AI for Plumbing Contractors.
Plumbing shops in the GTA live or die by which calls get answered first. AI won't fix a clogged main, but it can stop the dispatch board from being the bottleneck on a Friday afternoon.
- $$
Emergency triage that actually triages
An AI agent answers off-hours calls, asks the three or four questions that separate a real flood from a slow drip, and texts the on-call plumber with a priority and an address. The rest go in the morning queue.
- $
Quotes that don't sit in a notebook
A scheduled job pulls open quotes from your CRM or job sheet, drafts a check-in in your voice, and queues each one for one-click send. Mostly catches the ones you meant to follow up on three weeks ago.
- $$
Photo-to-parts-list before the truck rolls
Customer texts a picture of the leaking fixture or panel. AI takes a rough guess at the make, the likely failure, and what to load. Not a quote — a head start so the apprentice doesn't drive back for a part.
- $
Annual service reminders that reference the actual job
Pulls last year's service notes per customer and drafts a reminder that mentions their water heater, their backflow test date, the specific thing they had done. Generic reminders get ignored. These don't.
- $$
Permit and inspection paperwork drafts
AI takes the job notes and drafts the permit application or post-inspection summary in the format your municipality expects. Office still reviews and submits — they just don't start from blank.
- $
Review request tied to the actual visit
After a closed job, AI drafts a review ask that mentions what the plumber actually did. Specific reviews land 4-5x more often than the generic 'leave us a review' text.
Things people ask before getting started.
Can the emergency triage agent actually tell a flood from a slow drip?
Most of the time, yes — because it's asking the same three or four questions a senior plumber would ask. Where is the water, can you shut a valve, is anyone in the unit. If the answers are ambiguous, it errs toward escalation and texts the on-call. It won't always get severity right. The trade-off is a few false positives in exchange for never missing a real emergency at 11pm on a Saturday.
Does this work with my municipality's permit system?
It depends on the municipality. For drafting the permit application or post-inspection summary in the format your office expects, yes — the AI works from your past submissions and the current job notes. Actually submitting through a municipal portal is usually still a human step. A few GTA municipalities have decent APIs, most don't. I won't quote on full submission automation without seeing what your office actually deals with.
What happens if you get hit by a bus?
Fair question. Everything I build is on standard tooling — your existing CRM or dispatch system plus a thin layer on top, usually running on infrastructure you own or rent directly. No proprietary platform you can't leave. Documentation goes to you at handover, and another developer can pick up the code. I'd rather you not need me a year in than build something that traps you.
How long until I see the first real result?
For the smaller builds — quote follow-ups, review requests, reminder drafts — usually two to three weeks from kickoff to you sending the first AI-drafted message. The dispatch and triage agents take longer because they need to be tested against real after-hours calls before you trust them. Plan four to six weeks for those, and another month before they're handling calls without you double-checking everything.
Don't see your situation here?
Most engagements start with a use-case nobody's built before. Tell Echo what you're actually trying to make happen.
Talk To Echo About Your Situation