Turn the pilot into an evidence-based decision on whether to scale.
You have run a pilot on one document type, measured extraction accuracy, tracked credit consumption, and timed the review process. Now you package those findings into a decision brief that tells leadership whether scaling Nanonets across Registry's full document volume is worth the cost.
This guide walks you through assembling the pilot scope, accuracy results, credit burn, per-document cost, full-volume projection, and manual-hours comparison into a single recommendation that leadership can approve, defer, or decline.
Get the extracted data into the systems Registry actually works in.
Prove the accuracy on one document type before scaling to others.
Manage the $200 free credits so the pilot runs without a surprise bill.
Set out the pilot's scope. State the document type you tested and the batch size processed.
Report the extraction accuracy and the review time saved. Present the accuracy percentage measured during the pilot and the average time saved per document compared to manual reading.
Show the credits consumed and the measured cost per document. Pull the total credits used from your tracking record and divide by the number of documents to get a per-document cost.
Project the cost at full Registry volume across document types. Multiply the per-document cost by your real volumes to estimate total spend.
Weigh that against the manual hours it would replace. Calculate the total review hours saved at full volume and convert to staff cost or capacity freed for other work.
Recommend clearly whether to scale, extend the pilot, or stop. If the projected cost is lower than the manual labor cost and accuracy meets Registry standards, recommend scaling. If the pilot was too small or accuracy needs improvement, recommend extending. If the economics do not work, recommend stopping.
**Pilot scope** Document type: [Passport] Batch size: [N documents] **Accuracy and time saved** Extraction accuracy: [X%] Review time saved: [Y minutes per document] **Credits consumed** Total credits used: [Z credits] Cost per document: [Z/N credits] **Full-volume projection** Monthly passport volume: [M documents] Projected monthly cost: [M × (Z/N) credits] Annual cost: [[value] × M × (Z/N) credits] **Manual hours replaced** Time saved per document: [Y minutes] Monthly time saved: [M × Y minutes = H hours] Annual time saved: [[value] × H hours] Staff cost saved: [[value] × H hours × hourly rate] **Recommendation** [Scale / Extend / Stop] because [reasoning based on cost comparison and accuracy].
Report the pilot's accuracy and review-time saving, the credits consumed and cost per document, the projected cost at full volume against the manual hours replaced, and a clear scale recommendation.
Cross-check extracted values against what Registry already holds.
Teach Nanonets the exact fields Registry needs from one document type.