Security review

Security questions, answered

The questions a security or IT review tends to ask, each answered with how OKRs Tool actually works — written so your team can clear the review or drop the answers straight into your own questionnaire.

Last updated: 18 November 2026
EU-hosted — AWS IrelandGDPR with DPAAES-256 & TLS 1.2/1.3SSO & SCIM
EU-onlydata residency, Ireland
SOC 2 · ISO 27001audited infrastructure
72 hoursbreach notification
99.9%uptime SLA on Enterprise

No questions match that. Try a broader term, or send it to Steven and he'll answer in writing.

01
Data hosting & residency

Where is our data hosted and processed?

The database and authentication run on Supabase, hosted in the EU (Ireland) on AWS, with Cloudflare providing the CDN, web application firewall and DDoS mitigation at the edge.

Is customer data stored in the EU?

Yes. Core OKR content is stored in the EU (Ireland). A short list of sub-processors handles specific functions such as billing, email and AI, and where any of them processes limited data outside the EU it is covered by Standard Contractual Clauses.

Can we run in a dedicated, single-tenant environment?

Yes. On top of the Enterprise plan, a dedicated single-tenant EU instance gives you your own database, credentials and subdomain, isolated from every other customer. See the dedicated instance →

Who are your sub-processors?

We keep a current, public list of every sub-processor, what it does and where it processes data, at okrstool.com/sub-processors. A Data Processing Agreement covering them is available on request.

02
Encryption

How is data encrypted in transit?

All connections use TLS 1.2 or 1.3. Older protocols are disabled, and we do not accept unencrypted HTTP for any data exchange.

How is data encrypted at rest?

Data in our databases and backups is encrypted with AES-256.

03
Authentication & access

Do you support single sign-on?

Yes. SAML and Google Workspace SSO are available on the Enterprise plan. See SSO and SCIM setup →

Do you support automated provisioning and deprovisioning?

Yes. On Enterprise, SCIM provisioning and deprovisioning keeps accounts in sync with your identity provider, so access is removed automatically when someone leaves.

Do you offer two-factor authentication?

Yes, for every user on every plan, and Enterprise admins can require it org-wide. Two-factor is enforced in the database itself — until a user enters their code, no customer data can be read or changed, even outside the app.

What roles and permission levels exist?

Four roles — Admin, Manager, Member and Viewer — each with different permissions. Fine-grained privacy rules govern who can view or edit OKRs at the workspace level, and sensitive workspaces can be restricted to specific roles.

How is our organisation isolated from other customers?

Tenant isolation is enforced at the database layer through row-level security, so one organisation's data is never reachable from another. Where your policy requires physical separation, the dedicated single-tenant instance provides it.

Can we enforce session timeouts?

Yes. Admins set an idle timeout of 15, 30, 60 or 120 minutes and an absolute timeout of 8, 12 or 24 hours, enforced in-app. They apply to SSO users too, or can be switched off to keep sessions open until manual sign-out.

How do users sign in?

Sign-in is passwordless — an email magic link or Google — and rate-limited per IP address and email address. No passwords are stored.

04
Application & integrations

How are API keys secured?

API keys are locked to a single organisation, with permissions checked on every request, and secret database keys never reach the browser.

Are outbound webhooks signed?

Yes. Outbound webhooks are signed with HMAC-SHA256, so your systems can verify every payload came from us and arrived unaltered.

Do you run vulnerability scanning?

We run automated application and dependency scanning, with findings triaged and remediated by severity. Our infrastructure providers maintain their own independent testing under SOC 2 Type II and ISO 27001 audits.

05
Monitoring & audit

Do you keep an audit log?

Yes. Changes to OKRs, key results, check-ins, initiatives and KPIs, along with member and role changes, MFA events and sign-ins, are recorded with who and when. The log exports to CSV and is available to admins on the Enterprise plan.

How is the platform monitored?

Infrastructure monitoring runs 24/7 with automated alerting for anomalies, and performance and error monitoring run through Datadog.

06
Backups & availability

How often do you back up data?

Database backups run daily and are retained for about 14 days, then rotated out. Backups are encrypted with AES-256.

What is your uptime commitment?

A 99.9% uptime SLA on the Enterprise plan.

07
Incident response

Will you notify us of a security incident?

Yes. We detect and contain through 24/7 monitoring, assess scope within hours, and notify affected customers within 72 hours where legally required, or sooner where practicable, followed by a post-incident report.

Can we review your incident response plan?

Yes, under NDA on request, along with our Information Security Policy and related procedures.

08
Compliance & certifications

Are you SOC 2 or ISO 27001 certified?

Our infrastructure providers hold SOC 2 Type II and ISO 27001, and we can share their reports under NDA. OKRs Tool does not yet hold its own certification at the application layer, and we would rather tell you that plainly than imply otherwise.

Are you GDPR compliant?

Yes. Data is hosted in the EU, and a signed Data Processing Agreement is available on request.

Will you complete our security questionnaire?

Yes. Send us your template and we will complete it — SIG Lite, CAIQ or your own format.

Who at your company can access our data?

Access to customer data is restricted to the founder and our development platform, and nobody else.

09
AI & data handling

Do you use our data to train AI models?

No. Data sent to power AI features is not used to train models and is not shared across accounts. Only the data needed for a given request is sent, and the result is returned to your workspace.

Can we turn AI off entirely?

Yes. AI can be disabled at the account level, which stops the model calls — not just hidden in the interface.

How do we export or delete our data?

Your data exports at any time via JSON, Excel, audit-log CSV and the REST API. When an organisation closes its account, its data is deleted on request as a hard delete, not a flag.

Deeper detail

Need the full write-up?

This page answers the review. The security overview explains the posture behind each answer, and the trust center covers where your data lives and how it is handled.

Still have a question on the list?

Send your questionnaire or your open items and we will answer in writing — SIG Lite, CAIQ or your own template. No call required.