The requirement, in our words
A Trading Participant must run monitoring adequate to spot, immediately, a Trading Message or a run of them that harms, or could harm, a Market in its efficiency or integrity, or how a Trading Platform works. It must also run adequate post-trade surveillance able to spot, as soon as possible, a message or run that might have produced, or been meant to produce, the impression Rule 5.7.1 describes, weighing the matters in Rule 5.7.2. It must keep records of what it identifies and the measures taken for seven years from the message date.
What the rule is trying to do
The new rule creates a detection duty that sits before the interference and a surveillance duty that looks back for manipulation. The seven year clock runs from the message date, not from the alert or the review. The rule does not prescribe a technology, but it does prescribe immediacy for the real-time limb and as-soon-as-possible for the post-trade limb.
What good practice looks like
Separate the two limbs in the design: a real-time interference monitor and a post-trade manipulation surveillance suite. Calibrate alerts against the 5.7.2 factors. Record the message, the alert, the reviewer, the decision and the action. Test with a known interfering series and a known layering or spoofing pattern. Measure detection and review times.
What changed in the instrument
Item 69 inserts Rule 5.6.3A. There is no equivalent rule in the current compilation: the old regime relied on the general 5.6.3 controls and the manipulation rule.
Evidence an auditor or ASIC asks for
| Category | Record to keep | Retention |
|---|---|---|
| Monitoring design | The real-time monitoring rules, thresholds and the message fields they read, with the owner and the review cycle. | Current, plus version history. |
| Surveillance design | The post-trade scenarios mapped to Rule 5.7.2 factors, with the alert logic and the review procedure. | Current, plus version history. |
| Alert records | The alerts raised, the messages identified, the reviewer, the decision and the measures taken. | Seven years from the message date. |
| Effectiveness test | A test using a known interfering series and a known manipulation pattern, with the detection time and the outcome. | Current, with date and tester. |
Common gaps
- Surveillance covers client orders but not the firm own-account and algorithm activity.
- The alert record is kept for two years under a general policy, not seven years from the message date.
- The post-trade scenarios are not mapped to the 5.7.2 factors, so the review cannot explain its decision.
- Real-time monitoring detects a burst only after the market calls.
What to do before 17 March 2028
- Separate real-time monitoring from post-trade surveillance.
- Map every surveillance scenario to the 5.7.2 factors.
- Set retention to seven years from the message date.
- Test detection with a known pattern and record the elapsed time.
Dates and cross-references
Commences: 17 March 2028 (Rule 2 of F2026L01223 and the registration date). Retention: Seven years from the date of the relevant Trading Message.
Read with: 5.6.1, 5.6.3, 5.7.1, 5.7.2.
Source
This page reads ASIC Market Integrity Rules (Securities Markets) Amendment Instrument 2026/574 (F2026L01223), Schedule 1, item 69, against the current compilation Compilation No. 6 (F2024C01108). Retrieved 2026-10-19. The requirement above is our own statement, not the instrument's wording. Not legal advice; does not certify.