Service Level Agreement
Last updated: 13 November 2026
This sets out the availability we commit to, how quickly we respond to support requests, and what you get if we fall short. It applies to accounts on a paid plan and forms part of the Terms of Service.
1. Scope
This SLA applies to accounts on a current paid plan with no overdue balance. Free accounts and trials receive the same engineering attention in practice but are not covered by the credit commitments in Section 6.
The Core Service means the parts of the platform your business and your customers depend on in the moment: customer-facing menus, ordering, booking, appointment and ticket pages; the checkout flow; loyalty card issuance and scanning; and the dashboard needed to operate them.
Supporting features — reporting and analytics, bulk exports, third-party integration synchronisation, AI-assisted menu import and receipt scanning, and bulk messaging — are important but are not covered by the uptime commitment, because they depend on providers and processing we do not control end to end.
2. Availability commitment
99.5% monthly uptime on the Core Service
We commit to making the Core Service available at least 99.5% of the time in each calendar month, measured as described in Section 3. That allows for about 3 hours 39 minutes of unplanned downtime in a 30-day month.
We have deliberately not published a higher figure. LoyalMT is run by a small operation on managed infrastructure, and a 99.9% or 99.99% commitment would be a number we could not honour during a provider incident. We would rather commit to something we can meet and routinely beat than publish a figure that is a breach of contract waiting to happen. In practice we expect to exceed 99.5% most months, and you can hold us to the number we actually wrote down.
The Core Service runs on managed, redundant hosting with automated encrypted database backups and point-in-time recovery. Our recovery objectives in a serious incident are a recovery point objective of 1 hour (the maximum data loss we would expect) and a recovery time objective of 8 hours (the time to restore service). These are objectives, not guarantees.
3. How availability is measured
The Core Service is unavailable during any period in which it does not respond to valid requests, or returns server errors for substantially all requests, for reasons within our control. Availability for a month is:
(total minutes in the month − unavailable minutes) ÷ total minutes in the month
- Measured in whole minutes, aggregated across the month, excluding everything listed in Section 5.
- Degraded performance that still serves requests is not downtime, though we treat sustained severe slowness as a P1 incident under Section 4.
- A fault affecting a single feature while the rest of the Core Service works is assessed proportionately rather than counted as total downtime.
- Our monitoring and logs are the primary record. If your own monitoring disagrees, send it to us with your claim and we will look at both in good faith.
4. Support and response times
Support is provided in English and Maltese by email at hello@loyalmt.com, and through the dashboard.
Support hours
Monday to Friday, 09:00–18:00 Central European Time, excluding Maltese public holidays. For P1 incidents only we also respond outside those hours on a best-efforts basis, including evenings and weekends — which matters in hospitality, where Friday night is the busiest time of the week.
We do not operate a 24/7 staffed on-call rota, and we are not going to claim one. What we do commit to is automated alerting on Core Service failures at any hour, and a genuine effort to respond to a venue that is down during service.
Priority levels and response targets
| Priority | What it means | First response | Target resolution |
|---|---|---|---|
| P1 — Critical | The Core Service is down or unusable for you, or customers cannot order, book or pay. Revenue is being lost right now. | 1 hour during support hours; best efforts outside them | Continuous work until resolved or a workaround is in place |
| P2 — High | A significant feature is broken with no reasonable workaround, or data is displaying incorrectly in a way that affects operations. | 4 business hours | 2 business days |
| P3 — Normal | A feature is not working as expected but there is a workaround, or something is inconvenient rather than blocking. | 1 business day | 10 business days |
| P4 — Low | A question, a cosmetic issue, a configuration request, or a feature suggestion. | 2 business days | No commitment — handled as capacity allows |
We set the initial priority and will tell you if we assess it differently from you, with our reasoning. First response means a human reply that engages with the problem, not an automated acknowledgement. Resolution targets are targets; a fix that requires a third-party provider to act is outside our control, and we will tell you when that is the position rather than leave you guessing.
Helping us respond faster
Please include your business name, what you were doing, what you expected, what happened instead, the time it happened, and a screenshot if you have one. For a P1, say so in the subject line and say which venue is affected.
5. Exclusions
The following do not count as unavailability and do not qualify for a credit. The list is specific on purpose — a catch-all exclusion would make the uptime commitment meaningless.
- Scheduled maintenance carried out in accordance with Section 7.
- Emergency maintenance needed to address a security vulnerability or prevent imminent harm, where we notify you as soon as reasonably practicable.
- Third-party platforms you connected yourself — a POS, delivery marketplace, booking platform or your own website — being unavailable or changing their API.
- Payment provider outages at Stripe or Viva.com, and messaging or wallet platform outages at Resend, Twilio, Bird, Apple or Google. We will tell you what we know and chase them, but we cannot credit you for their downtime.
- Your own connectivity, devices, browsers or local network, including a venue’s Wi-Fi.
- Your configuration or use — incorrect settings, exceeding plan limits, exhausted messaging allowances, a suspended payment provider account, or deleting your own data.
- Suspension under the Terms of Service or the Acceptable Use Policy, or for non-payment.
- Beta, preview and clearly labelled experimental features.
- Force majeure as defined in the Terms of Service, including infrastructure failures at our hosting or database providers that are outside our control, denial-of-service attacks, and national network or power failures.
On provider outages, honestly
A large part of the Core Service depends on Vercel and Supabase. If one of them has a regional outage, LoyalMT goes down with it and there is nothing we can do in the moment beyond escalating and keeping you informed. We exclude that from the credit calculation because we cannot indemnify against a risk we have no control over at our scale — but we will always tell you what has actually happened rather than describe a provider outage as “maintenance”.
6. Service credits
If we miss the availability commitment in a calendar month, you are entitled to a credit against your fees:
| Core Service availability in the month | Credit |
|---|---|
| 99.5% or above | None — commitment met |
| 99.0% to below 99.5% | 10% of that month’s fees |
| 97.0% to below 99.0% | 25% of that month’s fees |
| 95.0% to below 97.0% | 50% of that month’s fees |
| Below 95.0% | 100% of that month’s fees |
How to claim
- Email hello@loyalmt.com with the subject line SLA credit claim within 30 days of the end of the affected month.
- Include your business name, the dates and approximate times of the outage, and what you observed. You do not need to calculate the percentage; we will.
- We respond within 10 working days with our assessment and the measurement we relied on.
Credits are applied against future fees. If your account is closing and a credit is due, we refund it in cash instead rather than leave you holding a credit you cannot use. Credits for a month cannot exceed 100% of that month’s fees, and where an annual plan is in place the monthly fee is taken as one twelfth of the annual amount.
Credits are the only remedy for downtime
Service credits are your sole and exclusive remedy for a failure to meet the availability commitment. We do not compensate for consequential losses such as lost covers, lost sales or reputational harm, and the limitation of liability in Section 24 of the Terms of Service applies. Nothing here limits liability that cannot lawfully be limited.
If availability falls below 95% in any month, or below 99% in three consecutive months, you may also terminate immediately without penalty and receive a pro-rata refund of prepaid fees for the unused remainder of the term.
7. Maintenance
Scheduled maintenance
- Carried out in low-traffic windows, normally between 02:00 and 06:00 CET, and avoided on Friday and Saturday evenings and on public holidays wherever possible, because that is when hospitality is busiest.
- Announced at least 48 hours in advance to the account’s registered email address where we expect any interruption.
- Expected to total no more than 4 hours per calendar month.
- Most changes deploy with no interruption at all and are not announced, because there is nothing for you to do.
Emergency maintenance
Occasionally we have to act immediately — a security vulnerability being the usual reason. We will do it as quickly and narrowly as we can and tell you as soon as reasonably practicable afterwards, including what it was for.
8. Incident communication
During a significant incident we will:
- acknowledge it to affected accounts as soon as we have confirmed it, rather than waiting until we understand it;
- give updates at least every 2 hours while a P1 is ongoing;
- say what is actually wrong, including when the cause is a provider rather than us; and
- provide a written summary within 5 working days of resolving a P1 that caused material downtime, covering what happened, the impact, the cause and what we are changing.
If an incident involves personal data, the breach notification obligations in Section 11 of the Data Processing Agreement apply in addition to this section, and run on their own faster timetable.
9. Data durability and your own backups
We keep automated encrypted database backups with point-in-time recovery, held on our provider’s rotation. In a data-loss event we will restore from the most recent usable backup.
Even so, please export your own data periodically. The export tools are in your dashboard and the data is yours. A business that holds its own recent export is protected against a category of risk that no SLA can fully remove — including the risk of an account being closed, a payment lapsing, or something going wrong on our side.
10. Changes to this SLA
We may update this SLA. We will not reduce the availability commitment or the credit levels for an existing paid account during a term you have already paid for without at least 30 days’ notice, and if a reduction disadvantages you, you may terminate and receive a pro-rata refund of prepaid fees for the unused remainder.
Contact
Support, SLA claims and contractual queries: hello@loyalmt.com
LoyalMT, Address supplied on request, Malta
Revision history
- 13 November 2026 — First published. Previously there was no SLA, and the Terms of Service disclaimed availability entirely while the site described the Service in terms businesses would reasonably read as a commitment. This document sets a 99.5% commitment that can actually be met, defines priority levels and response times, states plainly that there is no 24/7 on-call rota, and gives a credit mechanism with a termination right if availability falls badly short.