The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
I built WITNESS — customer friction, captured before it becomes churn
Customers rarely submit perfect support tickets. They leave voice notes, screenshots, incomplete explanations—or simply leave.
So I built WITNESS: a focused web app that turns a customer’s statement, intent, and private evidence into one structured Witness Packet.
Customers can speak or type what happened, attach files, and receive a private tracking reference. Teams can review the packet, assign ownership, update its status, and resolve it without exposing internal notes or raw evidence.
WITNESS uses Base44 for data, authentication, storage, permissions, backend functions, and deployment. ElevenLabs handles voice transcription, while AI creates summaries, suggests severity, scores likely spam, and drafts customer updates for human review.
The product feels premium because it is restrained, clear, and built around trust. Instead of another crowded support dashboard, it uses an editorial case-file interface where the customer’s words stay central and every important action leaves a visible record.
https://witness-c1b4526a.base44.app
Built for the Base44 Dev Build-Off.
Very interesting project. For the voice transcription how well does the AI perform when giving severity scores if the voice recording is messy or with a lot of disturbance?
Great question. Messy audio can reduce transcription quality, so WITNESS keeps the original recording, shows an editable transcript, and treats the AI severity score as advisory rather than final. If the transcript is unclear or confidence is low, the packet stays visible for...
That's well thought out. Good job
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Trending
Claude
Claude has entered the design space. How are you using it?
Contra University
Learn from expert creatives how to earn more using next-gen AI tools.
Brand Design
The best brand designers are on Contra. Scroll to see what's trending in brand design. What are you building?
creativeaiflow
Creative AI workflows are evolving. What tools do you use, and what are their strengths and weaknesses?
freelancerlife
Freelancer life is wins, pivots, and everything in between. What’s yours right now?
Related posts
We’ve just wrapped up another Kajabi project! 🎉
What makes this one especially meaningful is that it's our third project with Focal Dystonia Method. For us, there's no better compliment than a client coming back with another idea and trusting us to bring it to life.
This time, we built a complete enrollment system for their new Intensive Program – from a private sales page and application form to flexible payment options, automated email sequences, and a private community, all inside Kajabi.
Long-term partnerships like this are what we value most. Each project builds on the last, allowing us to create even better solutions as our clients continue to grow.
Thank you to Ruth and the Focal Dystonia Method team for trusting us once again. Looking forward to what's next!
A long lasting releatinshop with a customer who trusts you is one of the best feelings ever! Great design tho, wish you more & better sales!
Hi! I joined the Flowstep Challenge to build Partnero AI 👫
To move fast, I’m combining Claude Code Extension in Visual Studio Code with Flowstep MCP✨
The mission of Partnero AI is to help people cope with emotions and find creative ways to resolve difficulties.
😭 😱 😡 --> ☺️ 🥰 🥳
P.S. I'm new to Contra. Let's connect! ☺️
can't wait to see your build
Redouble is a complex B2B sales intelligence platform, the kind of product that's hard to explain in a sentence, let alone a video.
We distilled it into a tight 30-second paid ad that cut straight to the value without losing the depth of what the platform actually does.
Boiling a complex B2B platform down to 30 seconds without dumbing it out is the hard part everyone underestimates. What ended up on the cutting room floor that you really wanted to keep in?
