Compliance evidence for ISPs and operators
ISP regulatory compliance: what providers need to be able to show
Where deep packet inspection is concerned, ISP regulatory compliance comes down to being able to show three things: that the filtering requirements that apply to you are applied at your internet gateways, that you can report on your traffic and on the policies you applied, and that subscriber data is handled under your data rules. Your exact obligations come from your licence and the regulator. This page sets out the evidence a compliance team should keep for each area, and how Interkey Intelligent DPI supports it with classification, content filtering, policy execution and reporting.
A Saudi product, built by Interkey
For compliance officers, regulatory affairs and network teams at ISPs, data service providers and telecom operators. Reviewed November 2026.
- The requirement:in Saudi Arabia, CST supplies filtering lists to data service providers to apply on their internet gateways.
- The evidence:a record linking each requirement to a rule, its approval, its execution and its results.
- The gaps that hurt:encrypted traffic that is not recognised, over-blocking that is not measured, and changes made outside change control.
- What DPI contributes:classification, content filtering, policy execution feedback and a reporting layer for regulatory teams.
Where DPI touches ISP regulatory compliance
In Saudi Arabia, the Communications, Space and Technology Commission (CST) provides internet filtering services in the Kingdom and supplies filtering lists to data service providers to apply on their internet gateways. Source: CST internet filtering service
| Area | What you need to be able to show | What DPI contributes |
|---|---|---|
| Filtering | Each requirement that applies is applied, from the date it applies, across your gateways | Application classification and content filtering to meet national regulatory requirements, with a URL library |
| Encrypted traffic | Requirements are applied to encrypted and tunnelled traffic too | Classification from flow behaviour, without decryption |
| Accuracy | Rules act on the traffic they target and not on other traffic | Per-class results measured during the proof of concept and after significant updates |
| Reporting | You can describe your traffic and the policies you applied | Live View, real-time policy execution feedback and a reporting layer for regulatory teams |
| Data handling | Subscriber data is kept and accessed under your data rules | Metadata output; retention and access set by your data-governance rules and the contract; can operate entirely in the Kingdom |
| Change control | Every change to a list, model or rule is approved and recorded | Policies are deployed and adjusted in real time; your approval route decides what is released |
How filtering lists reach the platform, and the filtering capability itself, are covered on ISP content filtering for national requirements.
Building the compliance file
A compliance file is the set of records that lets you answer a regulator's question with evidence rather than assurance. Build it before the first rule is enforced:
1. Obligation register.Every requirement that applies to you, its source, the date it applies from and the person who owns it.
2. Policy catalogue.For each rule: the obligation it implements, its scope (application, subscriber group or cell), its action and its owner.
3. Approval records.Who approved each rule's move from report-only to enforcement, when, and on what evidence.
4. Execution records.Policy execution feedback showing each rule running as approved, kept for the period your rules set.
5. Accuracy results.Per-class detection and false-positive results from the proof of concept and from re-tests after significant updates.
6. Data inventory.Each type of record the platform produces, where it is kept, its retention period and who can access it.
7. Change log.Every list, model and policy update, with its approval and the date it went live.
Reports your compliance team will draw on
| Report need | Source in Interkey Intelligent DPI |
|---|---|
| Is each rule running as approved? | Real-time policy execution feedback |
| What does our traffic consist of? | Network-wide view with drill-down to application and flow level |
| How is our traffic changing? | Network- and application-level traffic trend analysis |
| What do we submit to the regulator? | The reporting layer for regulatory and capacity teams |
Monitoring latency is measured in seconds, so the view your compliance team sees is close to real time. Which reports your deployment produces is agreed with your team during scoping and documented in the integration plan.
Common compliance gaps
- A list applied without a record.The rule is live, but nobody can show when it went live or who approved it.
- Encrypted traffic not recognised.A requirement is applied to traffic the platform can identify, and encrypted or tunnelled versions of the same application pass unrecognised.
- Over-blocking not measured.False positives are never measured, so nobody knows what a rule acts on beyond its target.
- Retention left undefined.Records are kept for as long as the system happens to keep them.
- Changes outside change control.A model or policy update reaches the live network without approval.
Put compliance into the proof of concept
Interkey recommends a proof of concept in 3 stages. Use each one to produce compliance evidence:
1. Written acceptance criteria.Include the requirements in scope, the reports compliance needs and the data rules for the test.
2. Out-of-band classification against verified traffic.Measure false positives as carefully as detection, per class. Method: Measuring DPI classification accuracy.
3. Report-only policy, then the inline decision.Record what each rule would have done before it acts on traffic, and keep that record in the compliance file.
See also: How Interkey runs a DPI proof of concept
Questions about ISP regulatory compliance
Does a DPI platform make an ISP compliant?
Who decides what must be filtered?
How do we show that encrypted traffic is covered?
Does subscriber data leave Saudi Arabia?
Can we test a rule before it is enforced?
Proof of concept in three phases
Written acceptance criteria.Agree which rules are tested, on which traffic and subscriber groups, how results are measured and what counts as a pass.
Out-of-band classification against verified traffic.Measure false positives as carefully as detection for every class the rules depend on.
Report-only policy, then the inline decision.Each rule shows what it would have done; enforcement follows only after classification has proven itself.
Discuss your compliance requirements
Tell us what kind of provider you are, which requirements apply to you, what your compliance team reports today, and your timeline. Our team in Riyadh replies in Arabic or English.
- Our engineers read your request
- We agree the scope of a technical session
- You see the product on a use case you choose