A nonprofit does not need to be famous, wealthy or politically controversial to be attacked. Most attacks do not begin with someone choosing your organization by name. They begin with software scanning the internet for an old plugin, an exposed login page, a leaked password, a misconfigured server or an API that answers more than it should.
That distinction matters. “We are too small to be a target” is the wrong risk model. A better question is: if an automated attack finds us tonight, what can it reach, how quickly will we notice, and how cleanly can we recover?
This guide explains the current evidence, the controls that reduce risk most, and the security architecture we use when we build or harden websites, applications and protected workspaces for nonprofits and churches.
What the latest evidence actually says
The strongest reports do not point to one dramatic new trick. They point to several familiar weaknesses being exploited faster and at larger scale.
The 2025 Verizon Data Breach Investigations Report found that exploitation of vulnerabilities reached 20% of known initial-access vectors in the relevant breach set, a 34% increase from the prior report. Ransomware was present in 44% of the breaches Verizon reviewed. In its small and medium business segment, the report found ransomware involved in 88% of breaches.
Google’s Cloud Threat Horizons Report H1 2026, which analyzes activity observed during the second half of 2025, shows how quickly the entry point can change. In that cloud-specific dataset, exploitation of third-party software rose to 44.5% of initial access, ahead of weak or missing credentials at 27.2%. Google also reported that identity issues played a role in 83% of the major cloud and SaaS compromise engagements it examined.
These numbers describe different datasets and should not be combined into one universal probability. Their shared message is still clear:
- unpatched internet-facing software is being exploited quickly;
- stolen credentials and excessive permissions remain central;
- attackers increasingly target trusted integrations, tokens and backups;
- prevention alone is incomplete without detection and tested recovery.
The new OWASP Top 10:2025 reaches a similar conclusion for web applications. Broken access control remains first, followed by security misconfiguration and software supply-chain failures. Authentication, logging, cryptography, injection and insecure design remain on the list because a polished interface does not make the system behind it safe.
Why nonprofits have a distinctive risk
Nonprofits often hold more sensitive data than they realize:
- donor names, gifts, addresses and tax statements;
- children’s registrations, allergies, emergency contacts and authorized pickup adults;
- volunteer background-check status and schedules;
- prayer, pastoral-care or assistance requests;
- payment and fundraising integrations;
- grant, payroll and board documents;
- email lists and social-media publishing access.
At the same time, access is frequently spread across employees, volunteers, board members, contractors and former team members. One person may administer the website, donation platform, email, domain and social accounts from the same inbox. Shared passwords feel convenient until nobody knows who used them, who still has them or which other service received the same password.
The result is not a lack of caring. It is a systems problem: responsibility is distributed, but ownership and visibility are not.
How an attack usually becomes a breach
Attackers rarely need a cinematic “hack.” A practical attack path is much simpler.
| Stage | What happens | What stops or limits it |
|---|---|---|
| Discovery | Automated scanners find a website, admin route, server, plugin or exposed API | Small public surface, patched software, no exposed database or management port |
| Entry | A stolen password, phishing message, vulnerable dependency or bad configuration provides access | Phishing-resistant MFA, secure sign-in, rapid patching, secret management |
| Expansion | The account or server reaches data and other systems | Least privilege, separate roles, network and service boundaries |
| Persistence | New accounts, tokens, scheduled jobs or changed files keep access alive | Account review, deployment integrity, logs and alerts |
| Theft or disruption | Data is exported, pages are changed, email is abused or files are encrypted | Egress visibility, rate limits, isolated backups and a response plan |
| Extortion | The attacker demands payment or threatens disclosure | Tested recovery, legal and insurance contacts, preserved evidence |
Security improves when every stage has a barrier. Relying on one product creates one point of failure.
Build the security program around six functions
The NIST Cybersecurity Framework 2.0 is useful for small organizations because it organizes the work into six continuing functions: Govern, Identify, Protect, Detect, Respond and Recover. NIST’s small-business guide explicitly includes nonprofits, schools and other organizations with limited technical staff.
1. Govern: decide who owns the risk
Name one person who owns the security process, even if technical work is performed by a vendor. Write down:
- who approves access;
- who receives security alerts;
- who can authorize an emergency shutdown;
- who contacts the hosting provider, insurer, counsel and law enforcement;
- which systems must return first after an outage;
- how vendors must notify you about an incident.
Governance does not need a fifty-page policy. It needs names, authority and a review date.
2. Identify: know what you are protecting
Create a current inventory of:
- domains and DNS accounts;
- website hosting and cloud servers;
- databases and file storage;
- mobile apps and app-store accounts;
- email, fundraising, payment and messaging platforms;
- source-code repositories and deployment pipelines;
- administrator, service and API accounts;
- backups and the credentials needed to restore them.
Then classify the data. Public event information does not need the same controls as a child’s medical note or a donor’s tax statement. Collect less sensitive data when the program does not need it, and set a retention period instead of keeping every record forever.
3. Protect: make the common paths difficult
Identity and access
- Give every administrator an individual account; do not share one master login.
- Require phishing-resistant MFA for domain, email, hosting, source code, finance and administrator accounts. Security keys and passkeys are stronger than SMS.
- Give each role only the records and actions it needs.
- Remove access immediately when a staff member, volunteer or contractor leaves.
- Keep service accounts separate from human accounts and rotate exposed tokens.
- Store recovery codes securely and outside the account they recover.
The CISA guidance for civil-society organizations with limited resources prioritizes phishing-resistant MFA, account audits, fast offboarding and least privilege for exactly this reason.
Website and application
- Patch the operating system, framework, CMS, plugins and dependencies on a defined schedule.
- Remove software you no longer use. A disabled plugin can still be vulnerable if its files remain.
- Keep secrets out of source code, browser JavaScript, public repositories and screenshots.
- Validate every input on the server, not only in the browser.
- Parameterize database queries and restrict file uploads by type, size, storage location and access.
- Protect state-changing requests against cross-site request forgery.
- Apply rate limits and abuse controls to sign-in, password reset, contact, donation and registration endpoints.
- Use secure cookie settings, strict transport security, a content security policy and narrow cross-origin rules.
- Return safe error messages to visitors while recording useful technical details in protected logs.
- Review authorization on every request. Hiding a button is not access control.
The OWASP Top 10:2025 is a good minimum checklist for this layer, but a real review also considers the organization’s data, users and likely impact.
Server and cloud
- Do not expose the database, backup console or server-management port to the public internet unless there is a specific protected path.
- Separate production, testing and developer access.
- Encrypt connections and sensitive stored data.
- Restrict cloud permissions, firewall rules and service-to-service credentials.
- Apply security updates quickly when a vulnerability is actively exploited.
- Keep infrastructure configuration under version control and require review for high-impact changes.
4. Detect: make abnormal behavior visible
A control can fail. Detection limits the time an attacker has after that failure.
Record and alert on:
- administrator sign-ins and failed sign-in bursts;
- MFA, password, recovery-email and permission changes;
- new administrator or API accounts;
- unusual exports, downloads or database queries;
- deployments outside the normal process;
- changes to DNS, payment settings or notification recipients;
- repeated form abuse, scanning and rate-limit events;
- backup failures and missed security updates.
Logs need useful context: time, account, action, result and request identifier. Do not place passwords, full payment details or sensitive form contents in logs. Protect the logs from the same account being monitored, and keep them long enough to investigate.
Monitoring does not mean staring at a dashboard all day. It means turning a small set of high-value events into alerts that reach a responsible person.
5. Respond: prepare decisions before the emergency
A one-page incident plan should answer:
- Who decides that an incident is active?
- How do we isolate an affected account, application or server?
- Which credentials and tokens must be rotated?
- Where are logs and evidence preserved?
- Which vendor, insurer, legal adviser or authority must be contacted?
- Who communicates with staff, donors, clients or families?
- What facts must be confirmed before a public statement?
If you suspect an active intrusion, do not begin by deleting files or rebuilding the server. First contain access and preserve evidence. The correct sequence depends on the incident, contractual duties and applicable law.
6. Recover: treat backups as a working system
A backup is useful only when it can be restored.
Use the 3-2-1 idea as a starting point: multiple copies, more than one storage type and one copy isolated from the production environment. Protect backup administration with separate credentials. Keep at least one version that normal production accounts cannot delete.
The CISA ransomware guide recommends offline, encrypted backups and regular restoration tests because ransomware operators often try to find and delete reachable backups.
For each critical system, record:
- what is backed up;
- how often;
- where it is stored;
- who can restore it;
- the last successful restore test;
- how much data and downtime the organization can tolerate.
“Backup completed” and “organization can operate again” are not the same result.
A practical 30-day protection plan
In the first 72 hours
- Turn on MFA for domain, email, hosting, code, finance and administrator accounts.
- Remove former staff, volunteers and unused administrator accounts.
- Change reused passwords and revoke unknown sessions and tokens.
- Confirm that backups are running and that an isolated copy exists.
- Update exposed software with critical security fixes.
- Write down the person to call if something suspicious happens.
In the first seven days
- Complete the asset and administrator inventory.
- Assign roles and remove excessive permissions.
- Review public server ports, storage permissions and API endpoints.
- Add rate limits and spam controls to public forms.
- Configure alerts for administrator and recovery-setting changes.
- Run one restore test in a safe environment.
By day thirty
- Review application access control and the current OWASP risks.
- Document patch timing, access approval and offboarding.
- Separate production from development and personal accounts.
- Review vendors, integrations and the tokens they hold.
- Run a tabletop exercise: “Our main admin account was stolen. What happens during the next four hours?”
- Give leadership a short risk register with owners and target dates.
The goal is not to finish cybersecurity in thirty days. The goal is to replace invisible risk with a process the organization can keep running.
What NONPROFIT.SYSTEMS does
We design security into the projects we build for nonprofits and churches instead of adding it after launch. Depending on the agreed scope, that work can include:
- an inventory of websites, applications, servers, accounts and integrations;
- a threat and data-flow review before development;
- role-based access and separation of sensitive records;
- secure sign-in and MFA through the selected identity provider;
- server-side validation, abuse controls and safer file handling;
- dependency and configuration review;
- protected secrets and deployment controls;
- security headers, encrypted connections and restricted data access;
- activity logging, operational alerts and backup monitoring;
- isolated backups, restore testing and a written recovery procedure;
- an incident-response sheet written for the people who will actually use it;
- periodic security review and patching when included in the support plan.
We can begin with an existing website or system. The first deliverable is a plain-language map: what is exposed, what matters most, what should be fixed first and what can wait.
We do not claim that any website, app or server can be made impossible to breach. No serious provider can. Security is a managed reduction of likelihood, access and damage. The exact controls, monitoring hours, response obligations and third-party services must be written into the project scope.
NONPROFIT.SYSTEMS is based in Boca Raton and works with nonprofits and churches across South Florida and the United States. If you need a website, application or server reviewed, start with the contact form.
Questions leaders ask
Why would anyone attack a small nonprofit? Often nobody chooses it personally. Automated systems scan large numbers of websites and accounts for vulnerable software, leaked credentials and misconfigurations. Criminals can monetize access through fraud, data theft, spam, extortion or ransomware.
What are the three best first controls? Phishing-resistant MFA for critical accounts, rapid patching of internet-facing software, and isolated backups that have been restored successfully in a test.
Does MFA stop phishing? It blocks many account-takeover attempts, but not every one. Attackers also steal session cookies, use convincing approval prompts and call help desks. Passkeys or hardware security keys, careful recovery procedures and alerts are stronger than SMS alone.
Is a cloud-hosted system automatically secure? No. The provider secures its underlying service; the customer still controls identities, permissions, application code, firewall rules, data and third-party integrations.
Can a security service guarantee that we will never be hacked? No. A credible service reduces attack paths, limits access, detects suspicious activity and prepares recovery. Guarantees of perfect protection should be treated as a warning sign.
Sources and scope
Sources were reviewed on November 14, 2026. Statistics describe the populations and periods in their respective reports; they are evidence of patterns, not a prediction that any one organization has the same probability.
- NIST Cybersecurity Framework 2.0 Small Business Quick-Start Guide
- CISA joint guidance for civil-society organizations with limited resources
- CISA phishing guidance: stopping the attack cycle at phase one
- CISA and MS-ISAC Ransomware Guide
- OWASP Top 10:2025
- Verizon 2025 Data Breach Investigations Report
- Google Cloud Threat Horizons Report H1 2026
A practical first step
Free church website audit
Share your website. We will review the visitor experience, mobile layout and the next steps for your church or nonprofit.



