspected eligible data breach of an entity or to an entity's compliance with the requirements in Division 3 of Part IIIC, and lets the Commissioner, by written notice, require the person or entity to give information, produce documents or answer questions of the kind the notice specifies. How either provision might bear on your business is for your adviser to say.
Other parties make their own decisions on their own terms. If the business holds cyber insurance, what the policy asks it to do and when is for the business to read with its broker, and the OAIC's Part 2 notes that where an entity holds an insurance policy for data breaches, that insurer may have a pre-established panel of external service providers, or a hotline to assist in the event of a data breach.
Where a ransom or extortion payment has been made, ASD's ACSC report page says that under section 27 of the Cyber Security Act 2024 a reporting business entity has an obligation to report to the Government a ransomware or cyber extortion payment it has made, or is aware another entity has made on its behalf.
The Cyber Security Act 2024 guide covers that duty, and the cyber insurance guide covers what policy forms ask.
After the fix: where Clever Ops testing comes in
Forensics, incident response, malware cleanup and round-the-clock monitoring are not Clever Ops services.
A Clever Ops report is evidence for a decision someone else makes, and it is not a compliance opinion; whether an incident is notifiable stays with the entity and its adviser. While an assessment is running, the calls that come first go to the host, the business's legal adviser and an incident response provider, after checking what any cyber insurance policy asks the business to do before it engages one.
Clever Ops comes in once a specialist has cleaned the site or application and the fix is in, as the post-remediation tester in row 16: checking whether the weakness that exposed the information is still open, and what else the site exposes within the agreed scope.
The right service turns on who logs in. A site without customer logins fits the website security check, which looks from outside at what any visitor can reach without signing in: certificates and transport security, security headers, cookie settings, exposed files and admin paths. It does not go past the login.
On WordPress, the security audit follows each role that can log in, and for WooCommerce stores and member sites it takes in customer accounts, orders and checkout pages as well. Where customers or staff sign in to an application built for the business, the web application and API penetration test is the fit.
Its coverage includes sign-in and session handling, what each role can see and change, whether one customer can reach another customer's records, direct calls to the API rather than only through the screens, and file uploads. Live Shopify storefronts and Laravel applications sit outside these services.
Checks that found nothing go into the report as negative results for the build and accounts tested, alongside what was not checked, and after one consolidated retest every finding carries a status of fixed, partially fixed, still reproducible or not retested. Destructive actions are off by default for real people's accounts and records that existed before the test, and reset, lockout and deletion checks run only on test accounts and test records where the rules of engagement allow them.
Evidence has its session tokens, passwords, keys and personal details removed by redaction, is kept encrypted in Australia, is processed outside Australia only if the statement of work approves that, and is deleted when the data handling terms say.
What the entity and its adviser get is evidence to weigh in their own decisions; it is not a forensic investigation, a compliance opinion or a certificate.
Some of the work belongs elsewhere. Clever Ops does not test a vendor's own product or platform: the systems behind a form builder, booking engine or email marketing tool belong to their vendors. If a tender, panel or contract specifies a CREST-accredited provider, that work goes to a firm with the accreditation, since Clever Ops is not CREST accredited.
If the site is one Clever Ops built or hosts, the work is called a security review because it is not independent, and the report names the builder and the host.
Frequently asked questions
It is the scheme Part IIIC of the Privacy Act 1988 sets up for notification of eligible data breaches. The OAIC explains that under the scheme, an organisation or agency covered by the Privacy Act 1988 must notify affected individuals and the OAIC when a data breach is likely to result in serious harm to an individual whose personal information is involved. The Act's own tests are in sections 26WE to 26WL: what an eligible data breach is, the assessment of a suspected one, and the statement and notification that follow. The OAIC says the requirements apply to breaches that occur on or after 22 February 2018.
Under section 26WH of the Privacy Act 1988, an entity that is aware that there are reasonable grounds to suspect that there may have been an eligible data breach of the entity, and is not aware that there are reasonable grounds to believe that the relevant circumstances amount to one, must carry out a reasonable and expeditious assessment and take all reasonable steps to ensure that it is completed within 30 days after the entity becomes aware of those grounds to suspect. The OAIC says the Commissioner expects entities, wherever possible, to treat 30 days as a maximum and to endeavour to finish in a much shorter timeframe. Where an assessment cannot reasonably be completed within 30 days, the Commissioner recommends that the entity document this, so it can demonstrate that all reasonable steps were taken to complete the assessment within 30 days, the reasons for the delay, and that the assessment was reasonable and expeditious. When awareness began is a question of fact for the entity and its adviser.
No. Under section 26WE(2), an unauthorised access or unauthorised disclosure is an eligible data breach only where a reasonable person would conclude that it would be likely to result in serious harm to any of the individuals to whom the information relates, and a loss only where unauthorised access or disclosure is likely to occur and, assuming it were to occur, a reasonable person would reach the same conclusion. Under section 26WF(1), where the entity takes action before the access or disclosure results in serious harm to any of those individuals and, as a result, a reasonable person would conclude that it would not be likely to result in serious harm to any of them, the access or disclosure is not, and is taken never to have been, an eligible data breach; section 26WF sets out matching rules for a loss. Sections 26WD, 26WJ, 26WM, 26WN, 26WP and 26WQ set out further exceptions, and sections 26WS and 26WT set out exceptions to a direction from the Commissioner. Whether any of that fits a particular incident is for the entity and its legal adviser to decide.
Section 26WK(3) says the statement must set out the identity and contact details of the entity; a description of the eligible data breach the entity has reasonable grounds to believe has happened; the particular kind or kinds of information concerned; and recommendations about the steps that individuals should take in response. Under section 26WK(2) the entity gives a copy to the Commissioner as soon as practicable after becoming aware of the grounds. Under section 26WL(2) it takes reasonable steps to notify the contents to each individual to whom the information relates, or to each individual at risk, where that is practicable, and otherwise publishes the statement on its website (if any) and takes reasonable steps to publicise it. The OAIC says an entity should use its online Notifiable Data Breach form, and reminds entities to attach a copy of the template notification to affected individuals.
To some. The OAIC says small business operators generally do not have obligations under the APPs unless an exception applies, and that in certain circumstances a small business operator must comply with the APPs, and therefore with the scheme, including where it holds health information and provides a health service, trades in personal information, or has opted in under section 6EA of the Privacy Act. It also says entities with Privacy Act security obligations for particular types of information only, such as small businesses required to secure tax file number information, do not need to notify about data breaches that affect other types of information outside the scope of those obligations. Whether your business is covered is a question for your adviser.
No. A test looks at a website or application as it stands, on the build and accounts tested. After a fix, it can show whether the weakness still reproduces and whether similar routes share it. It cannot show who reached information in the past, what they took or whether serious harm is likely, and the text of Part IIIC does not mention security testing. Those facts come from a forensic responder's work on preserved evidence, and the decision belongs to the entity and its legal adviser.
Scope the right test for your application
Tell us what you have built, who uses it and when it needs to launch. You will get a plain-English recommendation: a website audit, a web application test or a referral to a specialist.