Name the destination and preserve the master
Photo handoff starts with a destination, not with a generic export button. Is the file intended for a web gallery, a client review portal, a press workflow, a lab print, or an in-house proof? Each destination changes what should be checked and how the receiver should interpret the file. Write the delivery intent into the job note before making delicate adjustments. That note anchors the display work: calibration gives you a steadier reference, while the delivery intent tells you what reference matters.
Soft proofing is not a command to make every image duller. It is a controlled preview of how a selected output condition may reshape the document. Inspect saturated colors, near-neutral tones, deep shadows, and highlight detail. Then decide whether any adjustment serves the photograph and the destination. Keep proof-specific changes separate from the master when appropriate, and name the proof condition. This preserves a clear relationship between the file you edited and the version you prepared for a particular output.
Compare a physical proof without chasing the lamp
When a print matters, compare it under a known viewing condition rather than under the nearest decorative lamp. Allow the print and your eyes a moment to settle, then compare broad tone, neutrals, important colors, and detail transitions. Do not chase tiny perceived differences by altering the monitor in the middle of the comparison. First decide whether the gap belongs to paper, proofing light, document settings, or the display baseline. A restrained comparison tells you more than a frantic series of compensations.
A client or collaborator cannot benefit from your calibrated desk if the delivered versions are ambiguous. Use names that distinguish master, proof, web, and final delivery states without inventing complicated codes. Include a short note when a file was prepared for a specific output condition. This gives the receiver a way to ask a useful question rather than saying the image looks different. Version clarity is part of color communication, even though it happens outside the profiling software.
| Layer | Question |
|---|---|
| Master | Is the creative decision intact? |
| Derivative | Does the named output need a bounded adjustment? |
Stage web and client review as separate conversations
A web image moves through browsers, operating systems, displays, and sometimes platform processing. Prepare the expected color space, preserve the profile when the path supports it, and inspect a known tagged reference in the browser and viewing context you actually expect. A calibrated monitor helps you see your own source consistently; it does not make an unknown viewer's display match. Set expectations accordingly, especially for subtle neutrals and near-black detail that can shift on bright consumer screens.
A good review note is brief and concrete: what file is being reviewed, what use it was prepared for, whether the client should judge color on a desktop display or a print proof, and where to place feedback. Avoid claims that the image must appear identical everywhere. A transparent note invites better feedback because it tells the client what kind of difference is meaningful. It also keeps routine display variation from being mistaken for a retouching error.
A photograph may be creatively approved before a final reproduction test, and those are different moments. Creative approval asks whether the image expresses the intended tone, crop, and emphasis. Reproduction approval asks whether a named output path preserves the relevant qualities within its limits. Keeping these decisions separate prevents one unexpected proof from reopening every artistic choice. It also makes your edits easier to trace if a later print condition changes.
Decide what the client is being asked to approve before sharing the link or proof. If the question is crop and content, say so. If the question is final print reproduction, explain the named proof condition. If the file is for general web review, encourage feedback on the image rather than treating every device difference as a retouching request. The review prompt guides attention to the decision the deliverable can support.
This staging protects the creative process. It keeps a client from feeling dismissed while avoiding a promise that a carefully prepared file will look identical on an uncalibrated phone, a bright office monitor, and a physical proof. Specific expectations make feedback more usable and reduce last-minute adjustment cycles.
Archive enough context to reproduce the question
Archive more than the pixels for important jobs. Keep the delivery profile or setting, a small proof note, the date of the monitor profile, and any physical proof observation. You do not need an elaborate database; a project folder note is enough. Months later, this context can answer whether an old file was intended for a particular printer, platform, or reviewer. It turns a recollection problem into a documented production decision.
If a proof differs sharply from the soft-proof expectation, repeat the comparison under the same light and confirm the exact file version. Then check the document profile, proof setup, and printer or lab information before altering the monitor. If a browser review differs, inspect the exported file and browser behavior before treating it as a print problem. Methodical escalation protects the master file from adjustments made to compensate for a single unexplained link in the chain.
For a print-sensitive job, make a compact proof packet: the exact file version, the selected output condition, a note about the intended viewing light, and the date of the display profile used during review. This packet need not be formal or client-facing. It keeps the final comparison tied to one named file and condition. When a print changes, you can locate the decision point rather than revising a master based on an unlabelled proof.
A packet also lets a collaborator reproduce the question. They can review the same version under the same stated condition and report whether the difference appears before or after output. That is more useful than exchanging screenshots of two uncontrolled displays, which usually multiply the unknowns instead of resolving them.
Archive one small reference crop or low-resolution proof alongside the final delivery notes when a job's color context matters. Do not confuse that sample with the protected master; it is a quick visual reminder of the intended balance, neutral relationships, and output version. Months later, it can help determine whether a current question concerns a changed display, a different export, or a misunderstood expectation.
Keep the archive honest about its scope. A small preview cannot establish print accuracy or recreate a monitor calibration. It preserves the production context that made a particular delivery reasonable. Paired with version names and profile notes, that context makes later conversations less dependent on memory.
Close the handoff without altering history
Review a photo in a deliberate order: confirm the master and document profile, inspect the intended proof or delivery preview, create a named derivative if the output requires one, then compare the derivative through the actual path. This order keeps an output-specific adjustment from silently becoming the new master. It also gives a client or lab a file whose purpose is explicit, which makes a correction request easier to understand and reproduce.
If a reviewer sees something unexpected, collect a short disagreement note: which version they opened, where they viewed it, what quality changed, and whether the concern affects content, creative balance, or reproduction. The note may reveal that the issue is crop feedback rather than color feedback, or that a platform preview is involved. A structured observation is kinder to both parties than a sequence of unlabelled revisions built from different screens.
At delivery, preserve the master, name the derivative, retain the relevant output note, and state the review purpose once more. A clear closing message reduces accidental re-exporting and makes later reuse safer. Plain language is enough. Include the file's intended use, any named proof condition, and the place to report a reproducible issue. The calibration baseline remains context for your work, not a hidden guarantee imposed on the recipient.
Before final delivery, compare the derivative with the protected master at equal zoom and in the same managed application. An output conversion can legitimately alter the image, so equivalence is not the goal. Identify what changed on purpose: a paper-dependent adjustment, a restrained gamut compromise, a resize, or a delivery-space conversion. Write those intended differences into the handoff. If an unexpected shift appears later, the team can separate a planned proofing decision from a damaged export or a viewer that ignored the file description.