TradeLedger

These are the three most likely causes of death, ranked by probability. They compound: #1 usually causes #2.

---

Cause #1: No trade access (probability ~45% of failure cases)

What happened:

• The founder had no one inside the trade. They picked fire extinguishers from a spreadsheet of scores, which was the right trade on paper.
• They built a clean asset register and a due-date engine over 4 months, then cold-emailed 400 shops from the NAFED directory.
• 6 replied and 2 took a demo. Neither switched: "We've got 11 years of history on paper, and my office manager would quit."
• The founder didn't know which city portals the shops used, what the tags look like, or that techs work gloved in a plant room with no signal.
• The product fixed problems nobody had and missed the ones they did. By month 9 there were 3 paying shops, all friends of friends.

Early warning signs:

• By day 14, you can't name 10 shop owners who would take your call.
• By day 30, fewer than 3 shops are paying for the hands-on service.
• Your interview notes use your words ("workflow", "asset lifecycle"), not theirs ("tags", "hydro", "six-year", "the city thing").
• You're writing code before anyone has paid.

Fixable? Yes, but only before building. Get a paid advisor or co-founder from the trade: a retired shop owner, an ex-tech or a distributor's service manager. Pay 0.5–2% equity or a monthly retainer. If you can't find one in 30 days, don't start this idea. The research is unanimous: Rentman and ZenMaid both came from inside the trade (10).

---

Cause #2: The onboarding wall (probability ~35%)

What happened:

• Shops liked the demo. Then every one needed its records moved: 2,000–8,000 units, some in spreadsheets, some on paper, some in an old desktop tool.
• Each migration took the founder 6–12 hours. At ~15 hrs/week, that meant 1–2 new shops a month.
• Shops waiting for migration went cold, and shops migrated badly churned in month 2.
• MRR reached $2.4k at month 18 and stalled. The founder was a data-entry service with a software bill attached.

Early warning signs:

• Average founder time per new shop is above 5 hours after shop #10.
• Signed-to-live time above 14 days.
• More than 20% of migrated units need manual fixes after go-live.
• You have a waiting list of shops that "want to start next month".

Fixable? Yes, with effort.

• Build the import wizard and capped AI extraction early (P0, 04).
• Charge a one-off migration fee for large archives, e.g. $1 per unit above 1,000 (price to test).
• Let shops go live with current-year units only and backfill history later.
• Use a part-time contractor for data entry from ~30 shops on.

---

Cause #3: Squeezed from both sides (probability ~20%)

What happened:

• From below: FireInspected and similar tools kept their prices at $49–99 and added import tools and portal exports. Their pitch, "same thing, half the price", landed with solo techs.
• From above: The Compliance Engine kept moving into the contractor's workflow. In June 2026 it launched Compliance Sync, $5 a report to re-key contractor PDFs (10). By 2027 it offered a free inspection app for contractors in its jurisdictions.
• From the side: BuildingReports, already integrated with Brycer, launched a small-shop plan.
• TradeLedger's report and portal advantage disappeared, and what was left was "a nicer due list".

Early warning signs:

• A portal or incumbent announces a free or cheap contractor-side app, or a small-shop tier.
• Lost deals cite price more often than features.
• Your differentiators are all about reports, not about the asset history and the revenue it finds.
• Month-6 churn is above 2%. Shops that leave say "the city app does it now".

Fixable? Partly. You can't stop portals from moving. You can build where they have no incentive to go:

• The shop's own revenue: the due list, deficiency → quote → job, and "revenue found". A portal serves the AHJ, not the contractor's sales.
• Across every jurisdiction a shop works in, because a portal covers only its own cities.
• Across product lines (extinguishers, hoods, lights, backflow), which no single portal covers.
• Migration and service, which big vendors don't do for a 3-tech shop.

---

Honourable mentions (less likely, still real)

• A liability event. A building fire, a unit that didn't work, and a lawyer asking why the due date was missed. Your terms, audit trail and "the licensed contractor is responsible" wording matter (07).
• Data loss. One failed sync that loses a day of inspections ends trust in a small, talkative trade.
• The niche is smaller than assumed. If the real count of extinguisher service shops is closer to 2,000 than 6,000, $10k+ needs module 2 or 3 much sooner (09).
• Founder hours shrink. At ~15 hrs/week, a busy quarter elsewhere means shops go unanswered. In a phone-first trade, that means churn.