Photo Evidence Checklist for Customer Support Teams

A photo evidence checklist for a support team should require three elements in the same set of images: the packaging with a legible shipping label, the product in the condition it arrived, and the defect in close-up next to a scale reference such as a ruler or coin. Without all three in original image files — never a screenshot — an agent has no reliable basis to approve a refund, replacement, or reopened case.

This article is provided for informational purposes and does not constitute legal advice. Regulatory references are accurate as of the publication date. Consult a qualified professional for decisions affecting your business.

What a photo actually needs to prove before an agent can act

A single close-up of the defect proves nothing on its own — it could belong to any other item, including one pulled from a reverse image search. An agent can only resolve a case reliably if the submitted images answer three questions together: is this the product that was ordered, what condition did it arrive in, and which specific order does this image belong to.

Requiring all three elements before any refund is approved mechanically cuts the volume of unfounded claims, because the burden of proof discourages opportunistic requests — a pattern documented by ReferralCandy's guide to damaged goods return policies. The logic is simple: a genuine customer loses nothing by documenting a case properly, while a share of fraudulent claims stop the moment proof becomes mandatory.

The checklist by claim type

A single generic checklist does not work well across claim types: the photos that matter for transit damage are not the ones that prove a wrong item was shipped, and a hidden-defect claim needs an extra piece of evidence.

For non-conformity claims specifically, the case usually rests on more than photos alone — invoice, delivery note, proof of purchase — which our document verification guide covers in more depth as a complement to the visual evidence.

The UK legal framework support agents should know

In the UK, the Consumer Rights Act 2015 requires goods to be of satisfactory quality when they reach the buyer, and Section 29 places responsibility on the retailer for the condition of goods until they are in the customer's physical possession. Within the first 30 days after delivery, a customer can reject faulty goods for a full refund; between 30 days and six months, the retailer must offer a repair or replacement first, and only after that attempt fails does the right to a refund or price reduction apply. A photo taken on the day of delivery, with the phone's native timestamp preserved, is what turns a verbal dispute into a documented one.

The House of Commons Library briefing on faulty goods, digital content and services summarises these thresholds for anyone drafting internal support policy, and Section 23 sets out the specific remedies — repair, replacement, or price reduction — a retailer must offer depending on how long after delivery the defect is reported.

Why disputed photos are getting harder to trust

An agent who receives a photo has historically had no simple way to check whether the image genuinely shows the customer's parcel, whether it was cropped to hide something inconvenient, or whether it came from an entirely different case. Three patterns recur constantly in support team write-ups:

1. The recycled photo — the same image of a broken product circulates across several claims, sometimes weeks apart, as covered in our complete guide to photo evidence in e-commerce claims.
2. The screen recapture — a genuine photo from a different dispute is displayed on a screen and rephotographed, which destroys the original metadata; see the limits of EXIF metadata on customer photos.
3. The AI-generated or edited photo — a defect that never existed is digitally added to a real photo of the product, or the whole image is synthetic.

The National Retail Federation's 2025 return fraud data found that 9% of all product returns in the US are fraudulent, amounting to roughly $76 billion in annual losses (NRF, 2025 Retail Returns Landscape). A structured photo checklist does not eliminate this risk, but it raises the cost of pulling off a fraudulent claim enough to filter out a meaningful share of attempts.

Consumer-grade image generation tools can now produce a convincing fake defect on a product in seconds, with no editing skill required. CheckFile offers AI-generation signal analysis as a complement to existing controls, flagging the inconsistencies typical of synthetic content — repeated textures, inconsistent shadows, unusual compression artefacts — on attachments submitted to support a claim. This is not a replacement for human review; it gives that review signals it cannot detect unaided at scale.

Building the checklist into the support workflow

A checklist that lives only in an internal wiki page does not help an agent working through a ticket backlog under time pressure. Three practical changes make the difference:

• Block case creation until the required photos are attached — a mandatory field on the claim form, rather than a follow-up request the agent has to remember to send.
• Use a canned response that lists the exact angles and elements expected, cutting the back-and-forth that otherwise stretches resolution time.
• Set an escalation path for when multiple suspicion signals stack up — a recycled photo flagged, inconsistent EXIF data, a repeat pattern on the same account — instead of approving or rejecting on a single agent's judgment call.

Our e-commerce document verification page covers this in more depth for support teams handling a high volume of claims, and the security page explains how uploaded images are processed without being retained longer than necessary.

What support teams actually ask on seller forums

On Amazon Seller Central, eBay Community, and Shopify Community, a handful of questions come up repeatedly and deserve a direct answer rather than a policy statement.

How many photos should we require, and in what format? Marketplaces typically cap attachments at a handful of images — often five — in JPG or PNG, with no PDFs or oversized files accepted. For an internal support desk, three photos cover most cases: the sealed box, the opened box, and the defect close-up with a scale reference.

Can we reject a case based on a photo that just looks suspicious? Not reliably, no. A photo that looks recycled or inconsistent should be treated as a signal to verify, not an automatic verdict — the same principle our analysis of claim photo review at scale sets out for insurers: cross-check several signals before deciding, rather than relying on one agent's visual judgment alone.

Frequently Asked Questions

What photos should a support team require before refunding a damaged item?

At minimum three images: the sealed box with the shipping label, the opened box showing the interior packing, and the defect in close-up next to a scale reference. All three should come from the same device and remain uncropped.

Is a screenshot of a photo acceptable evidence?

Not reliably. A screenshot strips the original EXIF metadata and can hide a crop or a recapture from a different case; always request the original image file instead.

How long does a UK customer have to report a delivery issue?

Under the Consumer Rights Act 2015, a customer can reject faulty goods for a full refund within 30 days of delivery. Between 30 days and six months, the retailer must first offer a repair or replacement before a refund applies.

How do you spot a recycled photo from a previous case?

A visual or hash-based comparison against the history of past claims catches the most obvious repeats. Beyond a small volume of tickets, this check becomes impractical to do manually and needs a dedicated tool.

Should AI-generated photos be handled differently?

Yes, where possible. These images often show no inconsistency visible to the naked eye, but they can carry artefacts detectable through dedicated generation-signal analysis, used to complement — not replace — human review.