# PRE-MORTEM: The $8 Productivity App **Date of autopsy: July 3, 2027. The plan is dead. Here is how.** --- ## The obituary *Internal post-mortem, drafted twelve months after commitment.* **July–September 2026.** The remaining 20% was not 20%. It was Stripe webhooks, password resets, onboarding flows, and the empty-state screens nobody counts until a stranger sees them — the build froze six weeks late, in mid-August. The beta invite went out to the 400 Twitter followers on September 8. It got 14 clicks and 9 signups, because 400 followers accumulated by a builder are mostly other builders, and builders don't test productivity apps — they screenshot yours and go back to theirs. Four testers opened the app twice. The plan said ~30. Nobody paused to ask what that gap meant; the beta month was spent fixing what the four reported. **October–November 2026.** Product Hunt launch, November 4, five weeks behind schedule. Eighty-seven upvotes, roughly 400 visitors, 51 free signups, 6 paying. $48 MRR against a $4,000 target. By this point $6,200 of the $15k was gone — contractor UI polish and a logo — spent, note, on the only part of the plan its owner could evaluate. Zero dollars had touched distribution, because "maybe some SEO articles" is not a line item. **December 2026 – March 2027.** Three SEO articles were published in December and ranked nowhere, because the domain was four months old and SEO pays out on an 8–12 month clock this plan never had. Build-in-public was daily in August, weekly by November; the last post is dated January 12. Every quiet week got answered the same way — in the code editor. Dark mode shipped. A Chrome extension shipped. Integrations shipped. MRR peaked at $136 in February and flattened when three of the seventeen subscribers churned in March, all citing "not using it enough." Nights kept being spent; they were just spent on the picturable part. **April–July 2027.** The partner's year expires this month. The plan's own finish line — "500 users within 12 months *of launch*" — was November 2027, a date that was never inside the lease. In May, at $128 MRR, hitting target required 31x growth in months that no longer existed. There was no shutdown announcement. The last commit is June 9. Cause of death recorded as neglect; the autopsy finds the illness was congenital. --- ## Causes of death, ranked **1. ILLNESS — The unresourced half of the plan.** Grows from: *"Marketing = build in public + PH launch + maybe some SEO articles"* — and confirmed by the 2am testimony: *"the building part of this plan is the only part I can picture."* The single word "maybe" — the only hedge in the entire plan — sits on the activity responsible for 100% of the 500 users. Meanwhile $15k is budgeted for "tools/contractor polish": money aimed at the skill already owned. The arithmetic never existed: 500 payers at a typical 2–3% visitor-to-paid rate needs ~20,000 engaged visitors, and nothing in the plan generates 20,000 of anything. **Irreversible earliest** — audiences compound, so an audience not started in month 1 does not exist by month 3, and cannot be retro-built on launch day. The 2am doubt named the failing organ correctly. It always does. **2. ILLNESS — The clock that never added up.** Grows from two lines placed side by side: *"500 paying users within 12 months of launch"* and *"my partner has agreed to one year of this, not more."* Launch is month 3; success is therefore measured at month 15 — a 15-month promise written on a 12-month lease. Compounded by *"it's 80% done"*: the last 20% of software historically eats half the calendar, and every slipped week comes out of a runway that was already three months too short. **Irreversible from day zero** — it is in the plan's own text, signed before the first line of remaining code. **3. NEGLECT — The retreat to the workshop.** Grows from: *"every night and weekend for the next year"* stacked on a full-time job, times the testimony above. When a tired person chooses nightly between a picturable task and an unpicturable one, they choose the picturable one — so marketing is what goes weekly, then monthly, then quiet, while the commit history stays green and the founder feels productive all the way down. This is the mechanism by which Cause #1 executes. **4. MURDER — ranked last, deliberately.** A funded competitor, a Product Hunt algorithm shift, the $8 productivity shelf being a graveyard. All real. Ranked last not because the market is kind, but because this plan dies at home months before anything outside reaches it. --- ## Tripwires **1. For the unresourced half — count beta signups at day 14.** Sign: fewer than 20 signups within two weeks of the beta invite (the plan predicts ~30 from 400 followers — a 7.5% activation rate that deserves testing, not trusting). Pre-committed response: month 2 splits — build stops getting the whole night, and $5k of the $15k is re-earmarked *that week* from polish to distribution (a launch consultant, newsletter placements, someone else's audience). Not "post more." **2. For the clock — the feature-freeze date, on a calendar, now.** Sign: the app is not frozen by July 31, 2026. Every slipped week is countable. Pre-committed response: ship the 80% into beta anyway — the beta's job is to learn whether strangers pay $8, not whether the app is finished. And book the go/no-go conversation with the partner for April 2027 (month 9) with a number attached, so the verdict arrives with runway left instead of at expiry. **3. For the retreat — public posts per week, on a spreadsheet.** Sign: two consecutive weeks under 3 posts. This precedes the MRR flatline by about six weeks; it is the earliest visible symptom of the workshop retreat. Pre-committed response: the marketing hour happens *before* the editor opens each night, or the night doesn't count. Three tripped weeks in a row → stop supplying labor that isn't happening and buy reach instead: $1k/month of sponsorships from the remaining budget. --- ## The hardening 1. **Re-clock the plan to the lease.** The real checkpoint becomes $500 MRR by April 2027 — month 9 of the partner's year — agreed with the partner now. "500 users / $4k" moves to the renewal conversation, not year one's pass/fail. A target the calendar can't contain isn't a target; it's a pre-written apology. 2. **Re-aim the money at the missing skill.** Minimum $6k of the $15k pre-committed to distribution before a dollar more goes to polish. The budget's job is to buy what the founder can't picture, not more of what he can. 3. **Charge from beta day one.** Founder price, $5/month. Thirty free testers measure bugs; ten paying strangers measure a business. Willingness-to-pay is the one number this plan cannot currently produce, and it's the only one that matters by April. **What NOT to add:** no second channel stack (no YouTube-plus-TikTok-plus-cold-email), no expanded roadmap, no cofounder search, no agency retainer, no process. The nights-and-weekends solo operation is the part that deserves to stay fragile — it is the cost of keeping the job, and armoring it destroys the speed that is this plan's only structural advantage. The core bet — a builder shipping a small, honest $8 tool to a niche — is sound. Leave it alone. --- ## The verdict **Most likely cause of death: the marketing this plan never resourced and its author already couldn't picture at 2am — and the hardening fixes it only if the $15k actually buys distribution and the finish line moves inside the partner's year. As written — a 15-month promise on a 12-month lease — this plan should not be signed. The re-clocked version can be.** ⟡⟁⟡ The funeral is over. The plan that walks out is not the one that walked in. ⟡⟁⟡