Lorixen.
EN

A security page for your software product: what buyers check

Short answer: A good security page tells a business buyer, in plain language, where their data lives, how it is protected, who can reach it, how you recover from failure and whom to contact when something looks wrong. Write it as a set of short, checkable statements, link a status page and a security.txt file, and never claim more than you can show. Done well, it shortens the security questionnaire that arrives with almost every serious deal.

For a small Swiss software company selling to firms across Europe, the first security conversation usually happens before anyone talks to you. Someone in IT or procurement opens your site, looks for a page called "Security" or "Trust", and decides whether you are worth a questionnaire at all. If that page is missing, vague or full of padlock icons and slogans, you have made their job harder.

Who reads it, and what they are trying to do

Three people tend to land on this page:

  • The champion who wants your product and needs something to forward internally.
  • The IT or security reviewer who will compare your answers with a checklist.
  • The occasional researcher who has found a bug and wants to report it.

Each needs different depth, so the page works best as a short summary at the top, detailed sections below, and a clear contact route at the bottom.

The blueprint, section by section

1. One-paragraph summary

Two or three sentences: what kind of data the product holds, where it is hosted, and a link to the contact for security questions. This is the paragraph the champion pastes into an email.

2. Hosting and data location

State the cloud or hosting provider and the regions your production data sits in, plus where backups are stored if that differs. Explain the split of duties honestly. AWS describes its shared responsibility model as security "of" the cloud (the provider protects the hardware, software, networking and facilities) versus security "in" the cloud (the customer's configuration of the services they choose). Your page should say what you rely on the provider for and what you configure yourself. Buyers notice when a small vendor implies the hyperscaler's controls are its own.

3. Encryption

Say what is encrypted in transit (for example, "all traffic uses HTTPS") and at rest (databases, file storage, backups), and who manages the keys. Keep it factual; "bank-grade" tells a reviewer nothing.

4. Access controls inside your company

  • Who on your team can reach production data, and how that access is granted and removed.
  • Whether staff use multi-factor authentication on admin tools.
  • Whether access is logged and how long logs are kept.

A two-person company can answer this as well as a large one. "Two engineers have production access, both using hardware security keys" is a strong, specific answer.

5. Customer-facing security features

List what your customers can switch on: single sign-on, two-factor login, role-based permissions, audit logs, IP restrictions, session timeouts. Mark which plans include which, if it varies. This section often decides deals, so keep it current with each release.

6. Backups and recovery

How often backups run, how long they are kept, where they are stored, and whether restores are tested. If you have a target for how quickly you can restore service, state it as a target, not a guarantee.

7. Availability and status

Link a public status page and say how incidents are communicated (status page, email to admins, in-app banner). A status page that shows past incidents honestly builds more trust than one that is always green.

8. Incident and vulnerability contact

Give a dedicated address such as security@yourdomain, a response expectation you can actually meet, and a link to your disclosure policy. Then publish the same information in machine-readable form (next section).

9. Certifications and documents

If you hold a certification or an independent audit report, name it neutrally and say how a customer can request the report. If you don't, say what documents you can share instead (an architecture overview, a completed standard questionnaire, a sub-processor list). Don't put badges for audits that are "in progress" next to ones that are complete.

Add a security.txt file

RFC 9116 defines security.txt, a small text file that tells researchers how to report vulnerabilities. The practical points from the standard:

  • Serve it at /.well-known/security.txt, over HTTPS (the RFC requires HTTPS for the file and the links in it).
  • A Contact field (an email via mailto:, a phone number or a web page) and an Expires field are required; the RFC recommends the expiry date be less than a year ahead so the file doesn't go stale.
  • Optional fields include Policy (link to your disclosure policy), Encryption (a link to your key, never the key itself), Canonical, Preferred-Languages and Hiring.
  • Signing the file with OpenPGP is recommended, with a Canonical field so the signature confirms the file's location.

The securitytxt.org project offers a generator for the file. Put a calendar reminder before the Expires date; an expired file is a small but visible sign of neglect.

A minimal example:

Contact: mailto:security@example.ch
Expires: 2027-10-31T23:00:00.000Z
Policy: https://example.ch/security/disclosure
Preferred-Languages: en, de

Using the page to answer questionnaires faster

Most buyers will still send a spreadsheet of questions. Keep an internal answer bank that matches the sections above, so every answer on the page and in a questionnaire comes from the same source. When a new question appears twice, consider adding the answer to the public page. Over time, many reviewers will accept "see section 5 of our security page" for routine items.

What not to write

  • Absolute claims: "unhackable", "100% secure", "zero risk".
  • Features on the roadmap described as live.
  • Long lists of acronyms with no explanation.
  • Legal language copied from elsewhere. Contract terms belong in your agreements; have them reviewed where you operate.

FAQ

Does a small startup really need a security page?

Yes, if you sell to businesses. A short, honest page answers the first round of questions and shows that someone owns security, even when that someone is a co-founder.

Should the page be called "Security" or "Trust Center"?

Either works. "Security" is what most reviewers search for; a trust center usually adds documents, status and policies under one hub. Link it from your footer and pricing page.

Can we share security documents only after an NDA?

Many vendors keep a public summary and share detailed reports on request. Say clearly on the page how to ask and how long it typically takes you to respond.

How often should we update it?

Whenever infrastructure, sub-processors or customer-facing security features change, and at least when you renew the security.txt expiry date.

When your product is ready for its first procurement reviews, we can help you plan the security page and the features behind it.

Sources

WhatsAppLet’s talk