FieldLine
The phone line that runs the business.
Hey IH 👋
I launched FieldLine on Product Hunt yesterday. Zero signups. Barely any traffic. Posting here partly because IH is where builders actually talk about what didn't work, and partly because I'd rather process this in public than pretend it didn't happen.
What FieldLine is
A phone system built specifically for US home-services operators — HVAC, plumbing, roofing, electrical, and the rest of the trades. The shape of it:
Dedicated phone numbers per technician and per ad channel (so you can tell whether the Yelp ad is actually working, instead of guessing)
Every call recorded and transcribed automatically
AI summary on every call — name, service, urgency, address, callback time, all extracted from the audio
A structured lead card the moment the call ends, ready to dispatch
No softphone, no CRM to learn, two-minute setup
The job it's trying to do: turn a 4-minute multi-speaker phone call into something a dispatcher can act on in 10 seconds without listening to the audio. That's the core piece of work the AI is doing, and it's the part I'm most proud of — it actually works.
Where it came from
I'm a solo developer. I've spent years building custom apps for businesses in different niches and for the past 2 years, i've built custom solutions for businesses that share one thing: they live and die on phone calls.
Studying how those businesses run, the same knots kept showing up — missed calls becoming missed jobs, voicemails and sticky notes leaking leads, no one remembering the details of the conversation, hours of manual data entry after every shift, and nobody able to answer "is the marketing actually working?" because the phone bill and the ad reports never speak to each other.
Existing tools solve pieces of it, but they're built for enterprise, or require real technical setup, or are just the wrong shape for a 4-person crew running between trucks. FieldLine is my attempt at the right shape.
What's interesting from a building perspective
Flat-tier pricing instead of per-user. Field teams have variable headcount and per-seat pricing punishes growth.
The AI extraction is doing real work, not cosmetic work. Turning noisy multi-speaker audio into a structured lead card reliably enough that a dispatcher trusts it without going back to the recording is harder than it sounds.
Stack: Next.js on Vercel, Supabase for Postgres + auth, Twilio for the phone layer, Claude Opus 4.7 for summaries and lead extraction, Stripe Billing for metered usage, Resend for transactional email.
What I think went wrong yesterday
Honest list, not fishing for reassurance:
No audience. I don't post on social, I don't have a newsletter, I don't have a community. Launching on PH with zero pre-built attention is launching into a void.
The buyer isn't on Product Hunt. PH is builders and tech-curious people. HVAC and plumbing owners are not refreshing the leaderboard.
The product might be solving a problem these operators don't feel acutely enough to switch tools for. I'm not sure yet.
The positioning may be off — "AI phone system" might sound like a feature, not a category, to the people I'm trying to reach.
What I'd love feedback on
If you're a builder: does the product read as authentic? Does the positioning land? What would you cut, what would you say differently?
If you've launched vertical SaaS into an industry you're not embedded in: how did you bridge the gap? Or did you just stop trying and pivot to something closer?
If you've had a flat launch and recovered: what was the unlock?
Brutal honesty welcome — that's actually what I'm here for. Encouragement is nice but feedback is more useful.
PH page → https://www.producthunt.com/products/fieldline
Site → https://fieldline.work
I'll reply to every comment.
Cheers!
Comment