This sheet is past its review date. Deadlines or advice may have changed since 5 Nov 2026.
Threat intelligence: which warnings deserve your time, and what do you do with them?
This month, write down the products, suppliers and sector you need warnings about, and subscribe to the free sources that cover them: CERT-Bund's WID, CISA's and ENISA's catalogs of exploited vulnerabilities and your vendors' advisories. Give every warning that matches your inventory an owner and a decision (patch, block, hunt or detect) within one working day, or the same day if it is exploited and exposed, and judge your intelligence by the actions it produced, not by what you read.
If you only do five things
Write down what you need to know.
List your internet-facing products and versions, critical suppliers, sector and countries. A warning that touches none of them is reading, not work.
Turn a warning into an action
Does it touch something on your list?
Match product, version and exposure against the inventory. CSAF, the JSON format of CERT-Bund's feed2, lets tools compare advisories with an asset database or SBOM7.
Is it being exploited?
Check KEV, EU KEV and the advisory. Verizon found the chance of renewed exploitation roughly halves 30 days after the last observed activity23, so when you must choose, put recent activity first.
Which action fits?
Patch or contain within the patching sheet's deadlines. Block indicators that are recent and specific. Hunt in your logs back to the first known exploitation. Add a detection when the advisory describes a technique your logs can show.
Who does it, and did it hold?
Open a ticket with one owner and a due date. Close it with the outcome (patched, blocked, nothing found or not affected) once a check confirms it.
Decide what you need to know
- Your technology. Edge devices, identity, remote access and what suppliers run for you. ENISA knew the vector of only 5.2% of unauthorized access incidents in the EU in 2025; 60.4% of those used a vulnerability24.
- Your sector and region. Public administration, business services, transport, manufacturing and finance were the most affected EU sectors in 2025, and ideology-driven operations made up 57.3% of incidents24.
- Your suppliers and your own alerts. Add your MSP and cloud suppliers, and feed phishing reports and blocked sign-ins back in.
- Strategic reports, once a year. Read your sector's part of the ENISA Threat Landscape24, the BSI's Lagebericht25 and M-Trends or the DBIR. Take one or two changes in priority from each.
Free sources worth subscribing to
- CERT-Bund's WID. Short advisories for organizations on the portal and by RSS1, and a CSAF feed for automated matching2.
- CISA KEV and ENISA's EU KEV. KEV lists 1,768 exploited vulnerabilities8, EU KEV those confirmed exploited against EU entities9.
- Vendor advisories for every product on your list. NIS2's rules for digital providers name suppliers as a channel to monitor11.
- The BSI-Portal and the ACS. Participants get BSI security notices (BITS), warnings and daily and monthly situation reports5.
- Microsoft Defender threat analytics. Reports show exposed devices and recommended actions, with any Defender license except Endpoint P126.
- Exposure alerts about your own addresses: see the attack surface sheet.
Indicators, techniques and ATT&CK
- Indicators expire. Hashes, IP addresses and domains are the easiest things for an attacker to change; techniques are the costliest20. Give each blocked indicator an expiry date, such as 30 days for IP addresses.
- Search the past. When an advisory lists indicators, search your logs back to the campaign's start. A hit means incident response, not just a block.
- Build lasting detections on behavior. Mandiant advises it because attackers rapidly change infrastructure and run malware in memory22.
- Tag rules with ATT&CK technique IDs. Start from the detection strategies MITRE has linked to techniques since v1817, but don't chase 100% coverage16.
Share what you see
- EU law encourages it. NIS2 Article 29 has member states let entities share indicators, TTPs and alerts voluntarily in sharing arrangements, including with suppliers10. DORA Article 45 covers financial entities12.
- Personal data needs a basis. NIS2 recital 121 says legitimate interest under GDPR Article 6(1)(f) could cover sharing IP addresses and URLs10.
- Check whether to tell the authority. DORA requires it12, and NIS2 asks member states to require it10. Under BSIG § 30(7), sharing on the BSI's platform adds no obligations6.
- Label with TLP 2.0. CLEAR is public, GREEN stays in the community, AMBER in your organization and its clients, AMBER+STRICT in your organization, RED only with the individual recipients13.
Is it changing anything?
- Actions per month. Patches, blocks, hunts and detections that started from a warning. Review any source that produced none for two quarters.
- Time from advisory to decision. For warnings that match your list. Aim for one working day, the same day if exploited and exposed.
- Learned too late. Exploited flaws in your products you first heard about from a customer, supplier or the news.
What changed in 2026
- The ACS moved into the BSI-Portal in January 2026. Existing participants must re-register there, with an ELSTER company login, by 31 January 2027 or lose their participation5.
- ATT&CK v19 split Defense Evasion into Stealth and Defense Impairment in April18. Re-tag your rules. v20 added AI techniques in October19.
- EU KEV will add manufacturers' reports. The CRA has required them since 11 September 20269.
What you can skip
- Paid feeds before free ones turn into tickets. Only 26% of KEV vulnerabilities in Verizon's data were fully fixed in 2025, and full fixes took a median of 43 days23.
- Indicator lists without context. Without campaign, date and confidence you can't judge them, and IP lists go stale fast21.
- Tracking actor names. Microsoft lists 12 other names for its Forest Blizzard, APT28 among them27. Track techniques instead.
- A threat intelligence platform before a process. A mailbox, a ticket queue and a one-page requirements list come first.
Mistakes that cost the most
- Reading without a ticket. A warning that ends in a chat has no owner. Every match gets a ticket, even when the answer is "not affected".
- Blocking feeds unfiltered. Feeds can include public DNS resolvers, popular domains and scanners. Check them against MISP's warninglists15.
- Forwarding labeled warnings to suppliers. TLP:AMBER covers your organization and its clients, not your MSP. Ask the source first13.
- Stale filters. Update feed filters and subscriptions when you add a product. NIS2's rules for digital providers require reviewing these channels at planned intervals11.
What this looks like at your size
Small team, no security staff
- Send WID's RSS feed and your vendors' advisories, filtered to your internet-facing products, to a mailbox two people read.
- Join the ACS, or ask your national CERT what it shares with companies.
- Ask your IT provider which advisories it acted on last month.
Growing, with IT or security staff
- Pull KEV, EU KEV and vendor CSAF feeds into tickets, matched against the inventory.
- Add one detection a month from an advisory relevant to your sector.
- Run MISP, which is free and open source14, or use your ISAC's.
Enterprise or regulated
- Review intelligence requirements with risk owners every quarter.
- Keep a register of the sharing arrangements you join and the notifications you made.
- Put TLP handling in your MSSP contract and report the three measures to leadership.