ScanMessage — AI fraud detection for WhatsApp & SMS in Latin America
Hey IH! Just launched ScanMessage (https://www.scanmessage.dev/) after a few weeks building it in my spare time.
The problem: fraud via WhatsApp and SMS is massive in Latin America. Existing tools detect it but don't explain it — and most are built for English-speaking markets.
What ScanMessage does: you paste any suspicious message and the AI returns a risk score, detected signals, and a plain-language explanation of why it's dangerous and what to do.
Biggest learning so far: fraud scoring requires signal co-occurrence. A single signal like 'requests bank info' isn't enough — you need urgency + impersonation + suspicious link + time pressure together before assigning a high risk score. Took a while to calibrate.
Currently free, B2C first to build real fraud data, then B2B targeting fintechs.
Would love feedback from anyone who's tackled fraud detection or B2B sales in emerging markets.
I can imagine it's tough figuring out market entry for something like this. I actually know a couple of B2B sales professionals in emerging markets who'd probably be happy to answer your questions about that.
That would be incredibly helpful, thank you! I'm especially curious about how B2B sales cycles work in fintech and telco in LatAm — procurement timelines, who actually makes the buying decision, and how to get a first conversation without a warm intro.
If you're comfortable making the intro, I'd really appreciate it. Happy to share more context about what I'm building first if that helps.
Strong problem, especially if you’re moving toward fintechs later.
One thing I’d watch early: ScanMessage explains the function, but it still feels like a small consumer utility.
The actual product is bigger than that.
You’re not just scanning messages.
You’re building a fraud signal layer for WhatsApp and SMS in Latin America.
That matters because fintech buyers won’t only judge detection quality.
They’ll judge whether the product feels trustworthy enough to sit near risk, compliance, and customer protection.
ScanMessage works for B2C testing.
But if the long-term play is B2B fraud infrastructure, the name may start feeling too narrow and lightweight.
Davoq.com would fit that direction much better if you ever want this to feel like serious risk/security infrastructure instead of a message scanner.
Thanks for the thoughtful feedback — really appreciate it.
You're right that the real play is bigger than a message scanner. The long-term vision is exactly what you described: a fraud signal layer for WhatsApp and SMS in Latin America, with B2B as the main direction.
Right now I'm in the first stage — B2C to build real data and validate the detection quality before approaching fintechs. The name works well for that phase.
The positioning shift toward infrastructure is something I'll keep in mind as the B2B conversations start. Good flag on how fintech buyers think about trust and compliance.
Will definitely revisit as the product evolves.
David, one practical follow-up on the ScanMessage thread.
I remembered this one because the product has a real transition problem: it needs to work as a simple B2C scanner now, but eventually be credible as B2B fraud/risk infrastructure for fintech buyers.
That is not just a copy issue. The name, category frame, trust language, and domain layer all affect how fintech teams will read the product once you start approaching them.
If useful, I can do a focused naming/positioning audit around that transition: current ScanMessage risk, B2C-to-B2B framing, fintech trust perception, compliance/risk language, and whether the brand can carry the fraud signal layer direction before more traction builds around the scanner frame.
Not a long consulting thing. Just a sharp written breakdown you can use before the B2B conversations start.
I’m doing a few of these at $99 while refining the format.
If useful, connect here and I can give you a clear outside read:
https://www.linkedin.com/in/aryan-y-0163b0278/
That makes sense for validation.
But I’d be careful with waiting too long.
If fintech is the real direction, the trust frame starts before the first B2B conversation.
Because when you show early traction under “ScanMessage,” buyers will still mentally file it as a consumer scanner first.
That creates extra work later:
prove the tech
then reframe the category
then rebuild trust around infrastructure
If the product is already collecting fraud signal data, the infrastructure story is already starting now.
B2C can be the data engine.
But the brand should probably start preparing for the B2B trust layer earlier than the sales motion does.
That’s where a name like Davoq makes more sense — not as a cosmetic rename, but as a cleaner risk/security frame before fintech buyers anchor you as a lightweight scanner.