# How to Vibe Code Your Own Sendspark (and Stop Paying for It) > Record and Share Personalized Videos at Scale - Site: https://sendspark.com - Category: Sales & Marketing SaaS - Verdict: **Serious undertaking** (38/100 vibecodeable) - Estimated effort: 6-8 weeks of part-time work ## Verdict Build a personal subset with Next.js and ElevenLabs, or keep paying if you need production email deliverability and CRM syncs. While the dashboard, landing pages, and basic video recording can be scaffolded quickly, orchestrating a robust batch-generation pipeline that merges screen recordings with cloned audio variables and custom web background frames introduces heavy async background job complexity. Furthermore, email deliverability for custom video thumbnails and integrating reliably with third-party sales sequencers requires navigating edge cases that will consume weeks of engineering time. ### What you can't replicate - Pre-existing email provider whitelists and enterprise deliverability trust - Native integrations across the entire modern outbound stack (Clay, Apollo, Salesforce) - Optimized enterprise-grade video transcoding and CDN infrastructure ## What it does Sendspark is a video messaging and AI personalization platform that lets users record a base video and use AI voice cloning and dynamic backgrounds to generate personalized video outreach at scale for sales and marketing teams. ### Core features - Screen and camera video recording interface using MediaRecorder API - Video asset management and hosting dashboard - AI text-to-speech voice cloning for variable insertion (e.g., recipient names) - Dynamic background overlay generation (website screenshots/URLs behind camera bubble) - Trackable custom landing pages with embedded calendars and CTA buttons - Animated GIF thumbnail generation for email insertion - CSV contact import and batch generation pipeline - REST API and webhooks for CRM/sequencing tool integrations ## The business ### Pricing - Solo: $49/mo — For individual sales reps and founders - Growth: $99/mo — For growing sales teams - Team: $299/mo — For scaling revenue teams ### Funding $350K raised. - Seed Investors: Active Capital, Xenon Partners Founded 2019. Team size: 24-27. ## The hard parts - Automated audio stitching and voice clone generation that sounds natural across variable names without latency spikes - Client-side screen capture reliability across diverse browser versions and OS permission states - Scaling background video rendering pipelines to stitch custom background frames with camera bubbles for bulk campaigns - Maintaining high email deliverability for generated thumbnail GIFs and tracking pixel redirects ## How to vibe code Sendspark ### Prerequisites - Node.js (free): Runtime for running the Next.js full-stack application - GitHub (free): Source control and deployment pipeline for Vercel/Fly.io - ElevenLabs API Key (Pay-as-you-go (~$5/mo starter)): Required for text-to-speech voice cloning and dynamic variable audio generation ### Recommended AI tools - Claude Code: Best-in-class agentic coding tool for scaffolding complex full-stack applications and background job queues - Cursor: Ideal for iterative UI work on video player components, dashboard views, and landing page templates ### Stack - Frontend: Next.js with Tailwind CSS and shadcn/ui - Backend: Next.js API Routes with Trigger.dev for background video generation jobs - Database: Neon (Serverless Postgres) - Auth: better-auth - Payments: none - Other: ElevenLabs API for voice cloning and variable TTS, Cloudflare R2 for video asset and thumbnail storage, Resend for transactional notifications ### Hosting - Vercel (Hosting the Next.js frontend and serverless API endpoints): $0-20/mo - Cloudflare (Storing recorded video files and generated GIF thumbnails in R2 storage): $0-5/mo - Fly.io (Running FFmpeg container worker jobs for video stitching and background overlays): $5/mo ### Build guide 1. **Project Scaffolding & Database Schema** — Initialize the Next.js project with Tailwind CSS, configure Neon Postgres connection, and set up better-auth for single-user or team authentication. ``` Create a new Next.js project configured with TypeScript, Tailwind CSS, and shadcn/ui. Set up Drizzle ORM connected to a Neon Postgres database. Implement better-auth with email/password authentication. Create database tables for users, videos (id, title, raw_url, duration, user_id, created_at), contacts (id, name, email, company, variable_data, user_id), and campaigns (id, title, base_video_id, status, created_at). Ensure clean modular file organization under src/app and src/db. ``` 2. **Video Recording & Upload Interface** — Build the browser-based video recording module using the MediaRecorder API for simultaneous screen and webcam capture, uploading directly to Cloudflare R2 storage. ``` Build a dashboard page with a video recording interface using the browser's MediaRecorder API to capture both screen share and webcam streams concurrently. Add start, stop, and preview controls with a countdown timer. When recording stops, package the blob and upload it directly to a Cloudflare R2 storage bucket via a signed URL API endpoint. Save the resulting video metadata record in the Neon database with status 'ready'. ``` 3. **AI Voice Cloning & Variable Insertion Pipeline** — Integrate the ElevenLabs API to generate personalized audio snippets for variable tokens like recipient names and wire up an asynchronous processing queue. ``` Create a backend service integrating the ElevenLabs API for text-to-speech generation and voice cloning. Build a batch processing route that accepts a base video ID and a list of contact names/variables from a CSV import. For each contact, generate an audio file pronouncing their name using the cloned voice model, calculate duration offsets, and trigger a background job on Fly.io to stitch the personalized audio introduction onto the base video asset. ``` 4. **Video Rendering & Background Stitching Worker** — Set up a containerized worker on Fly.io running FFmpeg to composite camera bubble overlays over dynamic website screenshot backgrounds and stitched audio tracks. ``` Write a Node.js worker service deployed on Fly.io that consumes video generation tasks. Using fluent-ffmpeg, take a base video recording, apply a circular crop or overlay effect simulating a camera bubble, place a target website background or screenshot behind it, merge the ElevenLabs-generated intro audio segment, and render out a finalized MP4 file along with an animated GIF thumbnail for email previews. Save outputs back to Cloudflare R2 and update the campaign contact status in Postgres. ``` 5. **Trackable Landing Pages & Call-to-Action Widgets** — Build public viewer landing pages displaying personalized videos, custom branding, meeting scheduler embeds, and view analytics tracking. ``` Build a dynamic public landing page route at /[username]/[videoId]/[contactId] that renders the personalized video player with custom branding, headline text, an embedded calendar widget (Calendly/Cal.com), and call-to-action buttons. Implement view analytics tracking that records viewer session duration, percentage watched, and interaction events back to the database via lightweight beacon requests. ``` 6. **CSV Contact Import & Campaign Dashboard** — Develop the campaign management dashboard for uploading contact lists via CSV, monitoring batch generation progress, and copying trackable email snippets. ``` Build a dashboard campaign view allowing users to select a recorded base video, upload a CSV file containing recipient names, emails, and company websites, and map columns to personalization variables. Display a live status table showing generation progress for each contact, and provide quick-copy buttons for formatted markdown/HTML email snippets containing personalized thumbnail GIFs and tracking links. ``` ### Cost vs paying **Starting costs (one-time):** - Domain name (optional): $12 one-time - Initial ElevenLabs & AI API credits: $15 one-time - Total: ~$27 one-time **Ongoing costs (monthly):** - Vercel Hobby/Pro hosting: $0-20/mo - Fly.io worker compute: $5/mo - ElevenLabs API usage for voice cloning: $5-15/mo - Cloudflare R2 storage & bandwidth: $0-5/mo - Total: ~$15-45/mo - Paying for the SaaS instead: $49/mo - $99/mo - Build time: 40-60 hours - AI tool credits: $20 (Claude Pro) + $5 (ElevenLabs credits) - Break-even: 1-2 months ## Sources - [Sendspark Official Website](https://www.sendspark.com) - [Prospeo Sendspark Pricing & Review 2026](https://www.prospeo.io) - [GetLatka Sendspark Revenue Profile](https://getlatka.com) - [PitchBook Sendspark Company Profile](https://pitchbook.com)