Product Management & Platform Strategy · 2026
ZIP FUELER
A product strategy case study — finding a market gap that free tools don't close, pricing the solution to measure demand, and shipping the whole stack solo.
Situation
If you drive for a living, or drive enough that fuel is one of your biggest monthly bills, gas prices aren't background noise. They move every day and vary from one block to the next, and the decision that actually saves money isn't where to fill. It's when, and how much.
The go-to tools today are GasBuddy and Waze, and both run on crowdsourced data. A station's price is only as current as the last driver who bothered to report it, and finding the cheapest option means scrolling a map and comparing stations one at a time. Both also track your location.
Neither one helps a driver decide whether to fill up today or wait, or whether to top off the tank or buy just enough to get by. Zip Fueler answers both questions. Its prices come from HERE Technologies, the same feed fleet-management systems depend on, not from user reports. Every morning, it emails the cheapest Regular and Diesel stations near your ZIP, already ranked, along with a daily read on whether local prices are likely to rise, fall, or hold. There's no app to open and no location tracking.
Who it's for: rideshare and delivery drivers whose fuel comes out of their earnings, long-distance commuters moving between home and work ZIPs, and tradespeople running a work truck on diesel. For each of them, catching a price dip before it ends adds up to real money every month.
Task
Build a product that answers two questions at once — where is the cheapest gas near me right now, and does today look like a good day to fill or should I wait — and deliver it in a form a driver actually uses without thinking about it.
Three constraints guided every decision:
- Solo-operator sustainable. No DevOps burden. No infrastructure that requires a team to maintain.
- Honest unit economics. The business has to break even in the low double digits of paying subscribers, not require venture funding to reach viability.
- Real-world pilot before public launch. Numbers, not gut, decide what happens next.
Action
Chose email as the delivery surface. No app to install, no notifications to manage, readable on any device that already handles email. Side benefit: because there's no app, there's nothing tracking where the driver goes. The ZIP they entered at signup is all the system ever sees.
Picked enterprise data, not crowdsourced. HERE Technologies — the feed auto manufacturers, logistics operators, and government mapping agencies use — not GasBuddy-style user-submitted guesses. Source attribution included in every email.
Built a three-layer architecture, deliberately boring.
- Data layer n8n pulls HERE Fuel Prices and Geocoding per subscriber ZIP each morning, caches aggressively, logs daily price history to Google Sheets.
- AI layer Claude Haiku receives current prices plus fourteen days of local history and the regional wholesale trend from the EIA weekly feed. Returns a structured read on direction and confidence, plus a one-paragraph explanation that cites the data.
- Feedback layer Every email carries thumbs-up / thumbs-down plus a free-text field. Claude classifies free-text by theme and sentiment, feeding both the weekly founder review and the next day's analysis prompt. AI triages; a human judges.
Built the ops layer before the first paying subscriber. Real-time dashboard, hard budget caps on daily/weekly/monthly spend, graceful-pause mechanics with operator alerts and subscriber notices. Because a solo operator can't afford to wake up to a surprise $500 API bill.
Designed the pilot as a price-discovery instrument, not just a soft launch. The headline price — $2.99/mo — isn't a conviction, it's a measurement tool. The pilot is engineered to answer one question before public launch: what will this market actually pay? A 60-day window. Twenty-five free testers in exchange for research answers. Up to three hundred paid subscribers behind an invite gate at $2.99. On Day 50, a willingness-to-pay survey asks subscribers what feels fair; on Day 60, actual conversion-to-paid shows what they'll do. Stated vs. revealed, side by side. That gap sets public-launch pricing from data instead of guesswork. Day-60 call: expand, pivot segment, or wind down, against pre-committed thresholds.
Result
Live and operating. Pilot subscribers receiving daily emails on their chosen cadence. End-to-end flows shipped: signup, invite-gated enrollment, Stripe-backed paid subscription, preferences self-service, one-click unsubscribe with HMAC-signed tokens, operator dashboard.
Stack footprint. Six n8n workflows. Ten Google Sheets tabs as the database. Five external services — HERE Fuel, HERE Geocoder, EIA Weekly Retail, Gmail, Stripe. One operator dashboard. Zero servers.
Unit economics that work at pilot scale. Each paying subscriber nets roughly $2.26 a month on the $2.99 price, about a 76% margin in the mid scenario at the three-hundred-subscriber cap. Break-even is roughly forty subscribers; everything past that is profit. Public-launch pricing will be set by the pilot's willingness-to-pay data.
What this project demonstrates.
- End-to-end product ownership: discovery → positioning → architecture → build → ops → go-to-market.
- Choosing boring technology on purpose, and writing down why.
- Pre-committing to decision gates before data arrives, so the next step isn't argued from feel.
- Taking operator responsibility seriously — budget caps, pause mechanics, and subscriber communication built in before launch, not after an incident.
Next: the day-ninety call, measured against the gates set at day zero.