Recurring network dropouts affecting 8 of their 12 staff. Every afternoon, the connection would fall over for 20-30 minutes. Developers couldn't commit code. Sales couldn't access the CRM. Internal estimates put the productivity cost at roughly £800 per incident.
Deep-dive audit of the existing infrastructure — discovered a failing managed switch with failing memory that was causing MAC address table corruption. Replaced the switch, reconfigured VLANs to isolate corporate traffic from guest, and set up proactive port-level monitoring on the replacement hardware.
Zero network incidents in the four months since. Support call volume dropped 40%. The monitoring alerts catch potential problems before they affect anyone.