The moment a SaaS product — an ATS, an LMS, a telehealth app — needs to add video calling, they either embed an existing conferencing tool and inherit all its waiting room friction, or they ask their users to download something. Job candidates trying to join interviews 'sometimes never get in due to technical issues.' Telehealth patients hit password errors. Students sit in blank waiting rooms with no feedback. The developers who feel this pain most acutely are the ones integrating someone else's product and watching their users churn at the join step — but they have no lever to fix it because the join experience belongs to the conferencing vendor, not them.

Conferencing vendors have no incentive to fix this for embedded use cases. Their product is designed around direct users who log in, not anonymous participants who arrive via a link inside a third-party app. The waiting room, the plugin prompt, the moderator-approval gate — these are features to the vendor, not bugs. The developer building the ATS or the telehealth scheduler is the actual buyer, but they're powerless to change behavior one layer down.

What's missing is a video calling layer that a developer can drop into their own product via an SDK — one where the join flow is completely browser-native (no plugin, no app install, no account creation), the waiting room shows actual meeting context the host app supplies, and the host-approval gate can be toggled off programmatically for contexts like interviews where it's just friction. The 10-second join time and the blank loading screen during large webinars both stem from the same root: the conferencing layer has no idea what the surrounding product needs.

This is a business and not a feature because the need recurs with every new customer the SaaS product adds. An ATS paying per active meeting or per seat has a direct line between join friction and candidate drop-off, which is a cost they can name.

What to build

Ship a WebRTC SDK that embeds directly into any web app, exposing a join flow with no plugin requirement, a customizable pre-join screen the host app controls, and a programmatic API to approve or bypass the waiting room — deployable in under a day via a JavaScript snippet and a webhook.

Where to start

Start with ATS vendors specifically — candidate no-shows due to join failure are a metric recruiters already track, so the pain has a dollar figure attached and the sales conversation is concrete rather than abstract.

The hard part

WebRTC infrastructure is expensive to run reliably across regions and NAT configurations — you'll hit a forcing function early where you either under-invest in infrastructure and get blamed for call quality, or over-invest before you have revenue to support it.

How it makes money

Usage-based pricing per participant-minute of connected video, with a flat monthly minimum for predictable infrastructure costs; expansion path as customers grow headcount and meeting volume.

See the evidence. The complaints behind this idea, the products they came from, and similar ideas in Video Conferencing.

More ideas in Video Conferencing