← June 2026
App Idea Cards 2026-06-06

CreditCadence

A use-it-or-lose-it statement-credit tracker for OfferStack — for every premium card in your wallet, CreditCadence maps each recurring credit to its real reset clock (monthly, quarterly, semi-annual…

CreditCadence

CreditCadence

A use-it-or-lose-it statement-credit tracker for OfferStack — for every premium card in your wallet, CreditCadence maps each recurring credit to its real reset clock (monthly, quarterly, semi-annual, annual), tracks which ones you've already burned this cycle, and tells you exactly how many dollars are about to expire this week before they vanish.

Problem

The refreshed $895 Amex Platinum, the $795 Sapphire Reserve, and the new "coupon-book" cards now bundle $3,500+ in statement credits across a dozen different reset cadences — $25 monthly digital entertainment, $100 quarterly Resy, $300 semi-annual hotel — and the average Platinum holder leaves $300+ unused every year simply because nothing tells them what's left and when each clock resets. On June 2, 2026 even the enthusiast blogs were warning of a "triple threat" of credits expiring at the end of the month, the quarter, and the half-year all at once. OfferStack's Wallet already knows which cards you hold, but it renders each card's credits as a flat text blob (the card_credits table) with no cadence, no used/unused state, and no deadline.

Target user

The Active Optimizer persona from STRATEGY.md §2 — the power user holding 5–15 premium cards, several thousand dollars in annual fees, every card its own coupon book. Job-to-be-done: one screen that says "you have $X in credits resetting in the next 7 days — here's exactly what to burn first," without rebuilding a 12-tab spreadsheet by hand. Secondary user is the Aspiring Optimizer with a single Platinum, overwhelmed by the coupon-book complexity, who needs guardrails so an $895 fee isn't quietly wasted in its first year.

MVP scope

  • A /tools/credit-cadence page plus a "Credits" tab on each card in /wallet (LIF-023), listing every recurring credit with its cadence, dollar value, enrollment requirement, and next reset date.
  • Extend the existing card_credits table from (card_id, credit_description) to a structured card_credit_benefits shape — {card_id, label, value, cadence, reset_anchor, requires_enrollment} — seeded from a repo-checked data/card_credits.json rule-pack so RA can edit policy without a code deploy.
  • A per-user user_credit_redemptions table (RLS owner-only, same plumbing as LIF-014/015) recording which credit was marked used in which period, with a one-tap "mark used" toggle per credit per cycle.
  • A cadence engine that computes, for each unredeemed credit, the exact reset/expiry date and groups them into "expiring ≤7 days / ≤30 days / this quarter" buckets — fed into Next Best Actions (LIF-031) as a new input.
  • A "Show math" panel mirroring LIF-035: total annual credit value on a card minus realized redemptions = net annual fee actually paid, so keep/close decisions sit on real usage rather than the sticker fee.
  • Free tier: single-card public credit calendar (no auth, no marked-used state) as the SEO hook. Pro: wallet-wide tracking, marked-used persistence, and the "expiring this week" email/push digest reusing the LIF-055 reminder pipeline.

Monetization

none (internal tool) — feeds the existing OfferStack freemium model. The public single-card "what credits does this card have and when does each one reset" calculator is the free SEO hook (keyword cluster: "amex platinum credits 2026 reset schedule", "sapphire reserve credit tracker", "what credits expire december 31"). Wallet-wide marked-used tracking, the cross-card "expiring this week" dashboard, and the proactive email/push expiry digest sit behind the existing $5–8/mo Pro tier described in STRATEGY.md §5. One recovered $300 unused-credit year pays for roughly four years of Pro — a concrete upsell story for the Active Optimizer.

Why now

The 2025–2026 premium-card refreshes pushed the "extreme couponing" backlash mainstream: The Points Guy ran a piece on premium cards now demanding an extreme-couponing mindset, Frequent Miler shipped a "Coupon Book Tracker 2.0" adding the new Amex benefits and Citi Strata Elite, and on June 2, 2026 Eye of the Flyer flagged a "triple threat" of month-, quarter-, and half-year credits expiring at the same moment. The pain is measurable value leakage — $300+ per cardholder per year — which is exactly the "tracks in-flight actions and deadlines to prevent value leakage" purpose in PROJECT_REQUIREMENTS.md §1. It slots cleanly into Phase 2 (started 2026-05-11): /wallet (LIF-023) is live and Next Best Actions (LIF-031) is in-flight, so a user_credit_redemptions table drops in next to them on the same LIF-014/015 Supabase + TanStack Query plumbing rather than as net-new infrastructure.

Risks & open questions

  • Demand-side: free trackers already exist (Frequent Miler, Travel on Points spreadsheets, MaxRewards). Does an OfferStack-native, wallet-integrated version convert at the 3–5% Pro rate STRATEGY.md §6.6 targets, or do power users stay on their own spreadsheets? Validate with a landing-page test before scoping the redemption table.
  • Build-side: reset rules are messy — some credits reset on the calendar year, some on the cardmember anniversary, some require a once-per-year enrollment click. The rule-pack must encode reset_anchor per credit or the expiry math is silently wrong and the trust-first positioning in STRATEGY.md §6.3 breaks.
  • Data freshness: issuers add and cut credits mid-year (the 2025 Platinum refresh did exactly this). data/card_credits.json needs RA ownership and a refresh cadence — most naturally a monthly Claude-drafted PR mirroring the LIF-047 CPP-update pipeline.
  • Enrollment gap: knowing a credit exists is not the same as having enrolled in it. MVP can flag requires_enrollment but cannot verify the user actually clicked enroll inside the issuer app — it must never imply a credit is active when it isn't.
  • Self-reported state: without bank-transaction integration (explicitly out of scope per PROJECT_REQUIREMENTS.md §13), "mark used" relies on the user's honesty. Acceptable for v1, but the dashboard must not overclaim precision it doesn't have.

Next step

Promote to a weekly prototype — build /tools/credit-cadence as a free-tier single-card public credit-reset calendar (read-only card_credit_benefits seeded from one month of data for the top five premium cards, no auth, no marked-used persistence) and validate the SEO + Reddit-sharing hook before scoping the per-user redemption table and the expiry digest as separate LIF-### items.

Sources

More from App Idea Cards