When a circuit fails, the contract shows its teeth. Teams assume an SLA will make them whole, but definitions, exclusions, and stop-clock clauses often flip the script. Before you sign, put the agreement under a bright light. These nine questions close the costly loopholes that turn a long outage into a small credit.
Downtime Cost Mini-Calculator
Inside the $500K Mistake: 9 Questions Smart Teams Ask Before Accepting a Fiber SLA
1️⃣ What counts as an outage and when does the clock start
- Is the timer tied to ticket creation or provider verification.
- Which events pause the clock: no site access, waiting on customer action, weather holds.
- Are severe degradation events covered or only total loss of light.
- Stop-clock language can turn a six hour failure into a two hour billable event.
2️⃣ Which events are excluded from uptime and credits
- Planned maintenance windows and notice rules.
- Force majeure and third-party dig-ups near the building.
- Transit or NNI issues outside last-mile scope.
- Common exclusions remove the most likely root causes from the SLA math.
3️⃣ How availability is calculated and over what window
- Exact equation for uptime. Example requested in the contract.
- Monthly vs annual window and whether maintenance is excluded.
- Whether downtime counts before a ticket is opened.
- 99.9 percent is about 8.8 hours per year. 99.99 percent is about 52 minutes.
4️⃣ Credit schedule, caps, and exclusive remedy traps
- Tiered credits by outage duration or performance breach.
- Monthly cap relative to MRC and per-event limits.
- Is a credit the only remedy or can you pursue damages.
- Many SLAs cap credits at one month while your loss can be far higher.
Pro tip
Add a chronic breach multiplier so credits increase on repeat incidents in a rolling window.5️⃣ MTTR commitment and real escalation
- Hard MTTR target in hours for priority one incidents.
- On-site dispatch rules and spares policy.
- Named escalation ladder with titles and contact method.
- A tight MTTR without stop-clock tricks is the difference between a hiccup and a headline.
6️⃣ Performance metrics and where they are measured
- Latency, jitter, and packet delivery targets per route or market pair.
- Measurement endpoints: core to core vs to your demarc.
- Sampling method and reporting frequency.
- Backbone-only metrics can look perfect while last-mile pain remains.
7️⃣ Diversity defined and proven
- Definitions for diverse entrances, handholes, and conduits.
- Route maps and splicing diagrams that confirm separate laterals.
- Remedy if both legs of a protected service fail.
- False diversity is common when both legs join the same duct bank near the curb.
8️⃣ Chronic faults and termination rights
- Chronic breach definition such as three priority one outages in 90 days.
- Right to early termination without penalty.
- Escalating credit schedule for repeated violations.
- Short outages can stack into a bad year with no path to exit.
9️⃣ Delivery commitments and penalties for late install
- Firm order commitment date and ready for service definition.
- Credit table for missed milestones.
- Whether install credits are also an exclusive remedy.
- Project slippage can delay revenue and stall migrations at real cost.
Pro tip
Tie FOC to your construction schedule, change freezes, and lease start so incentives align.SLA Question Generator
A strong SLA doesn’t fix outages, it makes them rarer, shorter, and more accountable. Use the checklist above to lock in clear outage definitions, stop-clock limits, real MTTR, measurable performance at your demarc, and documented physical diversity. Run the mini-calculator with your own numbers to quantify the real risk. When the next dig-up or equipment fault hits, you’ll have less downtime, faster restores, and remedies that actually matter to the business.

