Landing Page Conversion Tracker
Build an internal tool that logs landing-page visits and conversions, computes conversion rate per page and per source, flags your leaky underperformers, and ships a weekly report only after a human approves it.
A private dashboard that compares conversion rate across all your landing pages and traffic sources, flags underperformers against a threshold you set, filters obvious bot spam, and emails a weekly report once you approve it - plus a clean CSV export.
Before you start
- A free Vercel account
- A free Supabase account
- A free Resend account
- A hosted landing-page form or endpoint that can log submissions with page + UTM (or a CSV export of submissions)
The problem this kills
You run a dozen landing pages and you genuinely cannot tell which ones are winning. The numbers live in three places: form submissions in one tool, visit counts in your analytics, UTM tags scattered across campaigns. To answer the simplest question - "which page converts worst, and is it the page or the traffic?" - you end up exporting spreadsheets, doing VLOOKUPs at 11pm, and eyeballing rates you don't fully trust because a bot spike last Tuesday inflated one of them.
So the leaky pages stay leaky. The page that's quietly burning ad spend keeps burning it, because nobody has a clean, side-by-side view of conversion rate per page and per source that they actually believe.
What you'll build
A private internal tool - login-protected, just for your team - that logs every landing-page visit and every conversion (form fill or signup) with its page and traffic source, computes the conversion rate for each page and each source, and flags the underperformers against a threshold you choose. It filters out the obvious bot and spam noise so the numbers are honest. Every week it drafts a conversion report, and a human on your team reviews and approves it before it goes anywhere - so a bot spike never gets shared as if it were real.
No analytics API hooked up yet? No problem. You'll import a visits CSV alongside your logged submissions and the tool still works end to end - and it exports a clean conversion CSV in the exact columns you need.
What's inside the Implementation Plan
The plan opens by interviewing you about your business - your pages, your UTM naming, your typical and peak traffic, what "underperforming" means to you, and the messy edge cases like duplicate submits and bot floods. It reflects a short spec back to you for a thumbs-up, then tailors the data model and every build step to your answers. This is the difference between a generic template and a tool that fits how you actually run pages.
From there it's a step-by-step build: the data model, the submission logger with duplicate guards, the visits import, the conversion math, the underperformer flagging, the approval gate, the dashboard, the weekly email, and the CSV export. Every step ends with a ready-to-paste prompt - you copy it into your AI agent and it does the work.
The governance it includes (this is the point)
This isn't a throwaway script. The plan builds in the controls that make an internal tool safe to trust:
- Login so only your team can open it.
- Row-level security so you only ever see your own organization's data.
- A full audit trail - who logged what, who approved which report, and when.
- A human-in-the-loop approval gate - the tool drafts the weekly report and flags, a person reviews and approves, and only then is it shared. Bot spikes get a sanity check before they ever leave the building.
- Duplicate guards so the same submission can't be counted twice and skew your rates.
Who it's for
Marketing ops and growth people who run multiple landing pages and are tired of guessing which ones convert. If you can fill in a form and follow numbered steps, you can build this - no coding background required.
You've got this - paste the first prompt and let the agent interview you.