The latest articles on DEV Community by валерій повзун (@__8648d81b3c). https://dev.to/__8648d81b3c https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3977817%2Fc4509bc6-d6ed-4934-b734-e25c04aa093a.jpg https://dev.to/__8648d81b3c en валерій повзун Wed, 10 Jun 2026 13:54:17 +0000 https://dev.to/__8648d81b3c/ii-4i8i https://dev.to/__8648d81b3c/ii-4i8i <h2> The research </h2> <p>I went through complaint analyses aggregating thousands of posts from Reddit, X and dev communities. One pain point kept showing up outside the dev bubble:</p> <ul> <li>64.8% of people forget to cancel a free trial before getting billed</li> <li>the average US household spends ~$273/month on subscriptions while estimating ~$111 — roughly $3,200/year quietly disappearing</li> <li>89% of consumers underestimate their total subscription spend</li> </ul> <p>The punchline: the most popular "solutions" are subscription-tracking apps that themselves require a subscription — and want read access to your bank account.</p> <h2> The constraints </h2> <p>People's actual complaints translate into a spec:</p> <ol> <li> <strong>No subscription for a subscription tracker.</strong> Free, no paywalled "basic" features.</li> <li> <strong>No bank linking.</strong> Manual entry is fine — people have 5-15 subscriptions, not 500.</li> <li> <strong>No account.</strong> No email, no password, no onboarding.</li> <li> <strong>Privacy by architecture.</strong> If there's no server, there's nothing to leak.</li> </ol> <p>That spec has a natural shape: a single static HTML file with localStorage.</p> <h2> What it does </h2> <p>SubTracker (~34 KB, zero dependencies):</p> <ul> <li>monthly / yearly / per-day spend, multi-currency with user-editable rates</li> <li>per-cycle normalization (weekly/monthly/quarterly/yearly → monthly cost)</li> <li>trial flag → prominent "cancel before X or you'll be charged" banner</li> <li>usage rating per subscription → "you could save N/year" from the ones you barely use</li> <li>upcoming charges (30-day window) with countdown, auto-rolling billing dates</li> <li>category breakdown chart, search/filter/sort, JSON export/import, dark/light theme</li> </ul> <p>No framework, no build step. <code>Date</code> math for billing cycles, CSS variables for theming, a 10-color hash palette for service avatars.</p> <h2> Try it </h2> <p>Live: <a href="https://valera99686.github.io/subtracker/" rel="noopener noreferrer">https://valera99686.github.io/subtracker/</a> — click "Показати приклад" (Show example) for demo data.</p> <p>The interesting lesson for me: the gap between "what people complain about" and "what gets built" is often just packaging. Nobody needs another VC-funded fintech app for this — they need a file that opens in a browser and tells the truth about their spending.</p> <p>What would you add?</p> <p><strong>Update (June 12):</strong> added income tracking. Enter your salary or freelance income and SubTracker shows how much is left after subscriptions and what percentage of your income they eat.</p> showdev webdev javascript sideprojects