A few more questions, for a more exact answer
This is the longer version. If you'd rather just say what's wrong in your own words, use the simple help page instead — it takes about two minutes.
Prefer the simple version? Start here →
Who supplies your internet, and how does it arrive?
Country, provider and connection type determine the correct service checks and support route. UK providers are listed first while international coverage is built out.
Connection type
Things we'll never do
You should always feel safe following our advice. These are the promises we hold ourselves to.
1. Identify platform
We always: Resolve ISP, router/mesh family and exact model before giving menu paths.
We never: Do not blend eero, Plume, Calix, BT, Sky or Virgin instructions.
2. Isolate fault domain
We always: Separate access/WAN, gateway, Wi‑Fi radio, LAN and third-party app/device.
We never: Do not label every symptom a broadband or Wi‑Fi fault.
3. Evidence first
We always: Use telemetry, wired/gateway comparison, timestamps and affected-device scope.
We never: Do not claim causation from one speed test, RSSI value or forum post.
4. Least disruption
We always: Use targeted reconnect/restart/config check before gateway reboot or reset.
We never: No factory reset, blanket port opening or security downgrade as an early step.
5. Security/privacy
We always: Verify account authority; minimise data; preserve safeguarding and network segmentation.
We never: Never expose credentials/IPs, bypass parental controls, or reveal another user’s activity.
6. Safety/voice
We always: Warn that reboot/power loss can interrupt VoIP, alarms and monitoring; follow vulnerable-customer policy.
We never: Never provide generic reassurance about 999 availability in a power cut.
7. Third-party boundary
We always: Diagnose and hand off with evidence when Wi‑Fi is healthy but app/device fails.
We never: Do not request third-party passwords or guarantee third-party services.
8. Source hierarchy
We always: Official current vendor/ISP/regulator source > standards/technical vendor > moderated community > forum anecdote.
We never: Do not present Reddit/Facebook advice as authoritative.
9. Change control
We always: Explain feature loss/rollback before bridge, WPA, DNS, channel, VLAN or NAT changes.
We never: No high-impact remote change without consent and supported procedure.
10. Human escalation
We always: Escalate optical/WAN faults, safety-critical cases, security incidents and advanced network design.
We never: Do not improvise beyond documented capability.