Post-call recap video with an AI avatar for prospects: script and cost

A post-call recap video is a 30-second clip with an AI avatar that restates what you agreed on and the next step. A 30-second clip costs $5.52 at standard, $7.35 at plus, and $16.50 at max, with no product image. Sume avatars are async jobs: you submit a request, get a job id back, and read the finished video later. There is no live video session.

Recaps work because the call already happened live with a person. The clip comes after, when there is time to render, review and send. It is the prepared part of a conversation, written from your notes.

Script structure

Use four beats and about 75 words for 30 seconds: a greeting by first name, two things you heard, the one thing you promised, and the next step with a date. Keep the numbers and names that you can verify from your own notes. Do not add claims the call did not contain. A recap is a summary of a real conversation, and the avatar should sound like a tidy version of you, not a different salesperson.

Review before sending

The advantage of async is the review gap. Read the script, check the first frame with a preview if the clip matters, and approve the render before it goes out. Add captions so the clip works with the sound off. After it completes, fetch the result and put the media URL in the follow-up email, ideally with the same details written out in text, because some recipients will not play a video.

Be clear that it is synthetic

Tell the recipient that the video was made with an AI avatar, in the email or on the page. Do not use the likeness of a real person without their consent, and keep a record of who agreed. The avatar is a handle you created, so tie it to a person or a persona you own, and keep it consistent.

Batch cost for a week

If a team sends 20 recaps a week at standard, the weekly cost is $110.40, or $147.00 at plus. The clips are independent jobs, so you can submit them as each call ends rather than in a batch at the end of the day.

curl -X POST https://api.sume.com/v1/avatar-1.0/talking-video \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: recap-2026-10-07-acme" \
  -d '{
    "avatar_handle": "account_exec",
    "script": "Hi Dana, thanks for the time today. You told us the setup takes too long. I will send the two-step guide tonight. Can we meet Thursday?",
    "quality": "standard",
    "aspect_ratio": "16:9",
    "captions": {"enabled": true, "style": "slam", "language": "auto"}
  }'

When not to send one

Skip the video when the call ended badly, when the prospect asked for text only, or when the recap would contain anything sensitive that the prospect did not agree to put in a recording. Skip it too when you would not be comfortable saying the same sentences to their face. A recap is a promise in writing, and a video makes it more visible, not less.

A good habit is to send the text summary first, then the clip as an optional extra with a short line such as "Here is a 30-second version if you prefer to watch". That gives the prospect the choice and gives you a cleaner record of what was said.

Metrics worth tracking

Track the share of recipients who press play, the share who watch to the end, and the reply rate within a week, compared with a text-only recap. Because every clip has a known cost, you can compute a cost per reply. At $5.52 a clip at standard, ten extra replies from 100 recaps mean $55.20 per extra reply, a figure you can compare with any other follow-up channel.