8. If I Were You: The Next 90 Days
The single most important thing
Get trade access, then get 5 extinguisher service shops paying at least $99 a month for a hands-on service, before you write product code.
- Segment: US portable fire-extinguisher inspection and service shops with 1–10 techs, in 2–3 regions where you or your advisor know people.
- Job: "Know every unit we service, tell us what's due, and get our reports out without typing them twice."
- How: a concierge MVP. Take the shop's spreadsheets and paper, build their asset register and due list in Airtable, send them a monthly "due next month" list and PDF reports, and charge from week 1. Automate only what you've done by hand at least 20 times.
The plan runs in two versions. Full-time is about 40+ hrs/week. ~15 hrs/week is the same plan with the milestones stretched. The ~15 hrs/week budget is: 6 hours of live calls with US owners, 7 hours of build, 2 hours of content and automation.
Week by week
| Weeks (full-time) | Weeks (~15 hrs/week) | Goal | Exit criteria |
|---|---|---|---|
| 1–2 | 1–3 | Access and discovery. Sign a trade advisor (retired owner, ex-tech, or a distributor's service manager) if you aren't from the trade. Build a list of 60 shops from the NAFED directory and 3 city contractor lists. Interview 20 owners (10 at ~15 hrs/week). Ask: how do you track due dates, how long does a site report take, what do portals cost you each month, what software have you tried and dropped, and why? | You can write the top 3 pains in the owners' own words, with hours and dollars attached. You have a list of which portals they use. |
| 3–4 | 4–7 | Concierge pilot. Offer 8–10 shops a done-for-you service: migrate their records into Airtable, set up the due list, send PDF reports after each visit, and send a "due next month" list on the 1st. Charge ≥$99/month. | 5 shops paying (3 at ~15 hrs/week). If fewer than 3 will pay, the pain isn't where you think. Change the region or shop size, not the pitch. |
| 5–8 | 8–15 | Build the thin v1 (web + offline PWA): customers and sites, asset register with QR/barcode, due-date engine (editable rules), offline inspection (scan, pass/fail, photo, signature), per-site PDF, CSV import, Stripe invoice links. | Pilots move from Airtable to the app. Every pilot shop runs all its inspections in it for one full month. |
| 9–10 | 16–18 | Free hook and channels. A public NFPA 10 due-date calculator, 10 SEO pages ("6-year maintenance schedule", "how to submit extinguisher reports in [city]"), and a recorded demo. Start directory outreach: 20 personalised emails a week, sent in batches. | 30+ calculator uses a week and 5+ trial requests from search or outreach. |
| 11–12 | 19–20 | Measure and decide. | 12 shops paying on the app (6 at ~15 hrs/week), zero lost inspections, and at least one measured money outcome per shop (units found due, report hours saved). Hit: continue to the roadmap. Miss: back to discovery. |
At ~15 hrs/week the plan takes about 20 weeks, not 12. The gates in WHAT-TO-DO-AND-AVOID.md already account for that.
Automation from day one (what keeps the hours low)
- Calendly-style booking for owner calls, with a 5-question intake form.
- A templated migration checklist and a column-mapping sheet. Every migration you do by hand becomes the spec for the import wizard.
- An automated "due next month" email per shop on the 1st, from Airtable in the concierge phase and from the app afterwards.
- A recorded 10-minute demo on sample data, sent before every call.
- One weekly office-hours call for all pilot shops, instead of ad-hoc calls.
Kill criteria (write these down now)
If at day 90 (or week 20 at ~15 hrs/week):
- fewer than 5 shops are paying, or
- any pilot is still keeping a paper copy "just in case" after a month on the app,
then the wedge or the product is wrong. Don't respond by adding features. Respond by changing region, shop size or the core promise. If you still have no trade access at day 30, stop.
The one thing to stop doing
Stop designing for "any trade".
Every configurable field, workflow builder and "industry template" is time not spent with extinguisher shop owners. It feels like leverage, because one engine could serve many trades, but it's the clearest case of building to avoid selling. Rentman built for AV rental and ZenMaid for cleaning. Neither started general.
A close second: stop planning features before the first 5 shops have paid.
What stays on the long-term roadmap (only after proof)
- Deficiency → quote → repair job, and the "revenue found" dashboard.
- Compliance Engine and city-portal export packs, then direct integrations where an API exists.
- Customer read-only portal with certificates.
- Kitchen hood suppression and emergency lighting modules for the same shops.
- Backflow testing module (the runner-up trade).
- Canada or the UK, only after $8k+ MRR in the US.