Automated Personalised Video Pipeline (Creatomate)

I run a local SEO agency and need a reusable video template and batch-rendering pipeline built in Creatomate (or another tool you can justify as better for this specific use case — see "Open to Alternatives" below). The end goal: I give you a spreadsheet of prospects, you (or eventually I) hit render, and I get back a personalised ~30-second video for each row — a screenshot with their business name, suburb, and a short marketing script read aloud. This is a template-driven job, not a generative AI video job. No AI-generated footage, avatars, or scenes are required — this is about swapping data into a fixed, well-designed layout at scale. What the video needs to do Each video is ~30 seconds, built from a fixed structure (script and layout stay the same; only the data changes per person): 1. Hook (0–4s) — A screenshot I provide (a Google Maps "3-pack" search result) is shown, with a text/graphic overlay highlighting that the business isn't listed. 2. Problem (4–10s) — Voiceover only, over a fixed background visual (reused every time). 3. Solution/mechanism (10–18s) — Voiceover over a second fixed background visual (also reused every time, not personalised). 4. Credibility (18–24s) — Voiceover, fixed visual. 5. Call to action (24–30s) — Voiceover, includes the business name again, plus a CTA graphic/link. Voiceover will read a script with merge fields for business_name, suburb, service_keyword, and industry. What I'll provide ● The finished 30-second script (fixed, with merge-field placeholders) ● A spreadsheet (CSV) with one row per prospect, including: ○ business_name ○ suburb ○ service_keyword ○ industry ○ screenshot file (the Map Pack image) ● Fixed background visuals / B-roll for the non-personalised scenes (or budget for you to source/create simple ones — happy to discuss) ● Brand colours and any logo/fonts if needed What I need you to build 1. A Creatomate template implementing the 5-scene structure above, with all personalised fields as dynamic variables. 2. Voiceover integration — script text-to-speech generated per row, ideally using ElevenLabs (either via Creatomate's native TTS or connected directly — please advise on the most cost-effective approach and confirm actual voiceover cost per video, since some platforms charge TTS separately from render cost). 3. A batch-rendering process driven by CSV upload, so I can (ideally) run future batches myself without needing you every time. Please document this clearly with screenshots or a short walkthrough video. 4. Confirmation of actual cost per video at volume — I'll be running batches in the thousands (starting with a list of ~ 500 NSW family lawyers, expanding to other services and states over time). Please validate Creatomate's real render + TTS cost at this volume before finalising the build, not just headline pricing. 5. A test batch of 5 sample videos using real data I provide, for approval before building out full batch capability. Ease of use is a priority I'm not a video editor and want to be able to operate this myself after handover — uploading a new CSV and triggering renders — without needing to touch code or hire you again for every batch. Please keep this in mind in how you structure the template and any tooling around it. Open to alternatives I've landed on Creatomate based on my own research (visual template editor, CSV-driven batch rendering, reasonable pricing for template-based — not generative — video). If you have strong reasons to recommend a different tool (e.g. Shotstack, JSON2Video, Plainly) for this specific use case, I'm open to hearing your case, but please justify it against my priorities: ease of self-service use, and cost per video at scale. Deliverables ● Working Creatomate template (or agreed alternative), with account access/ownership transferred to me ● Documented, repeatable batch-rendering process ● 5 test videos using real sample data, approved before full build-out ● Short written or video walkthrough of how to run future batches myself To apply, please include ● Relevant experience with Creatomate, Shotstack, or similar template-driven video APIs ● A brief note on how you'd approach the TTS/voiceover integration and its cost implications ● Examples of similar personalised/batch video work you've done, if available