IVR System Overhaul for Pay-Per-Lead Network

A pay-per-lead network was matching inbound homeowner callers to home-services companies through a no-code platform that wasn't built for the volume. APIs were inconsistent. The platform had downtime that killed live calls. The automations sprawled across dozens of poorly-named steps that nobody could trace. Inbound calls landed at over 16,000 a month, and the platform was breaking under it.

I built a voice IVR system in Twilio that fronts the network's internal matching API. Inbound calls route through Ringba. When all referral specialists are on the line, the call falls back to the IVR. The IVR collects the caller's postal code and a menu-driven choice of issue type. It then sends a GET request to the internal API for a matched company that can take the call. If a match exists, the IVR reads the company info to the caller and transfers the call through Twilio. From there a Make scenario takes over: it creates a lead record through the internal API, sends an SMS through Twilio to the caller with the matched company's info in case the call drops, sends both a Twilio SMS and a Gmail email to the company with the caller's info, and posts a Slack notification to the internal team with the full lead and company context. The whole system replaced an aging no-code platform that couldn't hold up at this volume.

Inbound calls land in a system that's predictable, debuggable, and built for the volume. The IVR qualifies and transfers in real time. Every qualified lead generates four notifications (caller SMS, company SMS, company email, team Slack) so nothing slips. The team stopped firefighting platform downtime and started working leads.