A company Wikipedia article without a logo reads as unfinished. The infobox — the card in the top-right corner — is the most-looked-at part of the page, and an empty image slot is the first thing a visitor registers. The effects reach beyond the article: Google's Knowledge Panel frequently sources company logos and images from the Wikipedia–Wikidata–Commons ecosystem, and a complete infobox makes a page read as established rather than provisional. So the instinct to add a logo and a few photographs during page creation is correct.
The execution is where companies get burned. Images go through a copyright review that is stricter, faster, and more automated than anything that happens to article text: files with the wrong license, or no proof of permission, are typically tagged within days and deleted within a week or two. Deletion is not even the worst outcome — a cluster of polished marketing uploads from a new account draws scrutiny to the account itself, and from there to the article.
Commons only accepts images anyone can reuse — commercially, forever
Wikimedia Commons — the media library serving Wikipedia in every language — accepts only free content: files in the public domain or under a free license such as CC BY or CC BY-SA. "Free" has a precise meaning: anyone, anywhere, may reuse the file for any purpose — commercial use included — and may modify it, with nothing owed beyond attribution. The release is irrevocable. Non-commercial and no-derivatives licenses are not accepted at all.
Make that decision consciously. Putting the five-figure brand shoot on Commons is a donation, not a placement: a competitor may legally use those images the day after. The North Face learned a related lesson in 2019, when its agency swapped images on articles about outdoor destinations for product-placement shots to climb Google Images; the edits were reverted and publicly condemned by the Wikimedia Foundation. Upload only what you can afford to give away, and only where it genuinely illustrates the subject.
One consolation: a free license releases copyright, not trademark. Commons hosts thousands of trademarked logos under a trademark notice, so a free release does not weaken protection against impersonation or confusing commercial use.
Logos: simple marks are easy, complex marks take a different door
Logos split into two legal regimes, and the split decides the route.
Simple text logos — a wordmark in standard or modestly styled type, perhaps with basic geometric shapes — generally fall below US copyright's threshold of originality. They cannot be copyrighted at all and can go to Commons at full resolution under the PD-textlogo tag, trademark notice alongside. This is the best case, with a bonus: a Commons-hosted logo can be attached to your company's Wikidata item, one of the channels Google draws on for Knowledge Panels. The caveat: the threshold varies by country, Commons expects the file to be free in both the US and the country of origin, and borderline marks get debated and sometimes deleted.
Complex logos — mascots, illustrations, gradients, anything with visible creative authorship — are copyrighted works, and Commons will not host them unless you free-license the artwork itself, which almost no brand should do. The standard route is non-free use (fair use) on English Wikipedia directly, bypassing Commons. The conditions are narrow: a single low-resolution copy (around 0.1 megapixels is the working rule), used only in the infobox of the article about the company, with a written non-free use rationale. Upload the plain mark, not the version with the tagline. Two consequences follow: if the article is ever deleted, the orphaned logo is removed automatically about a week later; and because non-free rules are set per language edition — German Wikipedia accepts none at all — a fair-use logo never reaches Commons, Wikidata, or other language versions.
Photos: the photographer owns the image, not the person in it
The most misunderstood fact in this area: copyright belongs to whoever took the photograph — not the person in it, and not the company that paid for the shoot, unless a written agreement transferred it. Your CEO does not own their own headshot; the photographer or studio does. Most agency contracts license usage rights, not copyright.
Hence the most common corporate deletion: a marketing employee uploads the press-kit portrait as "own work." A polished studio image, a days-old account, and an own-work claim form a pattern reviewers recognize instantly. The file is tagged as lacking evidence of permission and, unless proof arrives within about a week, deleted.
What does work:
- Genuine own work. The person who actually took the photo uploads it — the staffer who shot the office on their phone owns that image. A true selfie is the one clean case where subject and photographer are the same person.
- Copyright transfer in writing, after which the company can release the file itself.
- A release from the photographer through the workflow below.
The VRT workflow: how permission is actually proven
VRT — the Volunteer Response Team, formerly OTRS — is how copyright holders prove a release. The sequence:
- Identify the real copyright holder. Photographer, studio, or the company if a written transfer exists.
- The holder emails a declaration of consent to permissions-commons@wikimedia.org from a verifiable address — the studio's own domain, not a personal inbox. The email names the exact file, states the specific license (CC BY-SA 4.0 is typical), and acknowledges that the release is irrevocable and permits commercial reuse and modification.
- A volunteer verifies and stamps the file with a ticket number. Expect days to several weeks of backlog; the file may carry a pending-permission tag in the meantime.
- The file is now defensible. A VRT ticket is treated as settled evidence in any later dispute.
From running this repeatedly: collect the release before the article goes live, bundle all files into one ticket, and if the photographer hesitates at "irrevocable commercial reuse" — many reasonably do — commission a small shoot with copyright transfer written into the contract, which is cheaper than fighting deletions afterward.
Previously published images and the precautionary principle
If the image has already appeared on your website, social channels, or press kit without a free-license notice, reviewers will assume it is copyrighted — because it almost always is. Uploads are routinely reverse-image-searched, by humans and by bots; a match on an all-rights-reserved page produces a no-permission tag or an immediate copyright deletion. Commons runs on a precautionary principle: significant doubt that a file is free means deletion, and the burden of proof sits with the uploader.
Two clean ways through: publish the free license at the source — a CC BY-SA notice next to the image on your own site, which reviewers can verify — or run the VRT release. And do the reviewer's check yourself first: the uniqueness pre-check in our Wikipedia image checker reverse-searches the file the way a Commons reviewer would, so you find the prior publication before they do.
The route map: what survives review
| Asset type | Route | Survives review? | Notes |
|---|---|---|---|
| Simple text logo (wordmark, basic shapes) | Commons, PD-textlogo + trademark tag | Usually | Below the threshold of originality; feeds Wikidata and Knowledge Panels; borderline marks get debated |
| Complex logo (mascot, illustration, gradient art) | Fair use on English Wikipedia only | Yes, within its limits | One low-res copy in the company article only; deleted if orphaned; never reaches Commons or Wikidata |
| High-quality executive portrait (agency shoot) | Commons via VRT release | Only with VRT | Photographer or studio holds copyright, not the executive; "own work" claims fail fast |
| Office and building photos | Commons as genuine own work, or VRT | Usually, if actually own work | Useful sparingly; watch for copyrighted art or signage in frame |
| Product shots | Commons via VRT, if at all | Often deleted | Promotional framing draws scrutiny; packaging artwork can itself be copyrighted |
The deletion reasons we see most often
Company images die for a short list of reasons:
- "Own work" that isn't. The press photo uploaded by marketing, tagged for missing permission, gone in a week.
- Previously published with no license at the source. Found by reverse search, deleted as a copyright violation.
- Fair-use logos used beyond their one permitted article — or orphaned when the article is deleted, then removed automatically.
- Watermarks and promotional framing. Commons treats watermarked corporate imagery as advertising, not encyclopedic media.
- Derivative works. A photo of your lobby is also a photo of the copyrighted sculpture in it; packaging artwork carries its own copyright.
- Threshold miscalls. A complex logo uploaded under
PD-textlogoinvites deletion — and a closer look at everything else the account has touched.
That last clause is the real cost: image mistakes direct attention to the account and the article, joining the broader pattern we cataloged in Wikipedia mistakes companies make.
Our own order of operations: text logo to Commons first, wired into Wikidata; an executive portrait only with a VRT-backed release; office photography only where it earns its place; product shots usually left out — the article is an encyclopedia entry, not a brochure. If the page already exists and just needs its infobox completed, the same rules apply as part of standard Wikipedia editing work.
Planning to add a logo or photos to your company's article? Run the files through our free Wikipedia image checker first — it maps each asset to its licensing route and reverse-searches for prior publication before a Commons reviewer does.