Photo Evidence in E-commerce Claims: The Complete Guide

Valid photo evidence in an e-commerce claim shows the product, its original packaging and the shipping label in one uncropped image, with the camera's original metadata intact. Without those three elements together, a retailer or marketplace is entitled to ask for more before processing a refund.

What a claim photo actually needs to prove

A single close-up of a defect proves nothing on its own — it could belong to any product from any order. A photo only becomes usable evidence when it links three facts at once: what arrived, what condition it arrived in, and that this specific item belongs to this specific order. That is why serious return policies ask for a minimum set: the sealed box with a readable shipping label, the opened box showing the original packing materials, then a close-up of the fault with something for scale.

Requiring photo evidence before approving a refund measurably deters opportunistic claims, because the documentation burden filters out casual fraud attempts before they reach a human reviewer — a pattern confirmed by ReferralCandy's guide to damaged-goods return policies, which notes that photo requirements only add friction for the claims that need scrutiny, not for genuine customers. Amazon and eBay seller forums repeat the same complaint: buyers occasionally submit a photo of a completely different, already-damaged item and expect the swap to go unnoticed.

The UK legal baseline: what the Consumer Rights Act actually requires

Under the Consumer Rights Act 2015, a trader must supply goods that are of satisfactory quality, fit for purpose and as described. Section 22 of the Act gives consumers a 30-day short-term right to reject faulty goods for a full refund, and during that window the trader cannot insist on a repair or replacement instead — a right explained in detail by Which? and the House of Commons Library briefing on faulty goods. After 30 days, the trader gets one opportunity to repair or replace before a final right to reject applies.

None of this removes the practical need to document damage quickly. Couriers typically require written notice of transit damage within a short window after delivery, and a photo taken on the day of receipt — with the phone's original timestamp preserved — is what establishes that timeline if the claim is disputed later. Citizens Advice recommends keeping all packaging and evidence until the issue is resolved, precisely because it strengthens the consumer's position if the retailer later questions when or how the damage occurred.

Why claim photos are challenged more often now

A support agent reviewing a photo has historically had no reliable way to confirm it shows the actual returned parcel, hasn't been cropped to hide something relevant, or hasn't already been used on a previous claim. Three patterns keep coming up in seller-side accounts of return fraud:

1. Recycled photos — the same image of a broken item circulates across multiple refund requests, sometimes weeks apart, the same underlying pattern our guide on detecting recycled fraud documents covers for paper evidence.
2. Screen recapture — a genuine photo from an unrelated claim gets displayed on a screen and rephotographed, which destroys the original metadata and introduces characteristic moiré artefacts; see our screen recapture detection guide.
3. AI-generated or edited damage — a crack, stain or dent is added digitally to an otherwise genuine product photo, or the whole image is synthetic. This is the fastest-growing pattern and the hardest for a human reviewer to catch on sight.

On seller forums, one question comes up constantly: how do you prove a buyer photographed a different item than the one you shipped? The honest answer is that a photo alone, without any technical check on metadata or scene consistency, rarely settles it — which is why cross-checking several signals beats relying on visual inspection alone.

Generative AI has changed what a "photo" can mean as evidence

Consumer image-generation tools can now produce a convincing product defect in seconds, with no photo-editing skill required, which pushes return fraud into territory that manual review alone no longer covers. CheckFile's AI-generation detection page describes how signals such as repetitive textures, inconsistent shadows and unusual compression artefacts complement — rather than replace — existing manual checks on photos submitted with a claim. The goal is giving reviewers signals they cannot reliably spot at scale on their own, not automating the final decision.

Checklist for support and fraud teams

• Always require the sealed box, the opened box and the fault as one photo set — never rely on a single isolated image.
• Reject screenshots of photos outright; ask for the original image file or a direct share from the phone's gallery.
• Check that EXIF metadata (timestamp, device model) is consistent with the order and delivery dates.
• Compare incoming photos against the history of previous claims, visually and by hash, to catch recycled evidence.
• Log every claim with a timestamped case reference, so the timeline can be reconstructed if a dispute reaches a certified Alternative Dispute Resolution provider, such as the Consumer Ombudsman.

At scale, this cross-check between documents, metadata and scene consistency follows the same principles laid out in our document verification guide, which applies equally to standard paperwork and to the photo attachments on a claim. Our document verification for e-commerce and marketplaces page covers this workflow in more detail for support teams handling this claim volume.

What an unchecked photo policy actually costs

The cost of a weak photo-review process isn't limited to the refunds issued in error — it includes agent time spent judging images by eye, unnecessary product write-offs, and a learning effect among customers who work out that a return policy is lenient. A document verification layer built into the claims workflow can route higher-risk cases to a stricter human review instead of treating every photo with the same level of scrutiny. Retailers that formalise this control typically document it alongside their data-handling practices, in the same way our security page sets out how customer data is protected.

Frequently Asked Questions

Is one photo enough to justify a refund?

No. A single close-up of the defect, without the packaging or shipping label visible, doesn't link the evidence to a specific order. Serious retailers ask for at least three angles: sealed box, opened box, and a close-up of the fault.

Is the customer required to keep the original packaging?

The law doesn't explicitly demand it, but keeping the packaging until the claim is resolved strongly supports the customer's case, since it helps establish whether the damage happened in transit or before shipping.

What should a retailer do if a photo looks recycled or AI-generated?

Compare the EXIF metadata against the claim history, check for inconsistent shadows or repetitive textures, and if doubt remains, request a fresh photo captured live through an in-app camera feature, which is harder to fake than a file sent by email.

How quickly must a customer report a damaged parcel?

Couriers typically require written notice within a few days of delivery. For a claim under the Consumer Rights Act's satisfactory-quality guarantee, the short-term right to reject lasts 30 days, after which the retailer gets one chance to repair or replace before a final right to reject applies.

Does the 30-day right to reject apply to marketplace sellers as well as retailers?

Yes — the Consumer Rights Act 2015 applies to any trader selling to a consumer, including third-party sellers on a marketplace, though the marketplace's own dispute process often runs in parallel with the statutory right.