When the link drops, the store stops
A single-ISP outage can stall everything: DMS logins, OEM portals, e-contracting/credit apps, VoIP, payment terminals, service lane tablets, parts lookups, inventory photos, even guest Wi-Fi customers expect. One cut fiber—or a neighborhood power blink at the provider—can turn a busy Saturday into paper tickets and angry customers.
The key is ISP diversity (not just “two circuits”)
A true backup must be physically and logically independent from the primary:
Different providers (e.g., fiber + cable, or fiber + fixed wireless/5G).
Diverse paths into the building (separate conduit/entry if possible).
Separate last-mile plants (avoid “same pole/same vault” risk).
Independent CPE and power (UPS on both modems/ONTs and the SD-WAN/router).
Recommended topology for dealerships
Primary: Business fiber (or high-tier cable) sized for photos, VoIP, and DMS traffic.
Secondary: Different ISP—cable, fixed wireless, or 5G/LTE enterprise.
SD-WAN edge: Active/active or active/standby with sub-second failover, application awareness (keep DMS/VoIP first), and automatic path steering.
Tertiary (optional): 5G/LTE for out-of-area disasters or construction cuts.
How big should the backup be?
Size for business continuity, not full showroom speed. As a rule, back-up bandwidth = 30–50% of peak or the minimum needed to keep VoIP + DMS + payments + e-contracts snappy.
Enable QoS so phones and authorizations stay clean even if the sales team uploads photos.
Why it pays for itself (quick math)
If your store loses the internet for 2 hours on a Saturday, you can easily lose multiple closed deals, delayed F&I funding, and a backed-up service drive. One modest outage often costs more than a full year of a secondary circuit. Redundancy turns “store closed” into “customers barely noticed.”
30-day rollout plan
Audit: Document current ISP, hand-off (fiber/cable), contract term, and known pain points; confirm VoIP/DMS requirements.
Source a second ISP: Prefer a different plant (cable vs. fiber) or fixed wireless/5G if construction is slow.
Install SD-WAN: Configure app-aware policies (DMS, credit, VoIP, payments, OEM portals > everything else).
Power & labeling: UPS for both circuits and edge gear; label WAN ports and patching clearly.
Cutover & test: Run live failover drills during business hours (see checklist).
Document & train: One-page “What happens if…” guide for GSM, F&I, Service, and Reception.
Failover test checklist (use during go-live)
Pull primary link → confirm VoIP calls stay up (or re-establish within 1–2 seconds).
DMS/OEM portals stay logged in (or reconnect seamlessly).
Credit app/e-contract submissions succeed.
Payment terminals process a small charge/refund test.
Service tablets sync ROs and parts lookups.
Bring primary back → verify automatic return without sessions dropping.
Common traps to avoid
Buying a “backup” from the same ISP/plant (one backhoe cut takes both down).
Letting both circuits enter through the same conduit.
No QoS on the backup (phones choppy when staff upload photos).
Forgetting IP allow-lists—ensure your credit processor, DMS, and OEM portals accept traffic from the backup egress IPs.
Never testing. Put quarterly failover drills on the calendar.
How AtoZ ISP can help
Provider diversity plan: we source a different-plant secondary circuit and negotiate terms.
SD-WAN deployment: sub-second failover, app-aware QoS for DMS/VoIP/payments, and cellular tertiary if needed.
Turnkey cutover & drills: live failover tests, documentation by department, and ongoing monitoring/reporting.
Bottom line: Two circuits only count as redundancy when they’re truly diverse. Pair that with SD-WAN and QoS, and your dealership keeps selling—even when someone, somewhere, yanks a fiber.
