← all guides
Foundations

This page is our promise to you, written so you can hold us to it. No ads, no sponsored ranking, no hidden tracking, and no brand pays us - ever.

How this app treats you - the anti-app

Almost every consumer app is trying to do something to you: keep you scrolling, harvest your data, nudge you toward whatever pays it, or make leaving feel harder than staying. Conscious Consuming is built on the opposite premise. If we are going to help people resist manipulative technology, the tool itself has to refuse the manipulative playbook.

The short contract. This app should help you decide and leave. No brand can pay to rank higher. Your values, saved items, notes, and contribution drafts live in your own browser unless you export them. The normal app does not need an account, email address, ad network, cookie banner, or behavioral profile. It should be inspectable: sources are linked, scoring is legible, exports are available, and deletion is local. Where a network action exists, it should be explicit, narrow, and named.

What anti-app means in practice

Common app patternAnti-app commitmentHow to check it
Sponsored rankingNo ads, affiliate boosts, brand payments, or pay-to-rankRead the disclosure and compare the source trail on any verdict
Behavioral trackingNo hidden analytics, ad pixels, cookie walls, or accounts in normal useOpen the browser Network tab and reload the app
Data captureValues, saved entries, notes, and drafts stay in local browser storageOpen the You page, export your data, or inspect localStorage
Engagement loopsNo infinite feed, autoplay, streaks, manufactured urgency, or pushy notificationsUse the app for one task; success is getting out with a decision
Lock-inExport and erase are first-class actionsUse the You page to export, import, or delete everything local
Opaque scoringScores are values-relative and source-linked, not a hidden editorial verdictOpen any item and follow the evidence links
Silent growthNew claims should be sourced, dated, and humble about uncertaintyTreat missing or stale evidence as a bug to report

The FTC's dark-patterns work is the public-interest anchor here: design can manipulate people into choices they would not otherwise make. The anti-app rule is simple: if a pattern would be manipulative in another product, we do not get to use it because our cause sounds nicer.

What stays on your device

The core app stores your personal state in your browser, not on a Conscious Consuming server.

Local thingWhat it containsWhy it exists
Values profileYour theme weights and category weightsSo every category can re-rank by what you care about
You settingsRegion, diet, allergy, language, and theme choicesSo the app can remember practical preferences
Saved listEntries you kept for laterSo you can compare without creating an account
Notes and suggestionsYour testing notes and contribution draftsSo feedback is useful before you choose to export it
App cacheStatic files needed for offline use where supportedSo the tool can keep working without a live backend

This is why the You page matters. It is the account replacement: export a JSON file if you want a backup, import it on another device, or erase it locally when you are done. Privacy is not only about hiding; NIST's Privacy Framework frames privacy as a risk-management design problem. Here, the simplest risk reduction is not collecting server-side personal data in the first place.

Where network use can happen

Local-first does not mean every click is metaphysically offline. It means personal state and ordinary ranking do not require a remote profile.

ActionNetwork behaviorWhat leaves your device
Open the app from a hosted siteBrowser requests static files from that hostStandard web request metadata to the host
Compare categories and guidesUses bundled/static data already shipped with the appNo values, saved list, or notes
Follow an external source linkLeaves the app for that source websiteWhatever your browser sends to that site
Use barcode lookup when an item is missingSends the barcode to Open Food Facts for that lookupThe barcode only; your values do not leave
Export, import, or eraseHappens in your browserNothing is sent automatically

If the public deployment ever adds aggregate visit counting, it should be disclosed plainly, configured without personal profiling, and kept out of local development. Hidden analytics would violate this page.

Feature gates before we add anything

Every useful feature still has to pass the anti-app test. A feature can be convenient and still shift power away from the person using it.

Proposed feature asks forIt must proveRed line
personalizationworks from local values or explicit settingshidden behavioral profile
contribution flowexports or submits only what the user choosessilent upload of notes, values, or saved items
reminders or promptsclearly user-requested and easy to turn offstreaks, guilt loops, or manufactured urgency
recommendationsbased on visible criteria and source trailspaid placement or undisclosed ranking boost
accounts or syncoptional, portable, and deletablemaking sign-in required for ordinary comparison
analyticsaggregate, minimal, disclosed, and non-profilingad pixels, cross-site tracking, or user-level surveillance
AI assistancenarrow, labeled, source-aware, and privacy-scopedhidden prompt upload or generated claims without review

This is not a ban on ambition. It is a gate that keeps useful growth from quietly becoming platform capture.

How to hold us to it

  1. Open Developer Tools, then Network. Reload the app and look for third-party scripts, pixels, analytics, or unexpected calls.
  2. Open Application or Storage. Look for the local Conscious Consuming keys that hold values, saved items, notes, and preferences.
  3. Use the You page. Export your data, import it, then erase it. A private tool should make departure boring.
  4. Follow the source trail. Open an entry, read the score basis, and check whether the linked source actually supports the claim.
  5. Inspect the files. The app is static: guides, lenses, datasets, and scoring code are built into readable files.
  6. Report the mismatch. If the app's behavior and this page disagree, the page is not marketing cover; the mismatch is a bug.

EFF's Surveillance Self-Defense is useful because it treats privacy as practice, not branding. This page should be read the same way: not "trust us," but "here is the behavior you can check."

The failure modes we guard against

Failure modeWhat it would look likeThe guardrail
Mission launderingEthical language wrapped around ad-tech behaviorNo ads, no paid ranking, no hidden trackers
Engagement creepStreaks, badges, urgency loops, personalized nudgesFinishable tasks, calm navigation, no manufactured urgency
Source theaterScores with links that do not support the claimDated provenance, source links, and correction paths
Phantom precisionA confident number where evidence is thinCoarse bands, honest missing data, no invented zeros
Lock-inSaved choices trapped in an account or subscriptionLocal export, import, erase, and no sign-in requirement
Product driftUseful features slowly becoming platform captureHandoff discipline, public docs, and anti-app review

Data portability matters because the right to leave changes the power balance. The Data Transfer Initiative works on easier movement of user data between services; this app takes the small-tool version of the same principle: your values should be portable because they are yours, not ours.

The promise is not purity

Not every respectful tool has to be tiny, offline, or local-first. A bank, a library system, a clinic portal, a messaging network, and an AI assistant cannot all behave like a static file. The standard is not "never use a server." The standard is: collect less, explain why, protect what you collect, avoid manipulation, make leaving possible, and let people choose with enough information to disagree.

That is also the product philosophy. Conscious Consuming ranks options by your weights, not ours. It should make tradeoffs legible instead of pretending they disappear. If a category is incomplete, if a source is old, or if a score is contestable, the honest answer is to show the limitation and improve the commons.

If a server becomes necessary

Some future features may need a server: voluntary contribution intake, opt-in sync, collaborative review, source-change monitoring, or account recovery. If that happens, the server should be treated as a narrow tool, not a business model.

Server useAnti-app condition
syncopt-in, encrypted where feasible, exportable, deletable
contributionsexplicit submit action with preview and edit trail
moderation or reviewtransparent queue and no private profiling
aggregate metricscoarse, disclosed, privacy-preserving, and not used for manipulation
notificationsuser-created reminders only, never engagement bait
AI helplabeled, scoped, and never a source of unsupervised verdict claims

The test is whether a person still understands what leaves their device, why it leaves, how long it stays, and how to remove it.

A reasonable default

Use the app when it helps a real decision, then close it. Set your values once, compare the category you came for, save only what matters, export if you want a backup, and erase if you are done. If a feature ever tries to make you stay for its own sake, it has crossed the line.

Useful anchors: FTC Bringing Dark Patterns to Light, NIST Privacy Framework, CISA Secure Our World, EFF Surveillance Self-Defense, and the Data Transfer Initiative.


Next, read digital literacy, healthy tech, or compare privacy-sensitive tools in digital services, AI assistants, password managers, and VPNs.

Read next
Reading the world that's reading you

Digital life feels weightless, but it is full of transactions. You trade attention, behavior, location, messages, searches, photos, contacts, and habits for convenience. Sometimes …

Where your money votes loudest

Not every purchase has the same weight. A $4 chocolate bar can matter, but it does not matter in the same way as a bank account, pension, phone, laptop, or monthly subscription you…

How the scores work — and how to read them honestly

A score you can't see into is just another opinion with a number on it. So here's exactly how the number is made — enough to trust it, and enough to argue with it.