Skip to content

InsightsSecurity

Put the Google Threat Intelligence you already pay for to work

For security teams with the license and no analyst: two weeks of onboarding, then weekly triage, monthly hunts and quarterly reports on a 6 or 12 month term.

Two weeks of onboarding set the intelligence requirements, profiles and monitors and end with a day-10 briefing; then alerts are checked every week, a briefing and hunt land every month and an executive report every quarter. The customer's owner approves every action.Onboarding · then 12 monthsWk 1Wk 2LensBriefMonthly briefing · huntQuarterly reportEvery week: alerts checkedin business hoursOwner approves actions
Written by
CloudMagic engineering team
Published
Reading time
13 minutes

The short answer

Our analysts can have your Google Threat Intelligence Enterprise or Enterprise Plus license working for a small security team within two weeks, counted from the day our analysts have access and your named owner is in place. Week one writes your intelligence requirements and sets up Threat Profiles, monitors and, if your license includes it, attack surface management. Week two connects Google Security Operations where you run it, triages every high and critical alert, tunes noisy monitors and ends with a baseline briefing on day 10. After that we run it in business hours every week, month and quarter, and serious incidents go to your incident response provider or your Mandiant Retainer.

This post is for a security lead with a team of three to ten, no threat intelligence analyst, and a Google Threat Intelligence Enterprise or Enterprise Plus license, bought alone or inside Google Security Operations Enterprise Plus or Google Unified Security. We run that license for you: two weeks of onboarding, then a weekly, monthly and quarterly rhythm on a 6 or 12 month term. We call this offer Threat Watch, and it is part of our cloud security work. This is the method we follow, and the paperwork is a service description and order form. Product details were checked against Google's documentation on November 14, 2026.

What has to be true before day one?

Your license, our access and one named owner. The two weeks count from the day our analysts can sign in to your tenant, because that approval is the step we cannot speed up. Two weeks is the promise only when all of this is done:

  • An active Google Threat Intelligence Enterprise or Enterprise Plus license.
  • Our named analysts in your Google Threat Intelligence group, with access to Digital Threat Monitoring and, if your license includes it, attack surface management.
  • Your main domains verified. Digital Threat Monitoring hides discovered passwords until an administrator verifies you own the domain; subsidiary and brand domains that surface in the workshop are verified in week one.
  • The attack surface scanner's addresses allowlisted, if you use it.
  • Access to Google Security Operations, if you run it, to set up the intelligence feed and one playbook.
  • Escalation contacts: your incident response provider or Mandiant Retainer, and who calls them.
  • A named security owner who approves every reset, block and detection rule and answers within one business day.

What happens week by week?

Week one points the product at what you care about, week two connects it to your tools and ends with a day-10 briefing, then it repeats.

WhenWhat happensWho we need from youGoogle settingsWhat exists at the end
Before day 1Access checked; order form signedSecurity owner, your adminAnalysts added; domains verifiedOwner named, escalation path written
Week 1Requirements, Threat Profiles, monitors, attack surface seeds if licensedSecurity lead; IT for domain, brand and network listsThreat Profiles; monitors; seeds; digests8 to 12 written requirements, tested monitors
Week 2Security Operations feed and one playbook; first triage pass; first exposure list; day-10 briefingSecurity owner; Security Operations adminApplied Threat Intelligence or the bring-your-own-license pipelineHigh and critical alerts triaged, baseline briefing, signed runbook
Every weekTriage in business hours; short noteOwner approves actionsNone newWhat changed and what to do
Every monthBriefing, hunt, patch-first list, tuningSecurity lead, 30 to 45 minutesRetrohunt; Security Operations searchesProposed detections for your approval
Every quarterExecutive report; scope reviewCISO or security leadCountry and Industry ProfilesReport for leadership, updated requirements

What happens in week one?

Week one turns what you worry about into written requirements and points Google's features at them.

Days 1 and 2 are an intelligence requirements workshop: the systems an attacker would want, your subsidiaries, regions, main suppliers, the executives whose names get impersonated, and your technology stack. The output is 8 to 12 written requirements, and every monitor, profile and report after that traces back to one of them.

Then we set up two or three Threat Profiles, Google's filter for your organization by target industry, region and other criteria. Google says the most accurate recommendations need a time period of at least 90 days, so we set at least that. Those 90 days are a look-back over Google's intelligence, and Google says a settings change can take up to 5 minutes to update the recommendations. What takes time is tuning the criteria until the recommendations match what you care about. We start that in week one and repeat it weekly.

Digital Threat Monitoring searches the surface and dark web, including paste sites and forums. We start from Google's four monitor templates (Domain Protection, Initial Access Broker, Compromised Credential and Ransomware), add custom monitors for your brands, executives and main domains, and test each one before it goes live. If your license includes attack surface management, we give it seeds such as domains and netblocks, and it builds an inventory of what you expose to the internet. Scheduled digests close the week.

Google's dark web intelligence is in preview, with its relevance system labeled Beta. We set it up only where your tenant has it and you approve, beside Digital Threat Monitoring as Google advises, and nothing in the two weeks depends on it.

What happens in week two?

Week two connects the intelligence to your tools and shows where you stand.

Where you run Google Security Operations, we connect the intelligence on days 6 and 7. Security Operations Enterprise Plus includes Applied Threat Intelligence, a pipeline Google manages, and we confirm it is flowing. On Security Operations Standard and Enterprise with an active Google Threat Intelligence license, we set up the bring-your-own-license pipeline and check that data arrives under the GCP_THREATINTEL log type. Then we build one playbook that enriches monitoring alerts and assigns each to an owner. If Security Operations access arrives after day one, this moves into the first month.

Google's Triage and Investigation agent in Security Operations, generally available, prepares an assessment of an alert with its evidence and reasoning. The agentic platform in Google Threat Intelligence, generally available on Enterprise and Enterprise Plus, answers questions and summarizes files. We treat what both produce as drafts. A CloudMagic analyst checks each one before it reaches your note, and your named owner approves any action that follows.

Days 8 and 9 are the first triage pass: every high and critical alert so far, with noisy monitors tuned so older findings do not bury new ones. Where attack surface management is licensed, the same two days produce the first exposure list, which joins what you expose with Google's vulnerability intelligence and evidence of exploitation. Day 10 is the baseline briefing and the signed runbook. Service levels start the day after.

What goes into the day-10 briefing?

A briefing for leadership should rest on intelligence a person has already checked, and on day 10 your Threat Profiles are a week into tuning. So the briefing draws mainly on sources that are ready from the first day: Google's Country and Industry Profiles (which Google updates quarterly from its threat intelligence group's curated data), Google's finished intelligence reporting and Mandiant research, and the first findings from your own monitors and attack surface.

Here is the outline for Corrandale Bank, the made-up regional bank from our demos. The company and every figure are invented to show the shape of the briefing, and none of it is a result.

  1. Who targets banks in your country: three actor groups from Google's Country and Industry Profiles, with their motive and what they went after in the last two quarters.
  2. What they use: the ways in and the tools behind those groups, from Google's finished intelligence, mapped to the 9 requirements written on day 2.
  3. Where you are exposed: the first exposure list, with 2 internet-facing services carrying a vulnerability Google lists with evidence of exploitation, 1 forgotten test host on a subsidiary's domain and 31 subdomains nobody had on a list.
  4. What the monitors found in their first week: 9 staff credential pairs across two old dumps, 2 lookalike domains registered this month, and no mention of the company on the ransomware sources Google monitors.
  5. The ten things to do first, ranked by evidence of exploitation, each with an owner and a date.
  6. What we will watch, requirement by requirement, and what goes to your incident response provider without waiting for the weekly note.

Threat Profile recommendations feed the weekly notes from week three, and after a quarter of tuning they carry real weight in the first quarterly report.

What happens every week, month and quarter?

Each week brings triage and a short note, each month a briefing, a hunt and a patch-first list, and each quarter an executive report.

Every week, in business hours, we triage each new alert within the time agreed in the order form, usually the same business day. You get a short note on what changed and what to do, and we chase credential resets and domain abuse until they close. Takedowns of lookalike domains are coordination: we gather evidence and file with the registrar or host, or route it through Mandiant where your service includes takedowns.

A weekly note is short enough to read on a phone. This one is for the same made-up bank, Corrandale Bank, in week five; every figure is invented to show the shape of the note, and none of it is a result.

Corrandale Bank, Threat Watch note, week 5.

What changed. 11 new alerts, all triaged by Tuesday. One Compromised Credential alert: 14 staff email and password pairs in a combolist posted to a forum on Monday, 9 for current staff and 5 for leavers. One Domain Protection alert: corrandale-login[.]example was registered on Sunday and already has mail records, with no site yet. Attack surface: the VPN gateway on the card-services subsidiary's netblock shows a vulnerability Google lists with evidence of exploitation. Threat Profile: one newly recommended actor, matched because it targets banks in your region; we read Google's explanation and kept it.

What to do. Reset the 9 current accounts: requested Monday, 7 done, 2 waiting on your yes because they are shared mailboxes. Patch the card-services gateway first; it goes to the top of this month's patch-first list. The evidence pack for the lookalike domain went to the registrar on Tuesday; we chase it weekly, and the registrar decides.

Waiting on you. Approval for the 2 shared-mailbox resets. Whether the card-services netblock stays in scope after the sale closes.

This week. 11 alerts triaged, 3 hours median time to triage, 7 credentials reset, 1 exposure closed (the forgotten test host from the day-10 briefing, now offline), 1 takedown filed.

Every month there is a 30 to 45 minute briefing and a hunt package: one hypothesis per relevant new campaign, the queries we ran, what we found, and proposed detections your owner approves before anyone turns them on. Where you run Google Security Operations, our analysts run those hunts across your own telemetry there. Without it, the hunt runs as Retrohunt rules over Google Threat Intelligence's file dataset, and you get the queries and indicators written out for your team to run in your own detection tools, because we cannot see your logs. You also get a patch-first list and service figures (alerts triaged, time to triage, exposures closed, credentials reset). Each quarter brings an executive report and a review of requirements and attack surface scope.

Three findings from Mandiant's M-Trends 2026 set this rhythm. Exploits were the most common way in for the sixth year running, at 32% of intrusions, and the mean time to exploit is an estimated minus 7 days, so the monthly list ranks fixes by evidence of exploitation. The median hand-off from initial access to a second threat group fell to 22 seconds, from more than 8 hours in 2022, so credential and access broker monitors run from week one. And organizations found 52% of intrusions themselves, which is what a weekly pair of eyes on your own monitors is for.

Why does this need Enterprise or Enterprise Plus?

Most of the service runs on features Standard does not include. Google's package sheet gives Standard no Threat Profiles, no finished intelligence reporting, no underground monitoring, no custom monitoring and search, and no agentic platform. Private file and URL scanning is a paid add-on on Standard, while Enterprise includes 100 of each a month and Enterprise Plus 1,000. Retrohunt drops to 5 jobs a month, against 25 on Enterprise. Country and Industry Profiles are Enterprise and Enterprise Plus only.

Google positions Standard for threat-intelligence-driven event triage and detections, and Enterprise and Enterprise Plus for proactive work on the actors targeting you; with Standard, your Google account team is the first conversation. Google also describes Enterprise Plus as being for organizations with strong cyber threat intelligence teams. If yours came inside Google Security Operations Enterprise Plus, you may hold that license without that team, and this service fills the gap.

How does this sit beside Mandiant?

Mandiant is part of Google. Where you hold a Mandiant service, we work inside it and escalate to it.

A Mandiant Retainer (formerly Expertise On Demand) gives you incident response with 2-hour response times on pre-negotiated terms, and Google Unified Security includes one. If you hold a Retainer, it is the first call in our runbook for anything serious. Mandiant's Essential Intelligence Access includes takedown requests, so if you hold it, takedowns go through it. Managed Digital Threat Monitoring gives you a designated Mandiant analyst who helps triage your monitoring alerts. If you hold it, that analyst keeps the triage, and we agree in writing which of the rest we take on.

Where does our service stop?

We do not respond to incidents, act in your systems, or put your data into a public lookup.

The service runs in business hours only, and we do not run detection and response, incident response or forensics. Serious incidents go to your incident response provider or your Mandiant Retainer through the path in the runbook. We prepare every change and its evidence, and nothing happens in your systems until your named owner says yes.

Google warns that any indicator submitted or queried through its web interface is added to the Google Threat Intelligence dataset, where the community can see it. So your files and URLs go through private scanning only, which Google says is not shared with third parties, and indicators that could identify you never go into a public lookup.

Your license stays in your tenant under your agreement with Google, and we never promise a takedown outcome.

What do you own, and how does it end?

You own everything, because it lives in your tenant: requirements, Threat Profiles, monitors, seeds, playbooks, detection rules, the runbook and every report. Our analysts are vetted and named in the order form, and when the term ends we remove them from your Google Threat Intelligence group. Nothing needs migrating.

A 12 month term matches Google Threat Intelligence's flat annual license and gives four quarterly cycles. A 6 month term works as a first term on a license you already hold and renews into 12. Terms renew automatically unless either side gives notice.

When do we tell a CISO not to start?

When it could not work or would duplicate what you have. On Standard alone, the upgrade conversation comes first. If nobody can approve resets and blocks within a business day, the notes go unread. If you need someone watching around the clock, or detection and response, Mandiant Threat Defense or a provider that runs that kind of operation fits better. If you are in the middle of an incident, call your incident response provider or Mandiant first. And if a Mandiant analyst already triages your alerts through Managed Digital Threat Monitoring, you may not need us.

Otherwise, talk to one of our engineers about your license.

Frequently asked questions

Do we need Google Security Operations for this?

No. The service runs on Google Threat Intelligence alone, and without Security Operations the monthly hunt hands your team queries and indicators to run in your own tools. With Security Operations, week two adds the intelligence feed and an enrichment playbook, and our analysts run the monthly hunts across your telemetry there.

What does it cost?

The service description and order form are already written; you choose the term and whether you run Google Security Operations, and sign. There is a fixed onboarding fee and a monthly fee, and a free call with one of our engineers to check your license and access comes before you sign. Ask us whether Google funding applies to your project.

Can you build custom automation on top?

Yes, as a separate build. Anything that needs custom code, such as an agent that drafts tickets in your ticketing system from alerts, follows our 30-day plan.

Sources

Bring us the question you are working on.

A senior engineer reads every message and replies within one business day.