How to Set Up and Test a Call Tree and Emergency Contact List

Call tree structure, a backup channel independent of corporate systems, a KVKK-compliant emergency contact list and how to measure an out-of-hours reachability test.
Technical summary
Technology
Call tree, SMS, mass notification, KVKK, ISO 22301
Topic
How to reach teams during a disruption
Level
Beginner–intermediate
Prerequisites
  • Defined crisis and recovery teams
  • A draft business continuity plan
In this article7 sections

In the first hour of a disruption, the scarcest resource is often not a system but being able to reach the right person. If no way of alerting teams has been set up in advance for when corporate email, Teams or the IP phone system goes down, the crisis team can’t assemble and the recovery team starts work hours later. Below we explain the structure of the call tree, communication channels independent of the corporate infrastructure, protecting the personal data on the list and how to test and measure the tree.

The purpose and structure of a call tree

A call tree is a hierarchical notification structure that defines, when an incident is reported, who calls whom, in what order and through which channel. Its job isn’t just to pass on the news. Confirming the message was received and reporting anyone who can’t be reached back up the tree is also its job. A good call tree works in both directions: notification goes down, confirmation comes back up. No confirmation, no tree.

Apply these rules in the design:

  • Branch width: Keep the number of people one person calls limited. With manual calls, branches of more than 4–6 people per caller slow down; the first link gets stuck on the phone and the branch below waits.
  • Backup caller: Every node has a backup. If the team lead is on leave or can’t be reached, the whole branch must not break off.
  • Skip rule: If someone can’t be reached within a set time (for example 15 minutes), the person calling takes on the branch below them or hands it to the backup node. If the time isn’t written down, everyone says “Let’s wait a bit longer.”
  • Role-based structure: Build the tree with roles, not names (IT on-call, facilities manager, finance process owner); keep the person–role mapping separate. When staff change, the structure doesn’t break; only the mapping is updated.
  • Separate trees: Separate trees or lists are needed for the crisis management team, technical recovery teams and all employees. Reaching all employees is a different job from assembling the crisis team and is usually handled with a mass notification tool.

Below is a small sample structure; in the real plan, each node also has its primary and backup channel details.

Incident report (IT on-call / security officer)
└── Business continuity coordinator  [backup: IT manager]
    ├── Crisis team leader (general manager)  [backup: deputy general manager]
    │   ├── Communications lead
    │   └── Legal / KVKK officer
    ├── IT recovery lead  [backup: system administrator]
    │   ├── Systems team (2 people)
    │   └── Network team (2 people)
    └── Process owners coordinator
        ├── Production planning
        ├── Sales / orders
        └── Finance
Confirmation flow: each branch lead reports the reached / not reached list for their
branch to the coordinator; at T+30 min the coordinator reports the overall status
to the crisis team leader. (The tree shows notification order, not the chain of authority.)

Primary and backup communication channels

When choosing a channel, there’s one criterion: will this channel still work in this type of incident? The table below compares common channels together with their dependencies.

Channel Dependency Strength Limitation
Corporate email / Teams Identity infrastructure, internet, cloud service Keeps records, group communication is easy May be the first channel lost in a cyber incident
IP phone system / desk phone Local network, PBX server, power Familiar to use Doesn’t work if the building or network is lost
Mobile voice call Mobile operator Independent of corporate IT Network congestion in a regional disaster
SMS Mobile operator Gets through on low bandwidth Limited delivery confirmation and response collection
External mass notification service Service provider, internet or operator Voice/SMS/app, confirmation report Requires a contract and data processing terms
Radio / satellite phone Its own equipment Independent of the network Cost, training, limited number of people

Define at least two independent channels: one used on a normal day and tied to the corporate infrastructure; the other completely independent of the corporate infrastructure. The order of channels can change with the type of incident. In a ransomware incident, there’s a chance the attacker is reading the corporate messaging platform and email. In that case, crisis team communication moves to the independent channel from the start. Don’t let the attacker read the plan.

Communication when corporate systems go down

A design mistake you may come across is the emergency contact list living only in corporate systems: the list in SharePoint, phones on the IP PBX, people in Entra ID. When the identity infrastructure goes down, neither the list nor the channel can be reached. Take these precautions:

  • Offline copy: Each crisis team member has a copy of the current list on an encrypted mobile device or printed in a sealed envelope. Copies carry a version date, which makes it easier to collect old copies.
  • A group independent of corporate accounts: A communication group for the crisis team that works independently of corporate identity is set up in advance, and its rules (which information can be shared) are written down. If a personal messaging app will be used, which information must not pass through this channel should be clearly defined; customer data and technical details shouldn’t be moved here.
  • Assembly point and time: What to do if no channel works is also in the plan: for example, “If communication can’t be established, the crisis team meets at address X in the first business hour after the incident is noticed.”
  • Recovery team access: The IT team must be able to reach break-glass accounts, the password vault and documentation through a route independent of corporate systems. Otherwise the call tree assembles the team, but the team can’t get into the console; if the password vault is on an encrypted server, that’s where the work ends.

For organizations using Microsoft 365, the article continuity in Microsoft 365 and cloud infrastructure explains in detail what dependencies Teams and email rely on and what needs to be planned on the cloud side.

Keeping the list current and protecting personal data

An emergency contact list can contain personal data such as an employee’s personal mobile, home phone and the details of a relative to call in an emergency. This data falls under Law No. 6698 on the Protection of Personal Data (KVKK) (for the law and the Authority’s guides, see kvkk.gov.tr). Preparing the list is a compliance task:

  • Data minimization: The list contains only the information needed to reach people. Fields such as home address or ID number aren’t needed for a call tree.
  • Legal basis and notice: Determine with the legal department which processing condition personal numbers are processed under, and inform employees with a privacy notice. Assess third-party data, such as relatives’ details, separately.
  • Access and distribution control: The whole tree isn’t distributed to everyone; each branch lead sees their own branch. The number of printed copies and who holds them is recorded.
  • Mass notification service: If you use an external service, sign a data processing agreement with the provider and assess where the data is stored and whether it’s transferred abroad.
  • Leavers: Add a step to the leaver process to remove the person from the emergency contact list and the mass notification service.

From the perspective of information security and continuity compliance, the most effective way to keep the list current is to link it to HR records and have people verify their own details periodically: once a quarter, employees are asked to confirm their details. Joining, role changes and leaving should trigger an update to the list. A hand-maintained Excel list goes out of date within six months.

Call tree testing and measuring reachability

A call tree test is the cheapest and lowest-risk functional exercise within ISO 22301 business continuity management; that’s why it can be run more often than other exercises. Set up the test with these steps:

  1. Standardize the test message: The message starts with “TEST” and asks the recipient for a clear confirmation (“This is an exercise. If you received this message, reply with 1”).
  2. Vary the timing: A test during office hours validates the list; out-of-hours, weekend or pre-holiday tests show real reachability. Get management approval before an out-of-hours test.
  3. Keep timestamps: For each node, the call/notification time, confirmation time and the channel used are recorded.
  4. Simulate a channel failure: In some tests, treat the primary channel as “not working” and confirm the backup channel can really be used.
  5. Report the result: The share of people reached, the time to the last confirmation, the number of wrong or outdated records and corrective actions.

A simple table like the one below is enough for the record; if you use a mass notification service, this data usually comes from the service’s report.

Role / node Notification Confirmation Channel Note
IT recovery lead 22:04 22:07 Mobile call —
Network team 1 22:09 22:31 SMS Call not answered
Finance process owner 22:06 — Mobile call Number out of date

Two indicators are enough for measurement: the share of people confirming within the target time, and the confirmation time of the last critical role. Derive the target time from the time the plan allows for assembling the crisis team. The test isn’t closed until the faulty records it found are corrected. The “number out of date” row in the table is a finding on its own. To see how the call tree fits into the annual program with other exercises, see the article types of business continuity exercise.

Checklist

  • The tree is built on roles, and every node has a defined backup.
  • Branch width is limited, and the skip rule and its time are written down.
  • At least one channel independent of the corporate infrastructure is defined and tested.
  • An offline copy of the list, with its version date, is held by crisis team members.
  • An assembly point and time are set for when no channel works.
  • There’s access to emergency accounts and documentation independent of corporate systems.
  • Personal data is minimized, notice has been given and access is restricted.
  • Joiner, role change and leaver processes update the list.
  • Regular testing, including out-of-hours tests, is done, and results and corrections are recorded.

The call tree is the field version of the activation section of the business continuity plan. To design the plan, roles and communication structure together, you can draw on business continuity plan and business impact analysis work.

Frequently asked questions

Is sending a mass SMS enough instead of a call tree?

Mass notification is a suitable method for informing all employees. For crisis and recovery teams, however, you need confirmation that the message was received, a backup brought in for anyone who can’t be reached, and clarity on who takes on which task. For critical teams, use a call tree with confirmation or a notification service that can collect confirmations.

Can employees’ personal phone numbers be kept on the emergency contact list?

They can; but this is a personal data processing activity. A legal basis must be determined, employees must be informed, only necessary data should be kept and access to the list must be restricted. Clarify implementation details with the organization’s legal department or KVKK officer.

How often should the call tree be tested?

There’s no fixed frequency; as the call tree is low-risk, it can be tested more often than other exercises. A common approach is to verify the list every quarter and run an out-of-hours, confirmed reachability test at least once a year.

Can Teams or email be the call tree’s primary channel?

In normal operations it can, but it can’t be the only channel. When the identity infrastructure, internet connection or cloud service goes down, these channels go too. In a cyber incident, corporate channels may also be monitored. Define and test at least one backup channel independent of the corporate infrastructure.

Support

Contact us for support or a discovery session on this topic.

Contact form
[email protected]

Eretek Compliance

We work with you on compliance with ISO 27001, IATF 16949, TISAX and similar standards, covering risk assessment, control design, documentation and audit preparation.

Knowledge Center

Other articles