DKDKCISSP
← Cybersecurity on the GroundCHALLENGE

Held to Account, With No Power to Act

By Deepak Kumar, CISSP

The problem: After an audit finding or an incident, leaders ask the head of security, “Why was this not fixed?” Security answers, “I asked IT and the project teams, and they did not do it.” Security is held to account for the result, but cannot give orders to the teams that do the work.

Words used here: accountability means being the one who must answer for a result. Authority means having the right to decide and give orders.

Think of a class monitor

A teacher makes one student the class monitor. When the class gets noisy, the teacher says to the monitor, “Why is it so noisy?” The monitor cannot give anyone a punishment or change the seating. The other students say, “You are not our boss.” The monitor gets the blame and has no power.

The class monitor
Teacher“Why is it so noisy?”
MonitorBlamed, but cannot give orders.
Students“You are not our boss.”
Held to account for the result, with no power over the work.

Security is often put in the monitor’s seat.

What it looks like at work

An audit finds a server that has not had its updates for a year. Leaders say, “This is security’s job.” Security says, “I have asked IT four times.” IT says, “It is a live server. Every change needs a slot agreed with the business, and updates sometimes break things.” The project owner says, “It belongs to a client. Please do not touch it.” The head of security still has to answer to the board.

How it goes wrong, step by step
  1. 1Audit findingLeaders ask who is in charge
  2. 2BlameSecurity has the title, so it answers
  3. 3No powerIt cannot change servers or projects
  4. 4Asks nicelyTeams say yes, then do nothing
  5. 5RepeatSame finding next time
Accountable for the result, with no say over the work.

Everyone is half right. Leaders are right that someone must answer. IT and the project owner are right that the system is theirs, and so is its uptime. Security is right that it cannot fix what it does not control. What is missing is a written split of who answers for each system, and what security can do when a fix stalls.

You will notice it when:

  • The head of security is blamed for findings in systems that other teams run.
  • “We will look at it” is the usual answer to security requests.
  • Nobody can say who owns a given system.
  • Security does the follow-up work, because nobody else will.
  • Leaders see security’s list as security’s own problem.

What exactly needs fixing

The first fix is not a stricter rule. It is a written agreement on who owns what, and what security can do when something stalls. This card shows the whole problem on one page:

Accountable, but with no power to act
WHAT IS BROKEN

Security answers for results it has no power to change.

WHERE IT IS

Who owns each system, and the path to escalate when a fix stalls.

WHAT COULD GO WRONG

Findings stay open, and blame lands on security after an incident.

WHAT IT COSTS US
  • Blame after a breach
  • Findings stall
  • Tired security team
  • Leaders lose trust
THE FIX, IN THREE MOVES
  1. 1
    Name an owner each
  2. 2
    Write the split down
  3. 3
    Agree how to escalate
WHO FIXES IT

The leadership team

BY WHEN

Signed in 60 days

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

Why it happens

  • Security was created to advise, but it is judged on results. The two do not match.
  • Nobody wrote down who owns each system. So when something is wrong, the question “whose is it?” starts an argument.
  • Teams are measured on uptime and dates. Security fixes are not on their scorecard.
  • Leaders want one person to blame. The title “head of security” is easy to point at.
  • Security has no tool to push. It can ask, but it cannot say “this stops until it is fixed.”
  • Asking nicely works only on good days.

What happens if you ignore it

The same findings come back, and security carries the blame each time.

  • Findings stay open, because nobody who can fix them feels responsible.
  • After a real incident, the blame lands on security, even if others held the key decisions.
  • The security team gets tired and good people leave.
  • Leaders stop believing the reports, because nothing improves.

At a glance

QuestionAnswer
How likely is it to hurt us?Very likely. It is one of the most common problems in security teams.
Worst case?A serious incident, with a fight about who should have fixed it.
What does it cost?Open findings, a tired team, and lost trust at the top.
What if we wait?The same pattern repeats, and the best people leave.
What does the fix cost?A few meetings and one signed page.

How to fix it, step by step

Separate who answers, who acts, and who checks. Then give security one real power.

  1. List the important systems. Start with the five that matter most. Name one owner for each, on the business or technical side.
  2. Write down who answers, who acts, and who checks. The owner answers. The IT or project team acts. Security checks and reports.
  3. Agree what security can do. Set the rule, check it, report it, and escalate in writing when a fix stalls.
  4. Agree an escalation path with times. For example: if a critical issue is open for 10 days, it goes to the head of the area. After 20 days, it goes to the leadership team.
  5. Agree one stop rule for critical issues. For example: a critical flaw can pause a release until the head of the area decides in writing.
  6. Put risk acceptance in the owner’s name. If an owner chooses to leave a gap, they sign for it.
  7. Report by owner, not by security. Show leaders a monthly page: owner, open findings, and days overdue.
  8. Start with one system. Prove it works, and then widen it.
Split who answers, who acts, and who checks
System ownerDecides and signs for the risk. It is their system.
IT or project teamFixes it by an agreed date. Reports progress.
SecuritySets the rule, checks it, and escalates if it stalls.
Security keeps one real power: escalating to the head, in writing.

The numbers in the steps are examples. Set your own with leadership.

Who to bring in: The head of security, the head of IT, the heads of the business or project areas, and one leader who can sign the agreement.

Tools and templates

Who answers, who acts, who checks

SystemOwner (answers)Team (acts)Security (checks)
Client portalHead of the projectProject team and ITSecurity lead
Finance systemHead of financeITSecurity lead
Office laptopsHead of ITITSecurity lead

Escalation ladder (examples)

How long it is openCriticalHigh
10 daysHead of the areaTeam lead
20 daysLeadership teamHead of the area
30 daysWritten risk acceptance, or a stopLeadership team

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 leak, lost trust
The fixThe exact action
Who fixes itOne name or one team
By whenA date, not “soon”

Email to the head of security’s boss

Subject: One page to agree who answers for each system

Hi [name]. Several security findings stay open because it is not clear who owns the system or who decides. I would like to agree one page that names an owner for each important system, and shows what security can escalate and when. Can we meet for 30 minutes? I will bring a draft for the first five systems.

Field tip

Do not ask for more authority. Ask for one written page that says who owns each system, and what happens when a fix stalls.

Words used in this chapter

  • Accountability: being the one who must answer for a result.
  • Authority: the right to decide and give orders.
  • System owner: the person who answers for a system and decides about its risks.
  • Escalate: take a stuck issue to a more senior person.
  • Risk acceptance: a signed decision to leave a weakness open, and carry the cost.