DKDKCISSP
← Cybersecurity on the GroundCHALLENGE

The Risk Scored Low While the SOC Sees Attacks

By Deepak Kumar, CISSP

The problem: The risk team scores the customer payment API as Low. The score comes from a yearly review, done in the calm of a workshop. Meanwhile the security operations team (the SOC) sees password-guessing attacks against that same API every day. The SOC mentions it in a chat channel. The risk register never changes, and the board report says Low.

Words used here: an API is a door that lets one computer system talk to another. The risk register is the list of risks the company has noted, each with a score. The SOC is the team that watches alerts and deals with attacks. Password guessing means trying many passwords against one account until one works.

Think of last year’s almanac

A forecaster looks in a book written last year. It says, “March is dry.” A farmer standing in a field in March is soaked to the skin. The book is not lying. It was true on average, and it was written by someone careful. But it cannot see the sky today.

The risk register is the almanac. The SOC is the farmer. Both are needed, and only one of them can see what is happening now.

The almanac and the rain
Almanac says: March is dry. It is pouring.
March is dry.
Almanac
I am soaked to the skin.
Farmer
The book is careful. It cannot see the sky today.

What it looks like at work

The yearly risk review scored the payment API Low, because it sits behind a login and the owner said it was well built. Since then, the SOC has seen about 2,000 failed sign-in attempts a day against it, from addresses around the world. These are example numbers. The analyst posts a note in the team chat and carries on with the day’s alerts.

The risk team never sees the chat. The register still says Low. The CISO builds the board report from the register. Three months later, an attacker gets in using a password reused from another leak, and the board asks why a risk with daily attacks was called Low.

How a Low score survives daily attacks
  1. 11. Yearly reviewPayment API scored Low
  2. 22. Attacks beginAbout 2,000 failed sign-ins a day
  3. 33. A chat noteThe SOC posts it in team chat
  4. 44. Register unchangedBoard report still says Low
  5. 55. BreachA reused password gets in
The two teams have no route between them.

Both teams have a fair point. The risk team scores on evidence gathered carefully, with the owner, and a yearly review is a reasonable rhythm. The SOC sees real attacks but each alert is one small signal, and they are busy. Nobody ignored anything on purpose. The two teams simply have no route between them.

You will notice it when:

  • The risk register is updated once a year, whatever happens in between.
  • The SOC says “we told them” and the risk team says “nobody told us.”
  • Attack information lives in chat messages and not in tickets.
  • The same system is Low in one report and “under attack” in another.
  • The risk team has never read a SOC monthly summary.

What exactly needs fixing

The fix is not a second yearly review. It is a small, fast route for real attack evidence to reach the risk score, and a promise to look again within days. This card shows the whole problem on one page:

A Low score under daily attack
WHAT IS BROKEN

The payment API is scored Low while the SOC sees password-guessing attacks against it every day.

HIGH
WHERE IT IS

The yearly risk register, and the SOC’s team chat.

WHAT COULD GO WRONG

A system rated Low is breached after months of visible attacks.

WHAT IT COSTS US
  • Wrong board picture
  • Late reaction
  • Register loses trust
  • SOC stops reporting
THE FIX, IN THREE MOVES
  1. 1
    1. Raise a flagOne short form when the SOC sees real attacks on a named system
  2. 2
    2. Re-score in 5 daysThe risk team re-scores or writes down why not
  3. 3
    3. Meet monthlyOne list: flags raised, scores changed or kept, and why
WHO FIXES IT

Who fixes it: SOC lead and risk lead

BY WHEN

By when: form and promise live in 2 weeks

A blank copy of this card is in the Tools section.

Why it happens

  • The risk register is a yearly snapshot. Nobody built a way to update it between reviews.
  • SOC findings go into alerts and chat. Not into a form the risk team reads.
  • The two teams have different goals. One protects today, the other plans for the year.
  • No one is named to carry the news. Each team assumes the other will ask.
  • A score feels final once it has been signed. Changing it needs a reason and a person.

What happens if you ignore it

The company makes decisions on a picture that is out of date.

  • Real attacks are missed in planning, so no money goes to protect the target.
  • The board is told Low and later learns attackers were trying every day.
  • Auditors see a register that never moves and doubt the whole process.
  • The SOC stops reporting up, because it seems to change nothing.

At a glance

QuestionAnswer
How likely is it to hurt us?Likely, wherever the register is updated yearly.
Worst case?A system rated Low is breached after months of visible attacks.
What does it cost?A wrong board picture, a late reaction, and trust in the register.
What if we wait?The next review will still say Low.
What does the fix cost?One short form and a monthly 30-minute meeting.

How to fix it, step by step

Raise a flag, re-score within five days, then meet monthly.

  1. Create a “live evidence” flag. One short form the SOC fills in when it sees real attacks against a named system.
  2. Set what triggers it. For example: repeated attacks over a week, or any attack that gets near success.
  3. Send the flag to one named person in the risk team. Not to a chat channel.
  4. Promise a re-score within 5 working days. The risk team looks at the evidence and updates the rating, or writes why not.
  5. Tell the system owner. The owner may know why the attacks matter, or what already protects the system.
  6. Meet monthly for 30 minutes. One list: flags raised, scores changed, scores left alone and why.
  7. Show the change in the next board report. With a line: “rating changed because of live evidence.”
Flag, re-score, meet monthly
1. FlagThe SOC raises a short form on a named system.
2. Re-scoreThe risk team looks again within five days.
3. Meet monthlyOne list of flags, changes and reasons.
The five-day promise makes the flag worth filling in.

The picture shows the order: raise a flag, re-score within five days, then meet monthly. The five-day promise is what makes the flag worth filling in.

Who to bring in: The SOC lead (raises flags), the risk lead (re-scores), the system owner (adds what they know), and the CISO (sees the changes in the board report).

Tools and templates

Live evidence flag

FieldWrite here
SystemThe name of the system under attack
What the SOC seesPlain words and numbers, for example failed sign-ins per day
Since whenThe date it started
Current scoreWhat the register says today
Raised byName and date
Re-score dueFive working days from the date raised

Blank finding card

FieldWrite here
What is brokenOne line, in plain words
Where it isThe system, team, or process
What could go wrongIn one sentence
What it costs usMoney, time, a failed audit, lost trust
The fixThe exact action
Who fixes itOne name or one team
By whenA date, not “soon”

Email to start the talk

Subject: A way for live attacks to reach the risk register

Hi both. The SOC has seen daily attacks on the payment API, but the register still says Low and the board report uses that score. I would like us to try a one-page flag the SOC fills in when it sees real attacks, and a re-score within five working days. Can we meet for 30 minutes this week? I am not looking for who should have known. I am looking for a route that makes sure the register moves when the world does.

Field tip

If you see the same attack three days in a row, stop treating it as noise. Fill in the flag. It takes two minutes.

Words used in this chapter

  • API: a door that lets one computer system talk to another.
  • Risk register: the list of risks the company has noted, each with a score.
  • SOC: the security operations team that watches alerts and deals with attacks.
  • Password guessing: trying many passwords against one account until one works.
  • Re-score: to look at a risk again and set a new rating.
  • Live evidence: proof of something happening now, not at the last review.