SECURITY & YOUR DATA

Your data is yours. Here is how we look after it.

In short: your records are kept apart from everyone else's, encrypted on disk, behind a sign-in you can protect with a second step, and never sold. This page says what we do about the things people worry about, where you can see it in the product, and what we do not claim.

Last updated: 22 October 2026

A database for each account

Your records are never mixed in with another customer's.

Encrypted on disk

AES-256 on every database, backup and uploaded file.

Two-step sign-in

Add an authenticator app to your password.

Card numbers never seen

Payments are taken by Stripe, not by us.

Never sold

No advertising, no analytics trackers.

Backed up every day

Kept for up to a year, and checked.

Who can see my data?

The first worry: is what I enter private, from other customers and from the people who run the service?

Can another CostBridge customer see my data?

No. Every account's records are kept in a database file of its own, and the service opens only the one that belongs to the account you are signed in to.

How this works
  • There is no shared table with a "which account is this?" column that a mistake could forget to filter on: another account's records are in a different file.
  • If a request ever reached the data without a known account, the service stops with an error rather than showing an empty or someone else's database.
  • Our automated tests include checks that one account cannot read another's ingredients, recipes, menus, suppliers, reports or photos. They run on every change.
  • A trial that has ended, or a cancelled subscription, is locked by the server on every request, not just hidden in the screens.

Who at CostBridge can see my data?

Only the few people who run the service, through an administrator area that is separate from the customer app and needs more than a password.

How this works
  • Administrators sign in with their password and a one-time code sent to their email. In production the administrator area also needs an authenticator-app code, and stays locked without it.
  • The codes are stored only as hashes, expire quickly and allow only a few tries.
  • Important administrator actions are written to an audit log.
  • We do not sell or hand over your records to anyone. The few companies that help us run the service are listed in the Privacy Policy.

See it:Who we share data with: Privacy Policy

How do I control who in my business or household sees what?

You decide who joins your account and what role they have. Taking someone out ends their access straight away.

How this works
  • The owner invites people and gives each a role; some pages, such as the budget and settings, are for owners only.
  • Removing a person, or switching them off, ends every sign-in they have open on the next request - nothing waits for a token to run out.
  • Invitations are single-use links that expire.

See it:Settings > Team

I keep my tenants' details in it. Is that safe?

It is kept in your account's own database, with the same protection as everything else. You decide what to enter; we handle it only to run the service for you.

How this works
  • For the records you enter about other people you are the "controller" and CostBridge is your "processor" under the UK and EU GDPR: we use it only to provide the service, on your instructions.

See it:Roles in the Privacy Policy

What if someone gets in, or steals something?

No system is perfectly safe, so the useful question is: what stands in the way, and what is left if something does go wrong?

What if the server's disk, a backup or a copy of the data is stolen?

It would be unreadable. Every database, every backup and every uploaded file is encrypted with AES-256, with a different random key for each account.

How this works
  • Databases (yours and ours) are encrypted whole with SQLCipher. Uploaded photos, receipts, voucher and property documents, kept statement PDFs and support attachments are each sealed with AES-256-GCM. Backups are copies of the encrypted database.
  • Each account's key is random and is itself stored locked by a master key that is kept apart from the data, so neither alone opens anything. The master key can be replaced without touching your data.
  • When an account is deleted its key is destroyed, which makes its database, its files and every backup of it unreadable for good.
  • What this does not do, and the limits we are open about, are further down this page.

Could someone guess or steal my password?

Guessing is cut off, and your password is never stored in a form that can be read - not even by us.

How this works
  • Passwords are kept only as slow one-way BCrypt hashes. They are never written to logs or sent back to you.
  • Five wrong passwords in a row lock that email address for 15 minutes, even for the right password, and there are limits per visitor on every sign-in, sign-up and reset route.
  • Very common passwords, and passwords that contain your email address, are refused when you choose one.
  • Sign-up codes, password-reset links and invitations are made from a cryptographic random source, expire, work once, and are stored only as hashes, so a copy of the database cannot be used to take over an account.
  • Prefer not to have a password at all? Sign in with Google or Apple: we never see or store one.

Is a password enough?

It does not have to be. Anyone can add a second step: a six-digit code from an authenticator app on their phone.

How this works
  • Works with Google Authenticator, Microsoft Authenticator, 1Password and other apps that follow the standard.
  • You also get ten single-use recovery codes, shown once, in case you lose your phone. A code from the app works only once, so one seen over a shoulder cannot be reused.
  • Wrong codes count against the same lock as wrong passwords.
  • Turning it off needs your password and a current code, and we email you when it is turned on, off or its recovery codes are renewed.

See it:Settings > My account > Two-step sign-in

What if I leave myself signed in on a lost or shared device?

You can see every device that is signed in, end any of them, or end all but the one in your hand.

How this works
  • A list shows each device ("Chrome on Windows"), when it was last used, and a Sign out button. "Sign out everywhere else" ends all the others at once. Signing out ends the session on the server, not just in the browser.
  • If someone signs in from a device you have not used before, we email you, so a sign-in that was not you is noticed.
  • Changing or resetting your password signs your other devices out.
  • The browser keeps only a short-lived key in memory. The long-lived part is in a protected cookie that no script on the page can read, and it is replaced every time it is used: a copied one stops working and the sign-in it belonged to is ended.

See it:Settings > My account > Where you are signed in

What about attacks on the website itself?

We close off the common ones and check the rest with automated tests.

How this works
  • Nothing you type is ever pasted into a database command, so the classic "SQL injection" has nowhere to go.
  • An uploaded picture is checked by what is in it, not by its file name, and is kept privately: it is only handed back to the account it belongs to.
  • The site is served over HTTPS only and tells browsers to keep it that way; the site cannot be put inside another website's frame, and browsers are told not to guess file types.
  • The public pages, support and sign-in routes are rate-limited.
  • Our payment notifications are accepted only with a valid signature from Stripe.

What about my money and my bank?

Two different worries: paying for CostBridge, and the bank or card statements that households may choose to import.

Do you see or store my card number?

No. Payments are taken by Stripe, a payment company that is built for it. Your card number never reaches CostBridge.

How this works
  • We keep only your plan, its status and dates, and Stripe's reference numbers for you and your subscription.
  • You manage or cancel the subscription from Settings, and Stripe sends the receipts.

See it:Settings > Billing

Are my bank statements safe?

Importing a statement is optional, and CostBridge can only read the file you choose: it has no connection to your bank and cannot move money.

How this works
  • The file is read by our own servers and not sent to another company. No AI service reads your statements.
  • A PDF is read only after you tick "I agree"; the file is thrown away once it has been read unless you ask us to keep a copy. A CSV is never kept.
  • Nothing reaches your budget until you have looked at the suggestions and pressed Import, and an import can be taken back.
  • The full explanation, with the table of what is kept and for how long and the steps to delete, is on its own page.

See it:Your data: how statements are handled

What if I lose my data, or want to leave?

Trust includes being able to get your records back, take them with you, and have them deleted.

What if my data is lost or damaged?

It is backed up every day, and the backups are kept for up to a year.

How this works
  • Daily backups are kept for 30 days, weekly ones for 12 weeks and monthly ones for 12 months.
  • Each backup is checked for integrity when it is made, and recorded with a checksum.
  • A backup is a copy of your encrypted database, so it is protected in the same way as the original.
  • A restore keeps a safety copy of the current data first, so a restore itself can be undone.

Can I take my data out, or have it deleted?

Yes to both. Your reports and lists download as CSV, Excel and PDF, we can send you a full copy on request, and you can ask for your account to be deleted.

How this works
  • If you cancel, your data is kept for 30 days in case you change your mind, and then your account and its database are deleted, or sooner if you ask. Deleting the database also destroys its key, so the backups of it cannot be read either.
  • You can delete a statement import yourself, with its lines and any PDF copy, and make CostBridge forget what it learned from your choices.
  • Your rights under the UK and EU GDPR - to see, correct, export, restrict, object and erase - are set out in the Privacy Policy.

See it:Your rights: Privacy Policy

Do you sell my data, track me or feed it to AI?

We do not sell it, we do not use it for advertising, and we use no analytics or advertising trackers.

How this works
  • The only outside request a page makes is for the website's typeface, from Google Fonts, which is listed in the Privacy Policy.
  • The "Ask AI" button on the free public recipe pages sends the public recipe to an AI service to write a method. Nothing from any account is ever sent, and there is no AI reading of your records.
  • There is no cookie banner because there is nothing to consent to: the browser storage we use is only to keep you signed in and remember choices like the theme.

See it:Cookies and storage: Privacy Policy

What if something goes wrong?

How we find out, how we tell you, and how you can tell us.

What happens if there is a breach?

If a breach puts people at risk we tell the people affected without undue delay, and the Information Commissioner's Office where the law requires it, within 72 hours where we can.

How this works
  • Important actions on an account - sign-ins from new devices, two-step changes, locked sign-ins, password changes, administrator changes - are written to an audit log, which is what lets us work out what happened.
  • If your data is in an account run by someone else, we tell them without undue delay so they can tell you.

See it:How we protect it: Privacy Policy

I think I have found a security problem.

Please tell us. Email support@costbridge.co.uk with "Security" in the subject, and say what you saw and where. We treat these as urgent.

How this works
  • Please do not read or change other people's data to prove a point, and give us a chance to fix it before you share it.
  • The same address is in our security.txt file, where tools and researchers look for it.

See it:security.txt

Industry standards we use

We use open, well-known standards rather than inventing our own. Each one says whether we use it directly, rely on a company that holds it, or only follow its guidance - and the ones we do not hold are listed too.

In use

AES-256 encryption

The encryption standard used by governments and banks, with 256-bit keys.

At CostBridge: Every database (SQLCipher) and backup is encrypted with it, and every uploaded file and stored secret is sealed with AES-256-GCM, which also detects tampering.

In use

HTTPS (TLS) and HSTS

The standard that encrypts everything between your browser and a website, and the header that tells browsers never to fall back to an unencrypted connection.

At CostBridge: The whole site and the service behind it are served over HTTPS, plain-http addresses are sent to HTTPS, and browsers are told to keep to it.

In use

bcrypt password hashing

A deliberately slow one-way function made for passwords, so a stolen list of them is very costly to crack.

At CostBridge: Passwords are stored only as bcrypt hashes with a per-password salt; they cannot be read back, even by us.

In use

TOTP two-step codes (RFC 6238)

The open standard behind the six-digit codes in Google Authenticator, Microsoft Authenticator, 1Password and similar apps.

At CostBridge: Any authenticator app that follows the standard works with CostBridge two-step sign-in. Each code is accepted once.

In use

OAuth 2.0 and OpenID Connect

The standards behind "Sign in with Google" and "Sign in with Apple".

At CostBridge: We check the signed identity token each provider gives us before anyone is signed in. Those accounts have no CostBridge password to steal.

In use

Signed tokens (JWT) with server-side sessions

A signed, short-lived pass your browser shows with each request.

At CostBridge: The pass lasts minutes, is kept in memory only, and is checked against a session on our server on every request, so ending a session takes effect at once.

Through Stripe

PCI DSS (card payments)

The security standard for anyone who handles payment card numbers.

At CostBridge: Card numbers are entered with Stripe, which is certified to the highest level of PCI DSS (Level 1). They never reach CostBridge, and Stripe's messages to us are accepted only with a valid signature.

Built to its guidance

OWASP Top 10

The industry list of the most serious risks to web applications: broken access control, weak authentication, injection, misconfiguration and so on.

At CostBridge: Our internal security review was organised around it. Each of the protections on this page answers one of its categories.

Built to its guidance

UK GDPR, EU GDPR and the ICO

The data protection laws for people in the UK and the European Union, enforced in the UK by the Information Commissioner's Office.

At CostBridge: Our Privacy Policy is written for them: who is responsible for what, why data is used, how long it is kept, your rights, and telling you about a breach without undue delay.

In use

security.txt (RFC 9116)

A standard place on a website where security researchers find out how to report a problem.

At CostBridge: Ours is at /.well-known/security.txt and points to the same contact as this page.

Not yet

Cyber Essentials, ISO/IEC 27001, SOC 2 and an independent penetration test

The independent certifications and tests that a company can pass to prove its security to others.

At CostBridge: We do not hold them yet, and we will not display a badge for one until we do. Until then this page tells you what we do so you can judge it yourself.

Your part

Security is shared. These few habits close the gaps that no service can close for you.

  • Turn on two-step sign-in, especially for the account owner.
  • Use a long password you do not use anywhere else, or sign in with Google or Apple.
  • Give each person the lowest role that lets them do their work, and remove people who leave.
  • Look at "Where you are signed in" now and then, and sign out devices you do not recognise.
  • Keep your own copy of important reports: they download as CSV, Excel or PDF.
  • Tell us quickly if you think someone else has got into your account.

What we do not claim

A security page that only says good things is not worth much. These are the limits.

  • CostBridge has not yet been independently certified (for example Cyber Essentials or ISO 27001) and we have no published penetration test to show you. Until we do, we describe what we do instead of showing a badge, so you can judge it yourself.
  • Encryption on disk protects against stolen or copied files, backups and disks. It does not protect against someone who takes over the running service, because the service has to be able to open your data. Sign-in protection, testing and monitoring are what stand in the way of that. Our current hosting has no hardware key vault, so the master key is held in the service's protected configuration, apart from the data and the backups.
  • No system is perfectly secure, and we will not tell you ours is. What we can promise is to tell you plainly if something goes wrong.
  • We do not use words such as "bank-grade" or "military-grade", because they mean nothing you can check.

Who helps us run it

A small number of companies handle data for us, each for one purpose and under a contract. The full list, with the legal detail, is in the Privacy Policy.

CompanyWhat for
SmarterASP.NETHosting: the servers and storage
StripeCard payments and subscriptions
An email delivery providerSending the emails the service needs
Google and AppleOnly if you choose to sign in with them
Still have a question?

Ask us anything about how your data is handled. A real person will answer.

Loading CostBridge… This can take a little longer on a slow connection. Please wait a moment.