Last updated: September 2026
Sparky is a tool built by an electrician, for electricians. This policy explains what data we collect, why we collect it, and what you can do with it. No marketing BS.
Account info: Email address and password (hashed) when you sign up.
Profile data: Name, license type, years of experience — only what you choose to enter. Ask Sparky uses role and years of experience to adjust explanation depth; it does not send your profile name to the AI provider.
Job data: Job names, locations, tasks, materials, notes, and photos you create in the Jobs tab. Currently stored only in that browser on that device.
AI conversations: Ask Sparky conversation messages are stored in Supabase for signed-in users. A verified support request is required for deletion.
Material-list scans: AI photo transcription is not active in the current field test. The prepared workflow requires explicit permission and human review before adding anything to a local Job.
Deletion requests: Your verified account ID and request time are stored so support can process an in-app account deletion request. A receipt confirms pending review, not completed deletion.
Retired credential records: The Credential Wallet no longer accepts new records. Any record saved before retirement may remain in Supabase until it is exported or deleted through a verified support request. License-number fields were encrypted on the server; associated metadata and files were not covered by that field encryption.
Usage and diagnostics: Calculator activity, page visits, account-related events, and error reports help us operate the app and investigate problems. During the private beta, Sparky records the most recent day a signed-in account used the app and coarse usage counts; it does not place job content, photos, calculations, or Ask Sparky messages in the owner activity dashboard. Do not put sensitive client information into feedback or error reports.
Account profiles, legacy credential records, and conversation history use Supabase. Jobs drafts and saved calculations stay in your browser for the signed-in account. Some preferences and older unassigned local data also remain on the device; older Jobs are copied to an account only after explicit confirmation. This local storage is not a cloud backup. Signing out does not erase it. Export Jobs backups before clearing site data, and avoid storing sensitive job information on a shared device.
Ask Sparky asks for permission before sending your messages, included conversation history, and profile role and years of experience to the configured AI service. You can withdraw permission for future sends in Ask Sparky; this does not recall prior requests. Production requests are sent directly to Google's Gemini API and are not automatically resent to another production AI provider. Provider handling depends on the Google account and billing settings. Google says content submitted on its Gemini Free tier may be used to improve its products; Paid-tier content is not used that way. Until Sparky confirms the paid-tier setting, do not send private jobsite information, customer secrets, private credentials, or other sensitive information in chat.
When it is enabled in a future release, material-list scanning asks separately before sending the selected photo directly to Google's Gemini API. Sparky's scan endpoint does not intentionally retain the photo or extracted text after returning the transcription, but the provider's handling and retention terms still apply. Do not scan private plans, access badges, passwords, customer secrets, or unrelated people. If you choose to keep the original image with the Job, that copy remains in this browser with the rest of the local Job draft.
Browser voice input asks for separate permission. Your browser or device speech service may process microphone audio to create text. That text is only sent to Ask Sparky when you submit it with AI permission enabled.
Hosting uses Vercel; billing uses Stripe. Analytics integrations include Vercel Analytics and, when configured, Google Analytics, Mixpanel and PostHog. Error alerts may be sent to Telegram. These services process data needed for their functions. Account checks and billing require an internet connection; reliable offline access to the full app is not currently guaranteed.
When configured, PostHog receives explicit action names such as app opened, tool opened, calculation completed, photo saved or failed, and share/export handed to the browser. Properties are limited to fixed tool identifiers, feedback categories and failure/method categories. A random browser identifier expires after 30 days; it is not linked to your account or email. Counts may be incomplete when offline or blocked.
This integration does not send calculator values, Job content, materials, photos, URLs, account identifiers or raw error messages. It does not enable session replay, autocapture, person profiles or location enrichment. The analytics provider still receives network connection information such as your IP address. Do Not Track and Global Privacy Control disable these beta events.
We do not sell your data.
We do not show ads.
Jobs drafts are not automatically synced to a crew or another device.
We do not ask you to send passwords or encryption keys in a support request.
If you are a California resident, you have the right to:
Know what personal data we have collected about you.
Request deletion of your personal data.
Opt out of the sale of your data (we don't sell it, but you have the right).
Not be discriminated against for exercising these rights.
To exercise any of these rights, email ianw@sparkyonsite.com with the subject line "Privacy Request." We'll respond within 45 days.
Request account deletion in Settings → Account Actions. A recorded in-app receipt means your request is awaiting manual review, not that deletion is complete. If you cannot access your account, email ianw@sparkyonsite.com from your account email with the subject “Account deletion request.” We will verify account ownership and confirm the scope and completion of the request. Do not send your password. Submitting a request does not immediately delete your account or cancel a subscription; mention any active subscription in your request.
We cannot remotely erase Jobs drafts, saved calculations, downloaded backups, or other copies on your devices. After saving anything you want to keep, clear Sparky site data in each browser and remove unwanted exported files. Support will explain any records that must be retained and any limits affecting backups or service providers.
We will update this page and its date when this policy changes. Review it before providing new information; automatic policy-change notices at login are not currently available.
Privacy questions?
ianw@sparkyonsite.com