Pricing that scales with your data volume.

A record is one row processed through the normalization pipeline. A row from Vendor A's CSV and the corresponding row from Vendor B that resolves to the same entity both count as separate records at ingestion. Entity resolution is applied after normalization and does not add to your record count.

Developer includes basic entity resolution: fuzzy name matching using Jaro-Winkler. It does not include custom matching rules, confidence threshold configuration, or the advanced blocking strategies available in Team and Enterprise. If you have complex entity graphs or multi-field matching requirements, Team is the right tier.

Yes. You can pause a paid subscription from the workspace settings. During a paused period, normalization jobs will not run but your config files and canonical schema definitions are retained. You are not billed during a pause. Billing resumes when you reactivate.

Jobs that push you over the limit are queued, not rejected. You will receive an email notification at 80% and 100% of your limit. You can choose to upgrade, purchase a one-time overage block, or wait until the next billing cycle. No data is lost.

Yes. We offer a 40% discount for verified academic institutions and registered nonprofits on the Developer and Team tiers. Email [email protected] with your institutional email and a brief description of your use case. We review and respond within 2 business days.

All payments are processed through Stripe. Prices are in USD. HyperNorm AI is a Bengaluru-based company (Prestige Tech Park III, Karnataka, India). Stripe handles currency conversion if your card is denominated in another currency.