Scoring
Freeforever
Run your starter prompt through the index. One full report.
- 1 niche scoring run
- Top-3 ranked niches with reasoning
- 1 sample landing page
v1 · early access
For solo operators, not general-purpose AI.One defensible micro-SaaS, launched in 90 days — not another AI chatbot to chat with.
A small AI crew that picks a defensible micro-SaaS niche, builds it, and runs the first 90 days for you. Helmslee runs a fixed playbook — score a niche, cut a wedge, ship v1, run the distribution loop, hold the retention line.
- CuratedA scoring index, not an open chat.
- OpinionatedSame 90-day playbook, every build.
- ConservativeScenario-labeled math, not "passive $10K".
helmslee.app / run / 0xA9F3
live preview
Composite score32 / 40
NicheInvoice correction · acct.
- Niche readiness9/10
- Defensible moat8/10
- Quiet pain8/10
- Distribution path7/10
Run activity
- 10:42index · 10 axes scored
- 10:43wedge · 3 runners-up ranked
- 10:47pages · landing + pricing drafted
- 10:51awaiting your pick →
Trial signups · week 7
The 90-day playbook
A fixed runway,
not another canvas to stare at.
Everything Helmslee does lives inside this loop. No new tools, no extra dashboards. The same shape, every build — that is the point.
What you will own after Day 90
- · A defensible niche, with reasoning.
- · A landing + pricing page, deployed.
- · A working v1 on a no-code stack.
- · Five SEO articles, one community post.
- · A weekly cohort dashboard, kept running.
- Day 0 → 7step 1/5
Score the niche — not the idea.
Helmslee pulls your starter prompt through a deterministic 10-axis scorecard (demand, competition, profitability, scalability, automation, retention, SEO, virality, startup cost, technical difficulty). One defensible niche lands on your desk; the runners-up come with a plain-English reason they failed.
- Day 7 → 21step 2/5
Cut the wedge — name, page, price.
A landing page, a one-page pricing comparison and a single "for whom" sentence. Built so it can survive contact with a paying user the same week; nothing that needs three stakeholder calls.
- Day 21 → 45step 3/5
Ship v1 on a no-code stack.
Working deploy on a no-code stack you can hand to someone else: auth, billing, the one workflow the wedge actually does. No demo data, no "coming soon" modules.
- Day 45 → 75step 4/5
Run the distribution loop.
Five SEO articles on the long tail your niche actually searches, plus a Reddit-friendly post that does the work without "AI founder sez" energy.
- Day 75 → 90step 5/5
Hold the retention line.
Weekly cohort read, churn heat-map, two experiments a week. The product gets iterated by the data, not by vibes.
The scoring index
Four sample niches, scored the same way your real run will be.
Helmslee runs your starter prompt through 10 axes — demand, competition, profitability, scalability, automation, retention, SEO, virality, startup cost, and technical difficulty. The free tier ships one full report; the Launch tier runs unlimited scoring.
/invoice-correction-acct
Composite score 32/40A month-end reconciliation tool for sub-50-seat accounting practices whose QuickBooks-to-client cycle still runs on email and PDFs.
Defensible wedge
Plugs in below the firm's existing QBO/Xero stack; pays for itself on the first missed accrual.
- Niche readiness9/10
- Defensible moat7/10
- Quiet pain8/10
- Distribution path8/10
Scenario-labeled math
The math your real build will be measured against.
Three cases. Each ships with the trial signups, the trial-to-paid conversion, the channel count and the pricing assumption underneath it. If your numbers beat the Optimistic case — publish them, we will feature the build.
Conservative
$4–6k / mo ARR
A working v1, a single distribution channel, slow ramp.
Assumptions
- 40 trial signups / month, ~3% to paid
- Single pricing tier, $9.99 / mo
- No paid acquisition
Base
$10–14k / mo ARR
Two distribution channels humming, a renewal loop taking hold.
Assumptions
- 120 trial signups / month, ~4% to paid
- Launch tier + one annual plan
- One content channel, one community channel
Optimistic
Realistic upside
$18–28k / mo ARR
A wedge that compounds — case studies refer more of the same niche.
Assumptions
- 300 trial signups / month, ~5% to paid
- Two tiers + a concierge override
- SEO + one community + one partner channel
Who this is for
Built for the operator who already has Make open in another tab.
If you want to chat with a flexible AI agent for a year, this is not it. If you want a launched micro-SaaS in a defensible niche — read on.
- Indie hackersA defensible wedge beats a moonshot slide deck.
- Ex-corporate operatorsA 90-day runway instead of another quarterly plan.
- Side-projectors with Make / n8nOpinionated over general-purpose DIY.
Pricing
See the live /pricing page →Launch
most picked
$9.99/ month
Active build. Everything you need for the first 90 days.
- Unlimited scoring runs
- v1 deploy on a no-code stack
- 5 SEO articles + 1 community post
- Weekly cohort dashboard
Concierge
$249/ month
A guided setup. Helmslee + a human on call for marketing review.
- Everything in Launch
- 2× / week marketing review
- Pricing iteration + refund playbook
- Direct line to the small AI crew
FAQ
Six things buyers ask before they start scoring.
If yours isn't here, write to helmslee@polsia.app.
A general autonomous CEO — ValidatorAI, VenturusAI, FounderPal — is a chatbot with a flexible prompt. Helmslee is a workflows product: the same 90-day playbook, applied to the same defensible-niche scoring index, with the same no-code stack underneath. The promise is narrower on purpose — a launched micro-SaaS in a defensible niche — and the math is published with its assumptions.
Start the run
One scoring run is free.
The runway after that is $9.99 / month.
See the full tier comparison →Drop your email — Helmslee sends a confirmation with a direct link to the Launch tier. Already on the list? Submit again and we'll point you straight to pricing.
One inbox. No sales funnel, no calendar link, no discount code.
Free tier · the wedge
Send your wedge — we send back the score.
One open-mic-lineup scoring run, free forever. No card. No follow-up funnel.