Getting started.
From signing up to processing your first live UK card payment. KYB-first onboarding, sandbox from minute one, and an honest path to production.
Help
Help is organised by topic across the site, not collected into one mega-FAQ. This page routes you to the canonical place for each question: developer reference, pricing detail, security and compliance documentation, the glossary, complaints procedure, or the live status feed. Eight categories, forty-eight topics, every link points to real Fluxa documentation.
From signing up to processing your first live UK card payment. KYB-first onboarding, sandbox from minute one, and an honest path to production.
UK card acceptance with Visa and Mastercard at launch. 3D Secure 2 on every CNP charge by default, Apple Pay and Google Pay where the device supports them.
Where the money goes, when it arrives, and how to reconcile it. Settlement is direct from the payments partner to your bank; Fluxa never holds customer funds.
The legitimate path to reversing a transaction, and the unwelcome one. Refunds you control; chargebacks you contest. Scheme rules apply to both.
Account hygiene that takes a minute to set up and saves an outage when something goes wrong. Two-factor on every account, role-based access by default.
REST API, signed webhooks, hosted checkout in PCI SAQ-A scope, a six-state payment lifecycle, and a Node.js SDK. Documentation lives at /developers and docs.fluxapay.co.uk.
UK card processing under the FCA-authorised payments partner, with Fluxa operating as a Referrer under PSRs 2017. PCI scope kept to SAQ-A; UK GDPR applied to every byte of personal data.
A live status page that is honest about partial outages, not a green-tick wall. Real-time uptime, every incident, and what we did about it.
The Fluxa team reads every help request directly. No outsourced ticket centre, no triage layer, no scripted replies. Most queries get a human reply within one working hour during UK business hours; complex ones within one working day. Urgent payments issues (checkout down, settlement missing) take priority.
Sign up with a UK business email at app.fluxapay.co.uk. You receive sandbox keys immediately and can integrate against test mode while KYB verification runs in parallel.
KYB verification needs: company number (we auto-pull from Companies House), beneficial owner list, business bank account (for settlement), and proof of address for the primary contact. UK businesses only at launch.
Typical verification time is one to three working days. Live keys are issued once KYB completes; sandbox keys keep working throughout.
Read the full documentation: Get started on /pricing
Every Fluxa account gets sandbox keys from minute one. The sandbox mirrors production exactly: same API surface, same webhook events, same payment-state machine, same dashboard. Only the money is fake.
Test cards simulate every scheme response: successful authorisation, soft decline (retryable), hard decline (lost or stolen card), 3DS challenge required, 3DS frictionless, AVS mismatch, and chargeback flow.
Sandbox data is wiped nightly at 02:00 UK time so a clean state is always close by. Live and sandbox use different key prefixes (sk_test_ vs sk_live_) so accidental cross-environment requests fail loudly.
Read the full documentation: Sandbox detail on /developers
Three integration paths, pick whichever fits your stack: hosted checkout (one HTTP call, we render the form, you handle the redirect), embedded SDK (drop-in component in your page, no PCI scope expansion), or direct API (full control, server-to-server, hosted checkout iframe for the card field).
The quick-start on /developers#quick-start walks all three with copy-paste Node.js, Python and curl. Time from sandbox key to working payment is typically under five minutes.
Read the full documentation: Quick start on /developers
Going live means swapping sk_test_ for sk_live_ and updating your webhook endpoint. That is it. Same API surface, same payload schemas, same event types.
Before flipping, make sure: KYB verification is complete (status visible in dashboard), production webhook endpoint validates HMAC signatures, you have tested all three failure paths (decline, 3DS challenge, chargeback) in sandbox, and you have a recovery procedure if your endpoint goes down.
Going live does not require code review or a sales call. Click the toggle in your dashboard.
Read the full documentation: Production checklist on /developers
The standard test card is 4111 1111 1111 1111 (Visa) with any future expiry, any CVV and any postcode. It succeeds with no 3DS challenge by default.
To simulate specific outcomes, use different last-four digits: ...0002 for hard decline, ...0010 for insufficient funds (soft decline, retryable), ...3155 for 3DS frictionless approval, ...3220 for 3DS challenge required, ...0259 for chargeback after settlement.
The complete table covering Mastercard, AVS responses, currency-conversion fees and dispute reason codes is on the developer reference.
Read the full documentation: Full test card list on /developers
Sign up is instant, sandbox keys are immediate. KYB verification typically completes within one to three working days for straightforward UK limited companies and sole traders.
Documents needed: business registration (Companies House number for limited companies, HMRC UTR for sole traders), beneficial owner list with passport or driving licence, UK business bank account for settlement, and proof of address for the primary contact.
Higher-risk verticals (financial services, gambling, regulated industries) take longer because they require additional checks. We tell you at signup if your MCC falls into one of those categories.
Read the full documentation: Onboarding detail on /pricing
3D Secure 2 is mandatory on most UK card-not-present payments under the Payment Services Regulations 2017 implementation of Strong Customer Authentication.
Fluxa runs 3DS 2 on every payment by default and handles the challenge flow inline: frictionless authentications complete with no visible step; challenges show the issuer’s page (biometric, OTP or password). Liability for fraud chargebacks shifts to the issuer on authenticated payments.
Exemptions are available for low-risk transactions (TRA, under £30, trusted beneficiary) and Fluxa applies them automatically where allowed. You do not configure 3DS; you configure the business rules around exemption appetite.
Read the full documentation: 3DS detail on /developers
Apple Pay and Google Pay work out of the box on hosted checkout and embedded SDK. The wallet shows when the visitor’s device supports it; otherwise the card form is presented.
Same headline rate as a direct card payment. No premium for wallet payments. The wallet wraps a Visa or Mastercard credential with a device-bound token plus a cryptogram, and biometric in-device authentication counts as SCA, so the customer never sees a 3DS challenge.
To enable on your own domain, the domain has to be verified with Apple Pay. Fluxa walks you through that in a single dashboard step.
Read the full documentation: Digital wallet detail on /glossary
Subscription billing engine, customer self-serve cancellation, automated dunning on failed charges, and SCA-compliant card-on-file storage for repeat charges.
Recurring charges use the MIT (merchant-initiated transaction) flag with the scheme, exempt from SCA on the recurring leg as long as the initial transaction was authenticated. Soft declines retry on a configurable schedule (default: day 1, day 3, day 7) before the subscription pauses.
Failed recurring payments trigger an email to the cardholder asking them to update their card; the dunning template is editable per merchant.
Read the full documentation: Recurring billing detail on /glossary
Mail Order / Telephone Order payments are supported but flagged differently to the scheme. MOTO transactions do not benefit from a 3DS liability shift, so the merchant carries the full chargeback risk on fraud disputes.
To process a MOTO payment in the dashboard, open the “New payment” flow and tick the MOTO box; the API takes a payment_method.type: moto flag. Card details are entered by the merchant, not the cardholder, and there is no 3DS step.
MOTO is permitted only for specific MCCs and is subject to enhanced risk monitoring. Not enabled by default; ask support to flip it on for your account.
Read the full documentation: MOTO detail on /glossary
Two categories. Soft declines (insufficient funds, daily limit, issuer outage, SCA required) are recoverable on retry. Hard declines (card reported lost or stolen, blocked by issuer, closed account) are final and should not be retried.
Every failed payment returns a structured failure_code and a retryable boolean. Your dunning logic should retry only when retryable: true. Recurring billing retries default to three attempts on soft declines; hard declines pause the subscription immediately.
Generic decline messages (no specific code) are returned for privacy reasons; the dashboard shows the full scheme response.
Read the full documentation: Soft decline detail on /glossary
Storing a cardholder credential after an initial cardholder-authenticated transaction so you can charge it again later. Every subsequent charge carries a stored-credential indicator and a CIT (customer-initiated) or MIT (merchant-initiated) flag.
Required for recurring billing, click-to-buy, and any post-purchase upgrade. Fluxa vaults the credential in the payments partner’s PCI Level 1 environment; you receive a token, never card data.
Cards stored before SCA (pre-September 2019) need to be re-authenticated by the cardholder before the next MIT charge. Fluxa handles that re-auth flow automatically when needed.
Read the full documentation: Card-on-file detail on /glossary
Standard settlement uses BACS three-working-day cycle: capture today, settle to your bank three working days later. Faster Payments accelerated settlement is available for an additional fee, typically settling the next working day.
Settlement happens daily for all captured transactions from the previous business day. Weekends and UK bank holidays push settlement out by one working day each.
Refunds within the settlement window net against the outgoing batch; refunds after settlement create a debit on the next batch.
Read the full documentation: Settlement detail on /pricing
Format is STL-YYYY-MM-DD-FX-NNNNN. YYYY-MM-DD is the settlement date, FX is the payments partner partner code, and NNNNN is the per-day counter.
The reference shows on your bank statement next to the credit line so you can match it to a specific settlement batch in the Fluxa dashboard with one grep. Each batch report shows every payment included, every refund netted, every fee deducted, and the net amount that arrived.
If a settlement batch is missing from your bank statement, search by the date portion of the reference; banks occasionally truncate the full string.
Read the full documentation: Settlement reference detail on /glossary
Flat-rate pricing, no per-transaction fees layered on top, no monthly subscription, no setup fee. 1.8% flat for UK businesses.
Monthly statements are generated on the first of each month and emailed to billing contacts. Each statement lists every settlement batch, every dispute fee, and any scheme fees that passed through. Downloadable as PDF and CSV.
Operational fees (chargebacks, refund scheme fees, dispute arbitration) are itemised separately so you can see exactly what each fee is for. No hidden fees, no surprise charges.
Read the full documentation: Rates detail on /pricing
Every settlement batch generates a CSV settlement report you can download from the dashboard or pull from the API. The report lists every payment in the batch with its ID, amount, fee, and net contribution.
The ledger state model exposes the full lifecycle: AUTHORISED, CAPTURED, SETTLING, SETTLED. A payment is on your dashboard but not yet in a settlement report means it has not settled yet; filter by SETTLED to see what is reconciled.
Discrepancies between dashboard total and bank credit are rare and almost always explained by refunds posted after settlement. Email support@fluxapay.co.uk with the settlement reference if anything looks off.
Read the full documentation: Reconciliation example on /pricing
If a settlement does not arrive on the expected day, the usual cause is a bank holiday, the BACS cut-off (16:30 UK on the working day before), or a name-on-account mismatch flagged by your bank.
Check the dashboard settlement page first: it will show status SETTLING (still in transit), SETTLED (payments partner has released it; your bank may be holding it for review), or FAILED with a reason. If it shows SETTLED but is not in your bank account by end of next working day, contact your bank with the settlement reference.
For payments partner-side delays (rare), check /live for any active incidents. Otherwise email support with the settlement reference.
Read the full documentation: System status on /live
Updating the settlement bank account requires re-verification because it is the highest-risk change on an account. The process: submit new bank details in the dashboard, upload a recent bank statement or void cheque showing the account in the business name, wait for confirmation (typically one working day).
New settlements route to the new account from the confirmation timestamp. In-flight settlements already initiated to the old account continue to that account.
The account name must match your registered business name on Companies House. Personal accounts are not accepted for limited company settlements.
Read the full documentation: Account changes on /security
Issue full or partial refunds from the dashboard or via the API. Refunds carry their own scheme fee (typically 10-15p depending on card type) but no platform fee from Fluxa.
Refunds appear on the cardholder’s statement within three to ten working days depending on the issuer. The credit references the original transaction ID and the original merchant descriptor.
Refunds within the settlement window net against the outgoing batch. Refunds after settlement create a debit on the next settlement; if there is insufficient outgoing volume, the difference is direct-debited from your operating account.
Read the full documentation: Refund detail on /glossary
A chargeback is a forced reversal of a transaction initiated by the cardholder via their issuing bank. You get notified within hours of the issuer raising it, with the reason code, the amount, and the response window.
Scheme reason codes group into: fraud (no liability shift on non-3DS payments; protected on authenticated 3DS payments), authorisation (issuer claims auth was not granted), processing errors (technical), and consumer disputes (goods not received, not as described).
Chargebacks carry a fee (typically £15-25) win or lose. The dashboard shows your chargeback ratio and a warning when you are approaching scheme monitoring thresholds (1% by count or 1% by value for Visa, 1.5% for Mastercard).
Read the full documentation: Chargeback detail on /glossary
Representment is your response to a chargeback: evidence submitted within the scheme’s response window (typically 7-30 days depending on reason code) that the original transaction was legitimate.
Strong evidence includes: signed delivery receipt, IP/device fingerprint matching previous successful orders, photo or screenshot of goods or service delivery, cardholder communications referencing the transaction, and any AVS or 3DS authentication data.
Fluxa builds the representment package in the dashboard, you add the evidence files, we submit through the payments partner. Win rates vary by reason code; we publish ours on /developers#disputes.
Read the full documentation: Representment detail on /glossary
Three fees on a chargeback. The chargeback fee (£15-25 depending on scheme) is charged when the chargeback is raised, win or lose. The arbitration fee (around £250) applies only if a case goes to scheme arbitration after a second chargeback. The reversal fee is refunded to you if you win.
If you win the representment, the original transaction amount is credited back to your settlement. The chargeback fee is not refunded; that pays for the scheme processing.
Excessive chargeback ratios trigger scheme monitoring programmes (Visa VAMP, Mastercard ECP) which add monthly programme fees on top.
Read the full documentation: Operational fees on /pricing
If the cardholder rejects your representment, the case can escalate to pre-arbitration (also called second chargeback). The cardholder’s issuer reviews your evidence and decides whether to send it to scheme arbitration.
At this stage the cost rises sharply. The scheme charges a non-refundable arbitration fee (around £250 per case) plus admin fees. Win or lose, those fees stick. Most pre-arbitration cases are resolved without going to full arbitration because the additional cost is rarely worth it for the merchant or the issuer.
Fluxa flags pre-arbitration cases prominently in the dashboard with a recommendation: accept and write off, or contest with stronger evidence.
Read the full documentation: Dispute escalation on /glossary
Two very different disputes need different responses.
Fraud chargebacks (reason codes 10.x on Visa, 4837 on Mastercard) claim the cardholder did not authorise the transaction. If 3DS authenticated, you have liability shift and almost always win on representment with the 3DS data alone. If not authenticated (MOTO, exempt low-value, recurring without prior auth), you carry full liability.
Service-quality chargebacks (goods not received, not as described, defective, refund not issued) carry no liability shift even with 3DS. The defence is delivery proof, customer communications, and your terms-of-service. These have higher merchant win rates than fraud disputes.
Read the full documentation: Chargeback types on /glossary
Three roles per account. Owner: full access including bank details, billing, and account closure (one per account, transferable). Admin: full operational access excluding bank details and closure. Operator: payment ops, refunds, dispute response; no settings access.
Invite teammates from Settings > Team. Invitations go by email with a one-time signup link; the invitee picks their own password and enrols in 2FA at first login.
Every action is audit-logged with actor, IP, user agent and outcome. The audit log is exportable to your SIEM via webhook for compliance retention.
Read the full documentation: Auth detail on /security
TOTP-based 2FA on every Fluxa account. SMS 2FA is not offered (SMS is a known weak factor under PSRs 2017 SCA requirements).
At first login, you scan a QR code with your authenticator app (Authy, 1Password, Google Authenticator, Bitwarden, etc.) and confirm the rolling 6-digit code. Ten single-use recovery codes are generated; save them in your password manager.
2FA is enforced for Owner and Admin roles by default; Operators can be policy-required to enrol via the Team settings. Account recovery without your 2FA device requires a recovery code or a human-verified flow.
Read the full documentation: 2FA detail on /security
If you lose your 2FA device: use one of the ten single-use recovery codes from initial setup. Each works once, can be regenerated after use, and should be kept in a password manager not on the device that does TOTP.
If you have lost both the device and the recovery codes, the human-verified recovery path applies. Email support from the email address on file with: your account name, the bank account details for settlement, recent transaction IDs, and an ID document.
Human-verified recovery takes one to two working days. It is deliberately not fast: the slowness is the security.
Read the full documentation: Auth detail on /security
Account closure is initiated by the Owner from Settings > Account > Close. Closure does not happen immediately: there is a 30-day wind-down during which active subscriptions complete, in-flight settlements process, and outstanding disputes resolve.
Closed account data is retained for the regulatory minimum (six years under Money Laundering Regulations 2017 for AML records; seven years for financial records under HMRC). Personal data is anonymised after the retention period.
If you change your mind during the 30-day wind-down, you can reactivate from the dashboard. After 30 days, reactivation requires a new application.
Read the full documentation: Account closure on /security
Restrict dashboard access and API key usage to specific IP ranges or CIDR blocks. Configurable per API key (so a production key can be locked to your production server IPs while the dashboard remains accessible from anywhere your team logs in).
Matching is bitwise CIDR, supporting individual IPs (203.0.113.42), ranges (203.0.113.0/24), and IPv6. Deny-by-default once enabled.
A break-glass override is available: if you lock yourself out, the Owner can disable allowlisting through a recovery flow that requires re-verifying the account email plus a 2FA challenge.
Read the full documentation: Auth detail on /security
Every action on a Fluxa account is logged: actor (user ID and email), timestamp (millisecond precision), IP and user agent, action (string identifier), affected resource, and outcome (success or specific failure code).
The log is browseable in the dashboard with filtering and search, exportable as CSV, and streamable to your SIEM via a webhook subscription to the audit.log.created event.
Retention is for the regulatory minimum: six years for actions touching customer data (UK GDPR Article 30 records of processing) or financial transactions (HMRC). Immutable: log entries cannot be edited or deleted by any role including the Owner.
Read the full documentation: Audit detail on /security
Three key types. Publishable (pk_) for client-side use, exposes only the operations safe to do from a browser. Secret (sk_) for server-side use, full account access; never put one in client code. Restricted (rk_) for least-privilege automation; scope it to specific operations (e.g. read-only, refunds-only).
Keys come in test and live flavours (sk_test_ vs sk_live_). Mixed-mode calls fail loudly with a clear error.
Rotation: generate a new key, deploy it, revoke the old one. Old and new can coexist briefly so there is no downtime window. Keys are shown once at creation; if lost, you have to generate a new one.
Read the full documentation: API authentication on /developers
Every state change emits a webhook to the URL you configured: payment created, captured, settling, settled, failed, refunded, chargeback raised, dispute updated, and so on. 29 event types at last count, full catalogue on /developers#webhooks.
Each event is signed with HMAC-SHA256. Your endpoint must validate the signature before trusting the payload. Sample verification code is provided in Node.js, Python, PHP, Ruby, Go and C#.
Delivery is at-least-once with exponential backoff (1, 2, 4, 8, 16, 32 minutes). Pair the event ID with your idempotency layer to handle duplicates safely. Failed delivery after 24 hours raises a dashboard alert.
Read the full documentation: Webhook detail on /developers
Official Node.js SDK at launch (npm install @fluxapay/node). Python, PHP and Ruby SDKs in development. The API itself is REST + JSON so any HTTP client works against it.
OpenAPI 3.1 spec is published at docs.fluxapay.co.uk/openapi.yaml for generating clients in any language. Postman collection at docs.fluxapay.co.uk/postman.json for interactive testing.
Webhook signature verification samples in six languages, a CLI helper for sandbox testing, and a hosted-checkout JS SDK for embedded payments are all in the developer docs.
Read the full documentation: SDKs on /developers
Errors return a stable error code (card_declined, insufficient_funds, 3ds_authentication_required) plus a human-readable message and a documentation URL. Cards declines also include a retryable boolean.
Rate limits are per-key per-second with a daily ceiling. The default for new accounts is 100 requests/second; live accounts can request an increase. Limit headers (X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset) on every response.
Exceeding the limit returns HTTP 429 with a Retry-After header. Implement exponential backoff; the SDK does this automatically.
Read the full documentation: Error reference on /developers
Send an Idempotency-Key header on every POST that creates or modifies. Any UUID works; pick one your code can deterministically reproduce on retry (e.g. a hash of the order ID).
Fluxa stores the response for 24 hours keyed by the idempotency key. A repeat request with the same key returns the original response, not a new operation. This makes retries safe even if the network ate the original response after the work completed.
GET and DELETE do not need idempotency keys: GET is idempotent by definition, DELETE is idempotent by API design (deleting a deleted resource is a no-op, not an error).
Read the full documentation: API design on /developers
Sandbox and live are completely separate environments. Separate keys, separate data, separate webhook URLs, separate dashboards. Sandbox data is wiped nightly; live data is retained for the regulatory minimum.
The API surface is identical: same endpoints, same payloads, same event types, same error codes. The only difference is that sandbox returns simulated scheme responses (configurable via test card numbers) and does not move real money.
Switch environments with a single env variable: FLUXA_API_KEY=sk_test_... vs FLUXA_API_KEY=sk_live_.... No code changes between sandbox and live.
Read the full documentation: Sandbox detail on /developers
Hosted checkout and embedded SDK keep merchants in PCI SAQ-A scope, the lightest tier of PCI DSS, available to merchants who fully outsource cardholder data handling.
How it works: card data is entered into an iframe served from the payments partner’s PCI Level 1 environment, tokenised, and returned to your server as an opaque token. Card data never touches your servers, your database, or your logs.
If you use direct API with raw card data (rare; almost no one needs this), you are in SAQ-D scope and we will tell you so explicitly during onboarding. We do not enable raw-card flows by default.
Read the full documentation: PCI scope on /security
Fluxa is a data controller for merchant account data and a data processor for cardholder data on the merchant’s behalf. UK GDPR Article 28 Data Processing Agreement is in your terms.
UK-only data residency: all merchant and cardholder data is stored in UK data centres (Neon Postgres EU-West-2 region). No data is transferred outside the UK without an explicit adequacy decision under UK GDPR.
Data subject access requests (DSARs) are honoured within 30 days; submit at /contact#data. Retention schedules vary by data category: AML records six years, financial records seven years, marketing data one year after last interaction.
Read the full documentation: Sub-processors on /security
Fluxa’s complaints process aligns with FCA DISP rules under the Payment Services Regulations 2017.
Submit a complaint by emailing complaints@fluxapay.co.uk or via the complaints form on /complaints. Acknowledgement within three working days. Resolution within fifteen working days for the majority of complaints, thirty-five working days for the most complex (we will tell you within fifteen days if your case will take longer).
You can escalate an unresolved complaint to the Financial Ombudsman Service within six months of our final response. The FOS is independent, free for consumers, and binding on Fluxa.
Read the full documentation: Complaints process on /complaints
Fluxa operates under the UK Money Laundering Regulations 2017 (MLR 2017) via the FCA authorisation of the payments partner. Every merchant goes through KYB checks at onboarding: company verification, beneficial owner identification, PEP and sanctions screening.
Transaction monitoring runs in real time on every payment. Patterns flagged for review include unusual transaction sizes, sudden volume spikes, structuring (sub-threshold layering), and recipient/sender risk indicators.
Suspicious activity reports (SARs) are submitted to the National Crime Agency. Affected merchant accounts may be frozen pending review; we cannot tell you a SAR has been submitted (legally prohibited).
Read the full documentation: AML detail on /glossary
The full list of third-party services that touch Fluxa data, kept current on /security#subprocessors. Each has a UK GDPR Article 28 contract and is included in Fluxa’s Data Processing Agreement.
Categories include: the payments partner (card data processing, PCI Level 1), Neon (Postgres database, EU-West-2), Cloudflare (CDN and WAF), Fly.io (container hosting), Zoho (transactional email), and a small number of additional services for specific subsystems.
Any change to the sub-processor list triggers a 30-day notification email to merchants. If you object to a new sub-processor, you have the right to terminate without penalty.
Read the full documentation: Sub-processor list on /security
The Financial Ombudsman Service is the UK’s independent dispute-resolution body for financial services complaints. Free for consumers and businesses below a turnover threshold (currently £6.5m).
You can escalate to the FOS if you are not satisfied with Fluxa’s final response to a complaint, within six months of the final response date. Apply at financial-ombudsman.org.uk.
The FOS decision is binding on Fluxa once accepted. We pay any FOS-awarded redress regardless of outcome on our part. The FOS award limit is currently £430,000 for complaints from April 2024 onwards.
Read the full documentation: FOS escalation on /complaints
Real-time uptime by component: API, dashboard, hosted checkout, webhook delivery, settlement. Rolling 90-day chart per component, plus the all-time incident archive with root-cause summaries.
Status pings every 30 seconds from monitoring nodes in London, Amsterdam, Frankfurt and Dublin. Component status updates within a minute of a degradation being detected. Incident posts within five minutes of an incident being confirmed by on-call.
We do not paint the status page green during partial outages. If a single endpoint is degraded, the relevant component shows degraded.
Read the full documentation: Live status feed
Every incident follows the same lifecycle: detection (monitoring alerts or merchant report), confirmation by on-call (within five minutes), public post on /live, mitigation in progress, mitigation complete, monitoring, resolved.
Affected merchants are notified by email to the registered operations contact, by dashboard banner, and by status page update. Severity-one incidents (full outage, settlement frozen, security breach) also trigger SMS to the registered phone number if provided.
Within seven days of an incident closing, a post-incident report is published on /live with timeline, root cause, customer impact, and remediation work.
Read the full documentation: Incident feed on /live
Planned maintenance happens off-peak UK hours (typically 02:00-04:00 UK time on a Tuesday or Wednesday) and never on a settlement day. Most maintenance is zero-downtime; the few that require brief outages are announced at least seven days in advance.
Notification goes to all merchants by email and dashboard banner. The maintenance window shows on /live with start time, expected duration, affected components, and what to expect (often nothing visible to merchants because we deploy behind a rolling load balancer).
Read the full documentation: Maintenance schedule on /live
How affected merchants find out about an incident. Email goes to the registered operations contact within five minutes of the incident being confirmed. Dashboard shows a top-banner with link to the status page. Status page (/live) updates with severity, affected components, and timeline.
Severity-one incidents (full outage of a critical component) also trigger SMS to the registered phone number if you have provided one. Severity-two and below do not SMS by default.
If you do not want incident emails, you can unsubscribe in dashboard preferences, but the dashboard banner cannot be disabled because it is the primary alerting channel.
Read the full documentation: Incident feed on /live
Programmatic access to the status feed for monitoring stacks that do not read email.
RSS feed at status.fluxapay.co.uk/feed.rss for subscription in any feed reader or notification service. JSON feed at status.fluxapay.co.uk/api.json for direct integration with your alerting (Slack, PagerDuty, Opsgenie, custom).
Feed updates within a minute of any status change. Includes incident metadata: severity, affected components, mitigation status, and post-incident report URL once published.
Read the full documentation: Live status feeds
Rolling 90-day uptime by component on /live: API, dashboard, hosted checkout, webhook delivery, settlement. Each component shows percentage uptime, longest incident, and total downtime minutes.
All-time incident archive with severity, duration, affected components, root cause, and remediation. Searchable by date or component. Useful for vendor evaluations or audit evidence.
Fluxa publishes the raw data, not a marketing version. Bad months are visible; we do not hide them.
Read the full documentation: Uptime history on /live