- Market
- Saudi Arabia
- Written for
- CISO, CIO
- Decision
- Choose a managed security operations centre and understand what it will actually watch
The path of an alert
How an alert travels through a managed SOC to your team
Most alerts die in triage, and should. The drawing shows the journey of the ones that matter — and the handover point where our analysts stop and your organisation acts.
01
Telemetry lands
Onboarded log sources stream into the SIEM, parsed and enriched, so a detection sees context rather than raw lines.
- Log sources
- SIEM
02
First-line triage
An analyst confirms or dismisses the alert against the runbook for that detection — is this the pattern the rule exists to catch, on an asset that matters.
- Runbooks
- Case management
03
Deep investigation
A confirmed alert is investigated across sources: what else did this account or host touch, when did it start, how far does it reach.
- Query tools
- Threat context
04
Escalation to your team
A confirmed incident reaches a named contact on your side, on the agreed channel, with what happened, what is affected and what to do first — not a ticket in a queue.
- On-call rota
- Agreed channels
Containment on your authority
Where you pre-authorise it, agreed containment actions — isolating a host, disabling an account — run while your team is being reached, so the clock does not wait for a phone to be answered.
The escalation path is agreed and rehearsed at onboarding; it is not improvised on the night it is first needed.
Coverage before promises
A SOC only sees what its log sources let it see
The uncomfortable truth of security monitoring is that coverage is decided at onboarding, not in the analyst chair. A SOC watching firewall logs but not the identity provider will miss the sign-in from a stolen credential; one watching endpoints but not the ERP will miss fraud in the system that holds the money. So the first piece of work is a map: which systems carry your risk, which of them emit usable telemetry today, and in what order the gaps get closed. Onboarding is then done properly — parsing verified, fields enriched, a test event traced end to end — because a source that streams unparsed lines into storage is coverage on a slide, not in reality.
Detection engineering is the second discipline, and it is continuous. Default rule packs fire on other people's estates; your detections have to know your admin accounts, your maintenance windows, your quirks — or the analysts drown in alerts that are technically correct and operationally meaningless. We tune rules against your real traffic, write new ones when an investigation shows a gap, and retire ones whose alerts have stopped earning their triage time. The rule set a year in should look different from the rule set at go-live; if it does not, nobody is doing detection engineering.
The backend decision
Where the SIEM behind a managed SOC should live
For Saudi organisations this is rarely just a cost question — regulated sectors often need security telemetry kept in the Kingdom. We run the SOC on either backend; the choice belongs to your constraints, and we put it on the table at scoping rather than after a contract.
| Option | Where the security data physically sits | Who carries the platform operations | How costs scale with log volume | How fast new detections and sources land | Best for |
|---|---|---|---|---|---|
| On-premises SIEM in your estate | In your own data centre, in the Kingdom, under your physical control — the cleanest answer where a regulator or contract requires residency. | Hardware, storage and upgrades are yours to run or ours to manage on your infrastructure — either way they are a real ongoing cost. | Capacity is bought ahead of need; a growth in log volume is a procurement, not a slider. | Changes move at the pace of your change process, which the SOC has to plan around. | Regulated estates where telemetry must stay in-country and infrastructure control matters more than elasticity. |
| Cloud SIEM in a region you accept | In the cloud region chosen at design time — including in-Kingdom cloud regions where the provider offers them and your obligations accept them. | The platform layer is the provider's problem; the SOC's effort goes into detections and triage rather than disk space. | Cost follows ingested volume, which is flexible and needs governing — noisy sources are filtered before they become a bill. | New sources and detection content land quickly, without a hardware step in the way. | Estates whose obligations are met by region choice, and teams that want the SOC's effort on analysis rather than platform care. |
Our role
What a managed SOC reports each month and what you keep deciding
- 01
The watch, run as an operation
Round-the-clock monitoring, triage and investigation against agreed runbooks, with every escalation documented — so the service is inspectable, not taken on faith.
- 02
A monthly review with substance
The service review covers what fired and why, what was escalated and how the handover went, which detections were tuned, added or retired, which log sources joined or degraded, and what the next month's onboarding and tuning priorities are — a working session, not a slide of green ticks.
- 03
Detection content as a living asset
The rules, runbooks and source documentation are maintained as deliverables you can read and, if you ever build your own SOC, take with you — the service should never hold your visibility hostage.
Which actions we may take without a call, who is woken at three in the morning, and where your data may live are your decisions; we run the watch inside them and show our work every month.
Common questions
What buyers ask
What does the monthly service review of a managed SOC actually cover?+
Four things, none of them decorative. First, the incident record: what was escalated, how the handover to your team went, and what the investigation concluded. Second, detection health: which rules were tuned, written or retired, and the false-positive burden trend. Third, coverage: which log sources joined, which degraded or went silent, and what that does to visibility. Fourth, the plan: next month's onboarding and tuning priorities. A review that only shows alert counts and a green dashboard tells you the service is being marketed to you, not managed for you.
Can our security logs stay inside Saudi Arabia with a managed SOC?+
Yes, and for regulated organisations it is often non-negotiable, so we treat it as a scoping input rather than an afterthought. The SIEM backend can run on your own infrastructure in the Kingdom, or in an in-Kingdom cloud region where your provider offers one and your obligations accept it. The monitoring service then operates against that backend — the analysts query the data where it lives rather than copying it out. What we will not do is promise residency vaguely; the design document names where every copy of the telemetry sits, including backups.
How does a managed SOC escalate an incident to a team that is asleep?+
Through a path agreed and rehearsed before it is needed. Onboarding names the contacts, the order they are tried in, the channels that reach them at night, and the severity levels that justify waking someone versus waiting for morning. For the window before anyone answers, you can pre-authorise specific containment actions — isolating a host, disabling an account — so the response is not paused on a ringing phone. And the path is tested with a drill, because an escalation matrix that has never been exercised is a document, not a capability.
How many log sources does a SOC need before the service is worth buying?+
The number is the wrong measure; the composition is the right one. A SOC watching five well-chosen sources — the identity provider, the endpoints, the email platform, the perimeter, and the system that holds your money or your data — sees more of your real risk than one ingesting forty sources chosen by ease of connection. Scoping therefore starts from where a compromise would hurt, works out which systems would show its traces, and onboards those first. Volume grows later, governed by what each new source adds to visibility rather than to the ingest bill.
In this cluster
Scope your monitoring with us
A scoping session that lists the log sources worth onboarding first, the escalation contacts on your side, and where the SIEM backend and its data should live.
What happens next
- 1We reply within one business day.
- 2A 30-minute scoping call with a senior consultant.
- 3A written proposal with scope, team, and timeline.
Or reach us directly

