Meta Lead-Gen Copy Performance OS | Track Which Angle Converts Qualified Leads Per Offer Type | Claude Projects
Your highest-CTR ad is probably filling your pipeline with the wrong people. The angle that wins for your demo will probably lose on your guide. Here's why that keeps costing you.A Claude Project that judges copy on lead quality, tracks the winning angle per offer type, and refreshes fatigued copy without throwing away the voice that works.You ran a test. The headline won. So you pasted it onto the next offer — because why wouldn't you? Then it flopped, and you're not sure if it's the copy or the audience or the offer itself. Meanwhile, a primary text with a 2.1% CTR got scaled as the winner three months ago. The sales team has been quietly filtering out the leads it brought in ever since. You found out last Tuesday.The tests are happening. The learning isn't.WHAT IT ISA permanent Claude Project — a Copy Performance OS — that tracks which copy angle produces qualified leads for each specific offer type you run, flags CTR traps before you scale the wrong winner, diagnoses whether declining copy is fatigued or just facing a different audience, and briefs every refresh to preserve the strategy that was working instead of rewriting it away for novelty.WHO THIS IS FOR The Meta specialist who's running demo and guide offers simultaneously and keeps wondering why the copy that crushes on one quietly disappoints on the other — and has never had a system to confirm why. The ads manager who scaled a winner on CTR, then got the sales team's feedback three weeks later and realized the leads were never right to begin with. Anyone who's refreshed fatigued copy, watched the new version underperform the old one, and can't figure out what they lost in the rewrite. The specialist who has months of test data sitting in spreadsheets that have never compounded into an actual answer about what works for this account, this offer, this audience. WHAT YOU GET A CTR trap flag on every review cycle — so when your highest-click variation is quietly your lowest-quality lead producer, you know it before you scale the wrong one. A confirmed winning angle tracked per offer type, under a strict no-cross-transfer rule — your demo winner and your guide winner learned separately, because they answer to different commitment levels and pretending otherwise is expensive. A fatigue-vs-audience-shift diagnosis for any declining variation — so you stop rewriting good copy for a fatigue it doesn't have, and stop blaming a new audience for copy that actually wore out. Refresh briefs that lock in your confirmed winning angle and doubt-neutraliser before anything changes — so every new execution compounds the learning instead of starting over. A Copy Intelligence Log that builds across months — by quarter three, you have a per-offer playbook with confirmed angles, confirmed voice, and confirmed cost-per-qualified-lead that makes every future brief start ahead. A single honest answer after every review cycle: which copy produced qualified leads, per offer type, judged on the metric that actually matters to the business. THE PROBLEM IT SOLVESThe pattern worth knowing — which angle produces qualified leads for which offer — only emerges across many tests and the lag until leads qualify. But every review happens mid-sprint, under time pressure, judged on the metric that resolves fastest (CTR), against the wrong benchmark (whatever won on a different offer). So the account never builds the one asset that would make every future ad better. You get smarter about ads in general and no smarter about this account specifically — and the tests that should have been paying compound interest just sit there as data you never quite decoded.WHY THIS ACTUALLY WORKSThe Copy Performance OS holds every test result tagged by offer type and judged on qualified-lead quality across months — which is what turns disconnected tests into a confirmed map. The no-cross-transfer rule is where the memory earns its place: knowing that the proof-led angle wins for demos while the curiosity angle wins for guides requires holding the whole test history segmented by offer type, not a single result read in isolation. And when copy fatigues, the Log is what lets the refresh preserve the confirmed winning angle instead of rewriting it away — because the brief specifies exactly what to keep and exactly what to change, so the next execution compounds the learning rather than resetting it.A single misread winner — one high-CTR variation crowned before the qualified-lead data resolves — can quietly waste weeks of budget and sales team time before anyone traces it back to the copy decision. This system costs $19 once and runs permanently inside Claude Projects, building a per-offer winning-angle map that makes every future test start with confirmed knowledge instead of a fresh guess. It doesn't replace your judgment — it gives your judgment something real to work with: months of tests, decoded, tagged by offer type, and held so nothing compounds into amnesia.If you're running lead-gen copy across multiple offer types, $19 to know your confirmed winning angle for each one — and keep knowing it through every refresh — is the most straightforward spend in the account.
Get it → aiprompt57.gumroad.com