An incident response plan is not a document that proves you are prepared. It is a set of decisions made before an incident, when nobody is panicking. For an SME, the plan does not need a security operations centre. It needs a one-page contact tree, a clear owner, a backup that restores, an evidence folder and a tested decision on who calls the bank, the regulator and the customer.
The economics are hard to ignore. IBM's 2026 Cost of a Data Breach report puts the global average at $4.99 million, up 12 percent, with an average of 247 days to identify and contain. Those are large-organisation numbers, but the pattern applies to smaller firms: slow detection and unclear ownership turn a contained event into a business continuity crisis.
Where things stand in 2026
Two risks dominate the current landscape. The first is AI-enabled attack volume: IBM reports that one in four malicious breaches were AI-enabled, a 56 percent increase, and that these incidents cost more. The second is third-party exposure. The World Economic Forum's Global Cybersecurity Outlook 2026 found that 65 percent of large companies see third-party vulnerabilities as their biggest challenge, up from 54 percent the previous year.
For an SME, the most likely incidents are not sophisticated. They are phishing, stolen credentials, a compromised supplier account, a misconfigured cloud bucket, ransomware on a shared drive or a fraudulent payment request. The plan should be designed for those, not for a movie-style APT.
2027: roles, evidence and the first tabletop
2027 should be the year the plan becomes operational. Name an incident lead and a deputy. List the people who can authorise a bank freeze, a password reset, a service shutdown or a customer notification. Keep the list offline as well as online. Test the contact tree with a real call, not an email.
Under Bangladesh's Personal Data Protection Act, 2026, personal data incidents can trigger notification and accountability obligations. The Act has a phased commencement, with some provisions taking effect later, but the operational lesson is immediate: know what personal data you hold, where it is, who can access it and how you would notify affected people. Do not wait for the regulator to ask.
- Evidence: preserve logs, emails, screenshots and timelines before resetting systems.
- Containment: decide in advance who can disable an account, revoke a token or block a payment.
- Communication: draft holding statements for customers, suppliers and staff.
- Suppliers: keep security contacts and contract clauses in the same place as the plan.
Run a 90-minute tabletop with the people who would actually be in the room. Use a realistic scenario: a finance employee receives a supplier email with a changed bank account, or a manager's email is used to request a urgent payment. The goal is to find the decision that nobody owns.
2028: AI-enabled attacks and supplier incidents
By 2028, deepfake voice and video, AI-written phishing and automated reconnaissance will be ordinary. The control that matters most is not a better email filter; it is a payment and access process that does not rely on a single channel. Call-back verification, dual approval for bank changes, phishing-resistant authentication and least privilege reduce the blast radius.
Supplier incidents will also be more common. A logistics provider, payment gateway or SaaS vendor can expose your data without your systems being touched. Map your top ten suppliers by data access and business dependency. Ask what they will tell you, when, and through which channel. Put a security contact and notification timeline in the contract.
Recovery testing becomes non-negotiable. A backup is a hypothesis until it restores. Test one critical system end to end, including the time to restore, the data loss window and the manual workaround. Record the result and fix one gap each quarter.
2029: resilience reporting and board oversight
2029 is the year incident response becomes a board-level operating metric rather than an IT chore. Report time to detect, time to contain, backup restore time, open critical findings and the status of supplier reviews. Use plain language. "We can restore the order system in four hours and we would lose two hours of orders" is more useful than a heat map nobody understands.
Cyber insurance renewals will ask for evidence: MFA coverage, backup testing, incident response plan, supplier controls and training. Keep the evidence in a folder that can be exported. Do not reconstruct it the week before renewal.
Regulatory expectations will continue to tighten. The direction is consistent across jurisdictions: faster notification, better evidence, named accountability and demonstrable oversight. Build the habit now, even if your current obligations are lighter.
2030: continuous response
By 2030, the plan should be a loop, not a binder. Incidents, near misses and tabletop findings feed a quarterly review. Controls are tested continuously. The contact tree is current. The backup restore is measured. The supplier list is ranked by dependency. The board sees the same five metrics every quarter.
The mature position is not "we have never had an incident". It is "we know what we would do, we have practised it, and we can prove it". That is a more credible promise to customers, insurers and regulators.
What to do in the next 90 days
- Write a one-page plan with an incident lead, deputy and decision rights.
- List the top ten suppliers by data access and business dependency, with security contacts.
- Test one backup restore and record the actual recovery time.
- Turn on phishing-resistant MFA for email, banking, cloud and admin accounts.
- Run a 90-minute tabletop using a payment-change or account-takeover scenario.
- Create an evidence folder for logs, contracts, training and test results.
How to measure progress
- Time to detect and contain: measured in hours, not impressions.
- Backup restore time: tested quarterly on a critical system.
- MFA coverage: the share of privileged and remote-access accounts protected.
- Supplier review completion: the share of tier-one vendors reviewed in the last 12 months.
- Tabletop actions closed: the number of gaps fixed after each exercise.
Incident response is a capability, not a document. The organisations that handle a bad day well are the ones that practised the boring parts beforehand.
Where this fits in our work
If you want help applying this to your business, start with Managed Technology & Cybersecurity. We publish scope and starting prices on the pricing page. For the next step in this series, read Third-party security risk: 2027, 2028, 2029, 2030.
Sources
- IBM, "Cost of a Data Breach Report 2026", July 2026.
- World Economic Forum, "Global Cybersecurity Outlook 2026", January 2026.
- DataGuidance, "Bangladesh: Personal Data Protection Act 2026", accessed 16 November 2026.
- NIST SP 800-61r3, Incident Response Recommendations and Considerations, 2025.
Responses (0)
No comments yet. Add the first one below.
Leave a response
Your email address will not be published. Required fields are marked *.