A law changed — which of your legal documents actually need reviewing?

In short

A legal change does not mean every legal document has to be edited. It means, first of all, that someone has to check whether and where it lands. Between “something changed” and “this paragraph in this document for this brand needs touching” sit several narrowing steps.

Those steps can be structured: does the change concern our activity at all? Which brands? Which context — audience, market, sales channel? Which document types? And finally: which concrete published versions? Only at the end of that chain is there a review worth doing.

Whether a change requires an edit in a specific case is a legal assessment and is not answered here. This is about the route to that question.

The notice is not the problem

There is no shortage of notices. Law-firm newsletters, trade associations, industry portals and vendor mailings reliably report upcoming changes. The problem starts afterwards.

A typical notice is written about a subject, not about a company. It says what changes, but not whether four of your seven brands are affected, whether it hits the shop but not the SaaS product, whether the English version even contains the passage in question. That mapping is what costs time — and it is rarely written down, so it starts from scratch next time.

Two opposite failure modes follow:

• Over-reacting. Every notice triggers a full re-read of every document. It burns capacity, wears people out and eventually leads to notices being ignored.
• Under-reacting. The notice is read, filed under “probably doesn't concern us” and never revisited. Whether that was right is unknowable later, because the assessment was never recorded.

Narrowing down in five steps

Rather than a yes/no decision, the question is better treated as a funnel. Each step reduces the set — and each step can be documented on its own.

1. Does the change concern our activity? Does it address an industry, a sales channel, a company size or a kind of processing that actually occurs here? Most notices drop out at this step — but the reasoning belongs on the record.
2. Which brands? In a multi-brand portfolio the answer is rarely “all”. One brand sells to consumers, one to businesses, one runs an information site with no ordering flow at all.
3. Which context? Market and audience often decide: a point that bites in the consumer business of one market may sit differently in B2B or in another target market.
4. Which document types? Only here does a topic become a document: legal notice, privacy policy, terms, withdrawal notice, product-specific conditions. Some changes touch exactly one type, others cut across.
5. Which published versions? The final step is the one most often missing: which concrete versions are live right now — per language, per market, per website? Without that list it is unclear what the review is even about.

A deliberately abstract example

Concrete statutes make poor examples: they age quickly, and interpreting them is not the job of a product guide. The mechanism can still be shown.

Suppose a new disclosure duty applies to distance selling to consumers in a particular market. A company operates four brands:

• Brand A sells to consumers in that market — relevant, most likely in the terms and the withdrawal notice, in that market's language version.
• Brand B sells to businesses only — not affected after review, recorded with that reasoning.
• Brand C runs an information site with no ordering flow — not affected, because no distance selling takes place.
• Brand D sells to consumers, but only in a different market — to be checked separately, because territorial scope is its own question.

One notice has become two serious review items instead of a re-read of every document of every brand — plus two recorded negative decisions that are reusable next time something similar lands. The legal assessment of that mapping remains, of course, a professional question.

Deciding to change nothing is also a result

Many processes only record what was changed. That is the smaller half. The sentence “we looked at this and concluded that nothing is needed for brand B” is a work product with its own value:

• It stops the same assessment being repeated three months later.
• It shows that a notice was assessed rather than missed.
• It reveals, when something similar happens later, which reasoning carried at the time — and whether it still carries.

Practically that means a review item needs a status that can express “assessed, no action needed”, not just “open” and “done”.

The quiet precondition: knowing what is published

The funnel assumes something that is often missing in practice: a reliable overview of which legal texts are being delivered where, in which version.

If the texts live scattered across CMS pages, shop backends, app constants and email templates, step 5 cannot be performed. All that is left is checking every channel by hand — and that is where review processes break down in practice. Why scattered storage gets expensive exactly here is covered in why legal content shouldn't live scattered across your CMS.

The converse also holds: where a single source with versioned, precisely published versions exists, the list of affected delivery states is a query rather than an investigation.

How TermShelf supports this funnel

Legal Change Monitoring is the Document Intelligence signal that starts exactly here. Publicly available legal sources are tracked and consolidated into change events. Such an event is then not broadcast as a mailing but assessed against your own material: against the brands, their profile and the documents actually published.

Where a possible need for review emerges, a review finding is created with evidence, status and history — attached to concrete documents rather than to a topic. A confirmed finding can become a proposed change in the form of an unpublished draft that runs through review and approval. Nothing is adopted or published automatically.

The check cadence depends on the plan; which signals run at which cadence is shown on the pricing page. The chain from event to publication is described in from change signal to review finding.

Limits of this approach

TermShelf does not produce legally binding content and is not a substitute for legal advice. Source monitoring does not promise to capture every relevant change: not all legal sources are machine-accessible, and relevance assessment is a pre-sort, not a determination. A finding says “this is worth a look”, not “there is an obligation here”.

The funnel therefore does not replace professional support — it makes sure professional capacity is spent where it changes something, and that the reasoning survives. A different trigger with an entirely different pattern is change on your own website: your website changed — should your legal content be reviewed too? The overall picture is in the hub keeping legal content current.