Effective Date: August 28, 2026
This Service Level Agreement ("SLA") is part of the Terms of Service and applies to your use of the Delegance Brokerage platform. Capitalized terms not defined here have the meaning given to them in the Terms of Service. If this SLA conflicts with the Terms of Service on the subject of platform availability, this SLA controls.
This SLA covers the Delegance Brokerage web portal and the authenticated application programming interface used to access it (together, the "Covered Services").
This SLA does not cover, and no availability commitment is made for:
We commit to a Monthly Uptime Percentage of at least 99.5% for the Covered Services, measured per calendar month.
Monthly Uptime Percentage is calculated as:
(Measured Minutes − Unavailable Minutes) ÷ Measured Minutes × 100
Measured Minutes are the minutes in the calendar month during which the monitoring service was itself operating and returning data. Minutes for which the monitor recorded no data are excluded from Measured Minutes and are disclosed on the status page. Unavailable Minutes are counted only within Measured Minutes: minutes of a confirmed incident that fall inside a monitoring gap are excluded from both counts. A monitoring outage is therefore never counted as platform downtime, and unmonitored minutes neither raise nor lower the percentage. The honest limit of this method: an interruption occurring entirely within a monitoring gap is not captured by the calculation, which is why gaps are disclosed rather than silently absorbed.
The monitor. Availability is measured by an independent third-party monitoring service operated outside our infrastructure. It requests the Covered Services from at least three geographic locations at intervals of no more than sixty (60) seconds. We do not measure our own uptime using the systems being measured.
What the monitor requests. The monitor requests a designated endpoint that exercises the portal application, its authentication layer, and a read against the primary datastore. It does not test every feature. This SLA measures the availability of that path, and does not represent that every individual feature is functioning.
What counts as a failed check. A check fails when the monitor receives no response within the configured timeout, receives an HTTP 5xx status, or receives any response other than HTTP 200 carrying the expected content marker. A response blocked by our own edge or bot-protection configuration is a failed check. Keeping the monitor authorized to reach the Covered Services is our responsibility, not yours.
When an incident starts and ends. An incident is confirmed when at least two monitoring locations each record a failed check on two consecutive checks. Once confirmed, Unavailable Minutes are counted from the timestamp of the first failed check in that sequence, not the second. The incident ends at the timestamp of the first check after which fewer than two locations record a failed check,counting a location that returns no data as still failing: recovery requires locations to report healthy, not merely fall silent. Recovery uses the same two-location quorum as confirmation, so one healthy observation point cannot end an outage that other locations still see. Requiring confirmation before counting means a single isolated failure at one observation point is not treated as an outage; counting from the first failed check means the confirmation requirement does not erase real downtime.
Sampling tolerance. Because availability is sampled rather than observed continuously, an interruption shorter than the confirmation window may not be recorded. We disclose this rather than imply the measurement is exact.
Month boundaries. An incident that spans midnight on the last day of a month is apportioned to each month by actual elapsed minutes. Monitor gaps are handled by the Measured Minutes definition above.
Current and historical availability is published on our status page at status.delegancebrokerage.com. The measurements recorded by the independent monitoring service are the authoritative record for this SLA.
The following are excluded from Unavailable Minutes:
Send support requests to [email protected]. That address is the only intake channel covered by this Section. The response clock starts when the request arrives there.
We target the following initial response times. An initial response means a human acknowledging the request with an assessment or a specific request for more information. These are targets for a first response, not for resolution.
| Severity | Description | Initial response target |
|---|---|---|
| P1 — Critical | Covered Services are unavailable to all users, or client data is at risk. | 4 business hours |
| P2 — Major | A core function is materially degraded or unusable, with no reasonable workaround. | 1 business day |
| P3 — General | Questions, minor defects, and requests with a workaround available. | 2 business days |
"Business hours" means 9:00 a.m. to 5:00 p.m. Pacific Time, Monday through Friday, excluding United States federal holidays. A business day is one full period of business hours.
Detection does not wait for business hours. Availability incidents are detected by automated monitoring and alerting that runs continuously, and we begin remediation when an incident is detected regardless of the hour. The targets above govern when you receive a human response, not when we start work.
We assign the initial severity. If you believe a request is under-classified, say so in the thread and we will reassess. The response clock pauses while we are waiting on information we have specifically requested from you, and resumes when you reply.
If we do not meet the Uptime Commitment in a calendar month, and you pay a recurring platform fee, you may request a service credit against that fee:
| Monthly Uptime Percentage | Service credit |
|---|---|
| Below 99.5% but at or above 99.0% | 10% of that month's recurring platform fee |
| Below 99.0% but at or above 95.0% | 25% of that month's recurring platform fee |
| Below 95.0% | 50% of that month's recurring platform fee |
How to claim. Email [email protected] within thirty (30) days after the end of the affected month, naming the month and describing the incident. We will decide the claim within thirty (30) days of receiving it and will show the monitor data we relied on. Credits apply to the recurring platform fee invoiced for the affected month, are applied to a future invoice, are not redeemable for cash, and total no more than 50% of that month's fee. Where the fee is prepaid for a longer term, the monthly fee is the prepaid amount divided by the months in the term.
If you do not pay a recurring platform fee for the Covered Services, service credits do not apply. In that case, if the Monthly Uptime Percentage falls below 95.0% in any month, or below 99.0% in any two months within a rolling six (6) month period, you may terminate your use of the Covered Services and the Terms of Service on written notice, at no cost and with no early-termination charge, and we will provide reasonable assistance exporting your data. Terminating under this paragraph ends your use of the platform. It does not by itself cancel any insurance policy, and it does not by itself terminate any separate brokerage engagement or broker-of-record appointment, which are governed by their own terms.
Scope of the remedy. The credits and termination right in this Section are your sole and exclusive remedy for a failure to achieve the Monthly Uptime Percentage. They do not limit, replace, or govern any remedy for a security incident, a breach of confidentiality, loss or corruption of data, our indemnity obligations, our brokerage or professional obligations, your statutory rights, or any other obligation under the Terms of Service.
Service credits are an adjustment to fees rather than damages. Nothing in this SLA increases the liability caps or changes the exclusions in Section 7 of the Terms of Service.
We may modify this SLA, subject to Section 15 of the Terms of Service. We will notify you by email or through the platform at least thirty (30) days before any change takes effect that materially reduces the Uptime Commitment, the service credits, or the support response targets, or that retires a Covered Service. Continued use after the effective date constitutes acceptance.
Each version of this SLA carries an effective date, and we keep prior versions available on request, so the version that governed any given measurement month can be established.
Questions about this SLA:
Delegance Brokerage
Email: [email protected]
See also our Terms of Service and Privacy Policy.