The law that governs the deal is the law as it was.
Every research tool will hand you the current consolidated text. For a dispute about something that happened in 2022 that is the wrong document, and it fails silently, because today's version looks exactly as authoritative as the one you needed.
Set a date. See what was in force.
4 versions of one provision across seven years. Drag through 2023 and something stops working.
Implementing regulations — procurement
Standstill period between award notice and contract signature
An amendment took effect on this date. We hold the record of the change but not the text it produced.
The superseded wording was not retained in a retrievable published form. This period is a gap, and it is shown as one.
A contract signed on this date is governed by wording we cannot show you. What you must not be handed here is the current text — it contains an exemption that did not exist yet, and nothing on the page would tell you.
The provision above is invented and labelled as such on the panel. The mechanism is not: dates, version boundaries and the gap are how the real material behaves.
Feb 2023 to Sept 2024 is a hole
An amendment took effect and the wording it replaced was never retained anywhere retrievable. We know the change happened and we know when. We cannot show you what it said.
Any transaction dated inside that window is governed by text nobody can produce, and a firm needs to know that before it advises, not after.
The tempting alternative
Serve the current text with a footnote saying it may have differed. It reads well, it fills the gap, and it is how the wrong provision gets into a memo — because the footnote survives one draft and the text survives to the client.
Baz stops instead. An absence you can see is worth more than a plausible substitute, and it points you at the archive, the issuing body or the counterparty's own file — which is where that wording usually still exists.
What actually exists to search
2 of these 6 sources are fully published. Knowing which is which is most of what separates competent Saudi research from confident Saudi research.
| Source | Availability | How current | What that means in practice |
|---|---|---|---|
| Statute (Nizam) | Published | Within days of issue | Published in the Official Gazette. The reliable end of this table. |
| Implementing regulations | Published | Within weeks | Often issued well after the statute they implement, so a law can be in force and not yet operable. |
| Ministerial and agency decisions | Partly published | Varies by body | Scattered across the issuing bodies rather than collected. Coverage is good and not complete, and we mark what we do not have. |
| Superseded versions | Partly published | Historic | The single biggest gap. Current text is easy to find; the text as it stood four years ago frequently is not. |
| Court judgments | Not published | — | Not comprehensively published, so there is no corpus to search. Any tool offering you Saudi case-law research is producing something else. |
| Regulator guidance and circulars | Partly published | Irregular | Sometimes only ever circulated to licensees. If you have it, put it in your own bank and it becomes searchable alongside the rest. |
3 rows are partial and 1 is not published at all. Those are the rows a research product should be loudest about, and they are the rows most of them quietly omit.
The Arabic is the law. The English is a convenience.
An English rendering of a Nizam has no legal force, however official its provenance looks. It is a reading aid, and treating it as the text is a category error that survives right up until somebody litigates it.
So a citation always resolves to the Arabic. The translation sits beside it, marked as what it is, and never becomes the thing being quoted.
يُنشر النظام في الجريدة الرسمية، ويُعمل به من التاريخ المحدد فيه.
The law is published in the Official Gazette and takes effect from the date specified in it.
Citations resolve to the panel above. Nothing in a memo will ever quote the panel below.
Where this stops
Research narrows the question. It does not answer it, and the last of these is the one that matters.
- A gap is not a guess
- Where the superseded text was never published we say so and stop. The alternative — showing the current text with a note that it may have differed — is how the wrong provision ends up in a memo.
- Arabic governs, and translations do not
- An English rendering of a Nizam has no legal force whatever its provenance. Translations are shown alongside the Arabic for reading speed and are never the text a citation resolves to.
- In force is not the same as operable
- A statute can be in force while the regulations that make it workable are still outstanding. The timeline shows both dates because acting on the first is a common and expensive mistake.
- This is research, not advice
- It finds the provision that applied and shows you how it changed. Whether it governs your facts is the question you are paid to answer, and nothing here is a substitute for it.
The rest of the platform
Assistant
Ask the file a question. Every answer cites its source.
Drafting
From your own precedents, not a generic bank.
Arabic and English
Two versions that say the same thing.
Contract review
Marked against your playbook, not somebody else's.
Document sets
Thousands of pages, ranked by what matters.
Matters
Every document, deadline and hour against the file.
Time and billing
Captured as you work, not reconstructed on Thursday.
Deadlines
Counted from the trigger, not from memory.