Amazon Chase Visa 5% cash-back lifetime calc — quantify the substrate’s payoff (parked 2026-05-24)

DARE.CO.UK · PARKED SKETCH · 2026-05-24

Mirrored from ~/.claude/.../memory/parked_sketch_amazon_5pct_cashback_calculator_2026-05-24.md. This is a design sketch parked for future build — read for context, not as a current deliverable.

Why we’re doing the Amazon email-evidence / GDPR-export / sidecar work. Dan needs two numbers: (1) lifetime cash-back earned at 5% on Amazon Chase Visa, year-by-year + total. (2) comparison framing — what alternative cards / strategies would have returned. The substrate already has all the data (orders.jsonl + payment-method field); just needs the rollup.


Dan 2026-05-24: “One of the Amazon calculations I need is how much is 5% per year saving me, plus, comparatively, the purchases, how much do I save. Park the question, but know this is why we are doing this work. 5% cash-back on Amazon Chase Credit Card”

The calculation

Two related numbers, both derivable from the existing Amazon substrate (pa/amazon/_data/orders.jsonl + sidecars):

Number 1 — Lifetime 5% earned

For each year, sum order_total where payment_method contains “Amazon Visa” → multiply by 0.05. Render as:

Year Eligible spend 5% earned
2018 $X $Y
2019 $X $Y
Lifetime total $X $Y

The Amazon Visa card gives 5% back on Amazon.com purchases (formerly “Amazon Prime Rewards Visa Signature” / “Amazon Visa Card” / now Chase-branded “Prime Visa”) and 2% on grocery/gas/restaurants. Only the Amazon.com 5% tier matters for this rollup — the other categories’ data lives elsewhere (or doesn’t, given Harvest’s vendor field issues).

Number 2 — Counterfactual comparison

Compare against what alternative cards/strategies would have returned over the same period:

Strategy Effective rate Lifetime “would-have” return
Amazon Chase Visa (actual) 5% on Amazon $Y
Generic 2% cashback card (e.g. Citi Double Cash) 2% flat $Y × 0.4
Chase Sapphire Preferred (1x base, ~1.5¢/point redeemed for travel) ~1.5% $Y × 0.3
Amex Gold (1x non-bonus categories) 1% $Y × 0.2
Cash / debit (no rewards) 0% $0
Bookshop.org / direct-from-brand (no Amazon at all) 0% from Amazon, but no marketplace tax

The Amazon Visa at 5% is generally the highest-cashback card available for Amazon-tier spending. The honest comparison framing is “5% is good when you’re going to spend on Amazon anyway; if you’d reduce Amazon spend, the 5% advantage shrinks and may invert.”

Where this goes in pa.gf.cx

Surface Section
pa.gf.cx/amazon/savings/ (NEW) The savings landing — per-year table + counterfactual table + lifetime headline
pa.gf.cx/amazon/ (existing landing) Add a stat card linking to /savings/ — “$Y earned this year via 5% cashback”
pa.gf.cx/amazon/years/<yr>.html (existing year pages) Add per-year “5% earned this year” stat

Build cost

Step Effort
Filter orders.jsonl by payment_method containing “Amazon Visa” / “Amazon Prime Visa” / “Prime Visa” / “Amazon Rewards” 15 min
Compute per-year sums + 5% multiplier 10 min
Generate counterfactual comparison table 15 min
Render /amazon/savings/index.html + integrate into existing pages 45 min
Total ~90 min

The reframing

The Amazon email-evidence work + GDPR exports + per-order detail pages aren’t just for insurance claims (future-loss-recovery framing from earlier today). They’re also for operating-intelligence: knowing exactly how much the 5% card has earned, year-over-year, to decide whether to keep it / use it / quit it.

This is the kind of question that’s effectively unanswerable without owning your purchase substrate. Amazon’s UI shows you order count + recent spend, not lifetime cashback math. Chase’s app shows you recent rewards, not lifetime-by-category. The substrate makes the math possible.

Resume conditions

Cross-references

Source: parked_sketch_amazon_5pct_cashback_calculator_2026-05-23.md · Rendered 2026-05-24 06:31