Zylo - Sell Project
Listed August 24, 2026
About Zylo
Zylo turns a prompt into a live, hosted website.
Project details
Website: http://www.myzylo.app
Next step: <p>The near-term work is documented in priority order, and the first three items</p><p>are days of work rather than months.</p><p><br></p><p>1. Dunning and card retries. Stripe smart retries and a dunning email sequence</p><p>are available and not currently enabled. 524 subscriptions have ended in</p><p>payment_failed against customer-initiated churn of 5.5% monthly, so the majority</p><p>of lost revenue is cards expiring rather than customers leaving. Manual recovery</p><p>has already returned $3,462 with no automation behind it. This is configuration,</p><p>not engineering, and it also removes the main recurring operator task.</p><p><br></p><p>2. Fraud screening. Stripe Radar is available and not enabled. An afternoon of</p><p>configuration that protects margin on every transaction from the day it is on.</p><p><br></p><p>3. Conversion tracking. GA4 key events are not configured, so visitor to signup</p><p>to paid is unmeasured. Hours of work, and it makes the results of everything</p><p>else visible.</p><p><br></p><p>4. The registered base. 5,212 accounts built entirely through organic search,</p><p>with no lifecycle email, reactivation campaign or onboarding sequence in place.</p><p>An owned audience with no acquisition cost attached to it.</p><p><br></p><p>5. Non-branded search. 98.8% of organic clicks are people searching the Zylo</p><p>name. The site already ranks for 33,800 non-branded impressions on</p><p>commercial-intent queries it has never been optimised against, including</p><p>"fitness website builder", "gym website builder" and the AI website builder</p><p>category broadly. Average position on those is 15.3. The demand is already</p><p>being served to the site and has never been converted.</p><p><br></p><p>6. Inference cost. LLM fees are the largest operating cost and were incurred on</p><p>premium frontier models, because for most of that period those were the only</p><p>ones producing acceptable output on this workload. Cheaper models now match or</p><p>approach it. Every model call routes through a provider-agnostic gateway, so</p><p>switching is a configuration change. No saving is assumed anywhere in the</p><p>financials.</p><p><br></p><p>7. Bring-your-own-key pricing. The gateway already supports customer-supplied</p><p>API keys with billing and spend controls resolving correctly. Packaging it as a</p><p>plan removes inference cost from those accounts entirely and suits customers who</p><p>want their own provider relationship. The plumbing exists; only pricing is</p><p>needed.</p><p><br></p><p>8. Service and build upsell. Customers have converted into paid build work with</p><p>no funnel or outreach in place. Agreement templates are provided at handover.</p><p><br></p><p>The business currently takes about 4 hours per week. Support runs on AI agents,</p><p>deployment is automated, sandboxes scale to zero when idle, and the product is</p><p>not under active development. A buyer executing the above would spend</p><p>considerably more time, by choice rather than necessity.</p>
For sale
Asking price: $27,000 USD
What's included
Domain name, Source code, All related data, All related design work
Four GitHub repositories with full commit history: backend API, product frontend, marketing site, and the sandbox runtime. The Firebase project in its entirety, including Authentication, Firestore, Storage, rules and indexes. 5,212 registered user accounts with full billing history, 3,373 tenants, 4,685 user records and all customer project data. 742 Stripe subscriptions, 46 currently active, migrated to the buyer's own Stripe account with stored payment methods intact via Stripe's migration team. Customers do not re-enter card details. The search footprint: 115 indexed pages, all published content, and the Google Search Console and Google Analytics properties. 30 days of transition support and a full written handover pack covering deployment, sandbox operations, billing operations, a credential register, architecture notes and any known open issues. Agreement templates for the custom build and service work some Zylo customers have purchased. Two obligations transfer with the assets and are stated up front rather than found in diligence. Zylo is the registrant of record for customer domains registered through Vercel, so the buyer inherits renewal responsibility and the customer relationships attached. A full register of those domains and renewal dates is provided at close. And the Stripe account itself does not transfer, as Zylo bills through an account shared with unrelated business, which is why customers migrate rather than the account changing hands. Clean IP title. All code was written by the founder and assigned to Humanik Technologies Inc. No contractor, agency or third-party code is embedded. No debt, liens, revenue-share agreements or investor claims attach to the assets. All three company founders have agreed to the sale. No trademark is registered on the Zylo name.
Reason for selling
I am building other SaaS products and working on projects with the Canadian government, including work backed by federal funding. Zylo is sound and still growing, but the upside documented in the data room needs an operator who can give it dedicated attention. Dunning, non-branded SEO and the service upsell are all sitting untouched for want of hours, not for want of a plan.
Project metrics
Average Monthly Visitors: Under 1K/mo
Average Monthly Revenue: $1K - $5K/mo