Source: https://restorebunny.com/guides/choose-photo-scanning-pilot-sample/

# Choose a Photo Scanning Pilot That Finds Problems Early

Select a small sample across size, surface, contrast and context so a successful scan test does not hide failures in the rest of your collection.

Scanning & preservation4 min readUpdated 11 October 2026

## Try it on your own photo

Have your scan ready? Repair a copy, then compare faces and details with the original.

[Restore my photo](<https://restorebunny.com/signup/?from=seo-content&source=%2Fguides%2Fchoose-photo-scanning-pilot-sample%2F>)1 credit per successful photo · Account required[See credit packs](<https://restorebunny.com/pricing/>)

Quick answer

## The short version

Choose a small sample that represents different capture challenges, not only the easiest prints. Include varied size, surface, tones and contextual writing where safely accessible. Test the complete capture-to-export workflow and record exceptions before committing a large batch.

## A pilot should expose differences

Ten similar glossy holiday prints may demonstrate a pleasant workflow without testing small portraits, pale handwriting or a dark group picture. Use the collection inventory to select representative challenges. A pilot is a deliberate diagnostic sample, not a random accuracy study or a claim that every unsampled item will succeed.

The National Archives advises choosing digitisation methods that handle originals carefully. Exclude objects that need specialist handling from an ordinary scanner pilot and record them as a separate work lane. A difficult source should be represented in planning, but you do not have to risk it to complete the sample. Use a safe substitute or document that the method cannot yet be assessed for that category.

## Select along four practical dimensions

Size affects framing and local detail. Surface affects reflections or visible texture. Tone affects shadow and highlight capture. Context affects whether borders, backs and album relationships must survive. Pick a few sound items that cover these differences and assign stable IDs before capture. One object can test several dimensions, so the pilot need not grow into a full project.

Include the downstream steps: naming, source preservation, an intended delivery export, a copy to the recipient and a check of the received file. A scanner preview alone cannot establish that the final archive package works. Keep settings and time notes, but do not interpret a tiny pilot's success count as a universal tool or service accuracy rate.

## Worked example: eight photos instead of eight easy scans

A fictional pilot contains two small portraits, two dark group photos, a glossy landscape, a textured print, a photograph with a border inscription and a sound front/back pair. They total eight objects, although paired backs produce extra captures. An adhered album is recorded as excluded pending a supported page-capture plan.

The small portrait tests local face detail, the glossy landscape tests lighting, the inscription tests crop completeness and the pair tests naming reconciliation. If the settings work for only six, the useful outcome is an exception plan for the other two. Do not quietly replace them with easy examples so the pilot can be described as perfect.

## Use a pilot acceptance matrix

- Sample ID and reason: list the capture challenge represented. If a challenge has no safe sample, label it untested and state the next handling question.
- Capture settings: record application, resolution, document mode, optional corrections and actual crop. Keep a baseline source for comparison.
- Acceptance criteria: choose focus, complete edges, acceptable reflections, preserved captions and reconciled backs according to the task.
- Delivery check: name the exported file and inspect its dimensions, orientation and visible detail after transfer. A successful working file is not sufficient.
- Exception rule: state recapture, different method, supported specialist handling or retained limitation. Keep the cause of rejection specific.
- Batch decision: proceed for the tested category, modify and retest, or pause. Record who assessed the files and when without inventing an external approval.

## Scale the method by material, not by optimism

If the sample reveals that textured prints need a different capture approach, group those exceptions deliberately rather than changing settings unpredictably midway through the main batch. Preserve the pilot's comparison files and notes. When new materials appear, add a small targeted test instead of assuming that the original pilot covered every case.

Use the scan calculator for sampling comparisons and the source-quality checklist for actual saved-file review. Keep expected output and measured results separate. A pilot helps estimate your own workflow and find operational problems; it does not guarantee historical fidelity, exact face preservation or a fixed processing time. Its value is a clear set of defaults and exceptions that makes the next batch easier to audit.

Keep one successful and one rejected comparison in the batch instructions when they explain an exception. Someone continuing the work can then see the actual crop, reflection or texture problem rather than interpreting a vague instruction to scan carefully. Do not use those two examples to claim a measured success rate for the whole collection.

## FAQs

Is a pilot a statistically valid quality benchmark?

This small deliberate sample is a workflow test, not a general accuracy study. State its scope and untested categories.

Should the most fragile photo be in the pilot?

Represent its handling problem in planning, but do not force it into a method that is unsafe or unassessed.

Must I test the export and transfer too?

Yes. Crops, orientation and resizing can change after capture, so inspect the final received file.

## Sources

Preservation and technical guidance reviewed for this article.

- U.S. National Archives: [Digitizing family papers and photographs](<https://www.archives.gov/preservation/family-archives/digitizing>)

## Related

- [Compare one processing setting at a time](<https://restorebunny.com/guides/test-scanner-automatic-corrections/>)
- [Plan a representative service handoff](<https://restorebunny.com/guides/prepare-photos-for-scanning-service/>)
- [Calculate scan pixels from the print size](<https://restorebunny.com/tools/photo-scan-dpi-calculator/>)

## Ready to restore your photos?

Upload an image, choose your settings, and preview results with a before/after slider.

[Restore my photo](<https://restorebunny.com/signup/?from=seo-content&source=%2Fguides%2Fchoose-photo-scanning-pilot-sample%2F>)[View Pricing](<https://restorebunny.com/pricing/>)

RestoreBunny Team

Photo workflow editors

Guides follow our [editorial policy](<https://restorebunny.com/editorial-policy/>) for sourcing, product claims, AI limitations, updates, and corrections.
