Technical video rejections come from five places: the codec, the frame rate, conversion artefacts, the file size ceiling and stray audio. The accepted lists differ by agency, so one export can pass on one and fail on another.
Commercial rules say describe the scene and keep real names and trademarks out. Editorial rules say name the person, the place and the date, exactly. The same frame needs two metadata sets.
Releases rarely fail because the model objected. They fail on formal details: an undated signature, the photographer acting as witness, one form covering several people, or pages digitally stitched together.
VectorStock wants an EPS and a matching RGB JPEG preview, and approved artists can send whole batches over FTP in a single ZIP. Here is the structure, the size caps and the mistakes that stall a batch.
Clearing review only means an asset is allowed in the catalogue. Being findable is a separate problem with its own causes: indexing lag, keyword order, missing filter attributes and editorial status each hide a file in a different way.
Motion Array publishes both a tag limit and a list of tag habits its reviewers push back on. The gap between the two is where most contributors lose search visibility without ever being rejected.
When an agency rewrites its metadata or content policy, the files already in your portfolio do not get re-reviewed, but they do get judged by the new rules the next time anything touches them. Here is how to audit a library against a change you did not make.
Motion Array publishes its upload rules in full, so almost every rejection maps to a line you can check before you submit. Here is the list, category by category, with the fix for each.
Transfer and submission fail differently. An FTP client can resume a half-sent file safely, but submitting the same file twice earns an Identical Submission or Approved on Site rejection. Resuming a batch means reading state from each agency, not re-running the job.
Motion Array publishes its metadata rules per content type rather than once for the library. The hard limits are a 40-character graphics title, a 25-word graphics description and a flat ban on accented and non-Western characters, with tag counts left unstated.
Shutterstock wants English, Adobe Stock wants your account language, and neither says much about accents. Here is what the guides state, where encoding breaks, and how to choose.
Storyblocks does not publish a per-download rate. It pays a share of a Member Library Revenue Pool on or about the 15th, net of third-party fees, with a threshold that depends on your payout method and a separate six-monthly payment for machine learning use.
Keywords rarely vanish because a platform rejected them. They vanish because a cap was hit, a comma split a value, or a field was overwritten between your editor and the agency. Here are the published limits and the five ways a list gets shortened.
Storyblocks splits uploading in two: a web uploader capped at 100 files and an FTP route for larger batches. The credentials differ, release forms cannot travel over FTP, and untagged content is deleted after 30 days.
A wrong category is rarely a mistake you made; it is a default you did not check. Adobe suggests one automatically, Shutterstock blocks until you pick, and both accept a batch with the first guess intact. Here is what each platform lets you change, and when.
Canva publishes far fewer keyword rules than most stock platforms: metadata is mandatory, elements need at least five comma-separated keywords, and titles must be unique. Everything else is judged months later, through the Creator performance review.
An FTP batch that dies halfway usually fails in one of five places, and only one of them is a network problem. Here is how to tell them apart before you re-send anything.
Canva pays creators monthly through a provider called Trolley, but only once a verified payout method, an accepted tax form and a balance over the minimum all line up. Here is what each of those actually requires.
Metadata that vanishes between your catalogue and the agency almost always traces to one of five specific causes, each with a named setting behind it. Here is how to identify which one you are hitting.
Canva takes uploads through the Elements Hub with a CSV manifest inside the collection folder, not over FTP. That changes how you name, stage and batch files across a multi-agency run.
Agencies flag keyword stuffing on relevance and repetition, not on a count. The exact wording from Shutterstock and Adobe Stock, the repetition their own examples permit, the limits that do exist, and where each keywording tool stops.
Santa Studio is a new product from Rastock AI: type in your child's name, their proud moment, their pet and the gift they hope for, approve every line, and Santa records a vertical HD video in a few minutes.
Every field Canva asks a contributor to fill, what is required versus merely preferred, the five-keyword floor, the file name match rule, and the two agreement clauses that change how you write metadata in the first place.
Similar-content rejections follow published limits, and the limits differ by agency. What Adobe Stock, Shutterstock and Storyblocks actually allow, which variations never count as different, and the pre-upload pass that prevents the loop.
What Storyblocks actually enforces on keywords and titles: the 6 to 100 range, the 15 to 30 recommendation, single-concept terms, the filename check, English-only metadata and the trademark rule.
A retail aisle is a trademark minefield before it is a keyword problem. How to describe shopping scenes in the words buyers search without naming a single brand.
Storyblocks keeps two rejection lists and a set of hard technical specs. Which one your file hit decides whether the fix is a re-export, a reshoot, or nothing at all.
Shelves, signage and packaging put a trademark in almost every retail frame, and agency standards say titles and keywords can never carry one. An honest comparison of the tools, and where each one stops.
Titles of 16 to 140 characters, six keywords minimum and 100 maximum, English only. The limits are the easy part. The harder part is that Storyblocks reads metadata from three different places.
Pond5, Shutterstock, Adobe Stock, Storyblocks, Artgrid and Envato Elements pay for footage in three different shapes. A comparison built only from what each platform publishes itself.
Shutterstock reviews content daily but publishes no turnaround target. What it does publish - queueing by media type, a 21-day submission deadline, and five reasons a file stays pending - explains most of the wait.
Science and lab frames fail metadata review for predictable reasons: brand names on the equipment, claims the picture cannot prove, and a literal layer that is too thin. Here is the order that survives review.
Canva does not publish per-file rejection codes. Here is what its Contributor Agreement and Quality Standards actually say about why content is refused, removed or re-reviewed.
Science and lab content is a permissions category before it is a metadata one: almost every real lab is identifiable private property, and Adobe Stock requires a property release for recognisable private interiors. Xpiks, StockSubmitter, Wirestock, PhotoTag.ai and Rastock AI compared for where each one stops.
Envato Elements has no per-download rate. Authors get 50% of net revenue on Individual and Teams plans, 25% on Enterprise, split by each subscriber's item point usage. Two authors with identical download counts are paid differently. The formula, the per-seat rule, the $50 threshold and the 15th-of-the-month calendar.
Event and concert photography splits into two routes with different metadata rules. Here is how to keyword each one, from the editorial dateline to the conceptual layer buyers search.
Envato publishes its tag limits, formatting constraints and word blacklist. Here is every keywording rule that gets enforced at review, plus the IPTC fields Envato actually reads.
Event and concert content is mostly editorial, and that decides your tooling more than any feature list does. An honest comparison of the desktop apps, managed pipelines and metadata tools, plus where each one stops.
Envato delivers photos over FTP or through Portfolio Manager, reads your embedded IPTC and XMP fields in a fixed priority order, and deletes FTP uploads after 24 hours. Here is the whole pipeline from export preset to review queue.
Remote work frames are bought for the situation they illustrate, but reviewed against what is literally visible. Here is how to build both keyword layers without crossing the contextual metadata rules at Adobe, Shutterstock and Envato.
Envato's reviewers work from a published list: focus, noise, exposure, trademarks, added graphics, editorial content and near-identical batches. Here is each reason, and the soft-versus-hard rule that decides whether you can try again.
Remote work is one of the most saturated briefs in stock, which makes metadata the deciding factor rather than the photography. Xpiks, StockSubmitter, Wirestock, PhotoTag.ai and Rastock.ai compared on what each is actually for, with prices from each vendor's own pricing page.
Getty Images announced a prepackaged Chapter 11 restructuring on 12 October 2026, backed by up to $533 million from the Getty Family and Koch Equity Development. What it means for contributor royalties, exclusivity and iStock, and the records worth gathering now.
Green and ESG buyers search for an idea, not an object — but agency metadata rules require that your keywords describe what is actually visible. Here is how to satisfy both.
Envato Elements reads your title, description and tags out of embedded metadata before a reviewer ever opens the file. Here are the exact limits, the blacklisted words and the specs photos have to meet.
Sustainability imagery is searched by abstraction and shot as objects, which makes metadata the whole job. Six options compared on what they actually do: AI metadata, multi-agency delivery, revenue share and pricing model, with the claim rules that get green files rejected.
Vecteezy runs two unrelated payout models. Free pays a flat $5 per 1,000 downloads; Pro pays 50% of one subscriber's net billing cycle divided by their download count, so the same download can be worth a hundred times more or less. The formula, the $25 threshold and what Vecteezy does not publish.
A ceramic bowl on an oak board has no story to describe. It has a material, a surface, an angle, a quality of light and an amount of empty space, and all five of those are what buyers type.
Vecteezy's pipeline has four states and one decision that locks your metadata for the rest of the review cycle. Here is the whole path, and the four places files quietly stall without telling you.
One product session yields forty near-identical frames, and every agency wants them described differently enough not to read as duplicates. How Xpiks, StockSubmitter, PhotoTag.ai, Wirestock and Rastock AI handle that, and which half of the work each one actually solves.
Vecteezy weights your first five keywords, wants one singular form of each word, bans content-type and AI terms, and locks keywords once you submit. Here is every rule, including the two keyword caps Vecteezy publishes that do not match.
A macro keyword set has to answer a question the picture does not: what the buyer will call this. The four layers that do it, the species-name decision, and which tools actually help.
Vecteezy sorts rejections into ones you can correct and ones that are final. Here is the full list for photos, vectors and video, and what each one actually asks you to change.
In macro the first tool that matters is a focus stacker, not a keyworder. Nine routes to market compared on what each one actually fixes, and where each one stops.
Shutterstock meters submissions in a rolling seven-day window, so the upload workflow is a budgeting problem as much as a technical one. The limits, the two routes in, and the order that avoids rework.
In a night frame the light is the subject, and the words that describe it best are the ones agencies treat as technical noise. How to name the light and the mood without reaching for ISO numbers or gear.
Vecteezy caps keywords at 30, wants titles in title case under 100 characters, and does not index titles or descriptions at all. Here is every field, the rules behind it, and the four-column CSV that carries them.
In low light the first tool that matters is a denoiser, not a keyworder. Seven routes to market compared on what each one actually fixes, and where each stops.
Pond5 treats exclusivity as a video question and nothing else. The rates by media type, the $25 threshold, the two-day dashboard lag and the payment method that costs 4%.
Nobody licenses a finance photo for the hand holding the card. They license it for the idea it stands in for. How to name concepts that are true of the frame while stripping the brands the frame is full of.
Transferring the file to Pond5 is the easy part. Processing strips the embedded metadata, the file stops in Needs Edits, and nothing enters review until you submit it by hand. Here is the whole sequence.
A drone shot is bought for where it is and what it can be used for, not for the objects in frame. How to keyword aerial photos and footage so buyers and reviewers both find them.
Adobe holds royalties until five conditions are true at once: a US $25 balance, a first sale 45+ days old, a linked payment provider, verified contact details and a validated tax form.
Aerial work turns a metadata problem into a transfer problem. Six routes to market compared on what each is genuinely good at, and where each one stops.
Two official Pond5 pages give two different keyword counts, and neither is dated. Here is what the documentation does agree on, and how to work with the part it does not.
A gradient has no subject to name, so literal keywording collapses. Abstract and background files sell on material, mood, layout use and colour instead.
The payout side of 123RF is written into the contributor agreement, not the dashboard. Thresholds vary by payment method, payment is automatic once you clear one, and a missing tax document costs 30%.
In wedding work the release decides what your keywords are allowed to claim, and tradition and moment carry the searches that remain. Here is the order that reaches buyers and the vocabulary that differentiates a file.
Pond5 separates returned files from rejected ones, and the difference decides whether you edit metadata or re-export the clip. Here is what each status means and the technical list behind most outright rejections.
Buyers in this category search by breed and species, not by mood. Here is the keyword order that reaches them, the hierarchy Adobe publishes, and the release rules that cap what your text may claim.
123RF runs three separate FTP hosts with their own storage quotas, and files do not enter review until you process the folder from inside your account. Here is the whole sequence, with the limits that decide it.
Abstract and background files have no subject to name, which makes automated keywording either the most useful thing in the workflow or the most damaging. Six tools compared on that specific problem.
123RF rejections fall into three groups: files that fail an automatic check, metadata that is incomplete, and frames the agency cannot license at all. Two of the three are fixable across a whole batch.
Finance frames are conceptual, brand-dense and full of numbers on screens. Six tools compared on the three things that actually matter for this niche: metadata quality, policy filtering and whether the file reaches the agency.
123RF's keyword minimum is seven and its own guide calls thirty to fifty ideal. The gap between those numbers matters less than four rules that get files pulled, and one that makes every mistake permanent.
Automotive frames are the hardest category for brand rules, because the trademark is bolted to the subject. Here is how to keyword cars, EVs and commercial transport so buyers find them and reviewers do not pull them.
Pond5 enforces three metadata fields and leaves eleven optional. Here are the exact limits, the field names its CSV uploader expects, and the keyword count its own documentation contradicts itself on.
Agriculture buyers license by crop, task and season, not by scenery. How to build a keyword set that names the right things, and why the brand on the tractor has to come out.
123RF will not review a file that arrives without adequate metadata, and the text cannot be edited after acceptance. The fields, limits and banned terms, taken from the platform's own documentation.
Wedding content is a releases problem before it is a keywording problem, and most tool comparisons skip that entirely. Here is what each of the main tools does for this category, where each one stops, and which setup fits which contributor.
Depositphotos takes files over FTP and over the browser, but neither route attributes them for you. Here is the published file ceiling, the approval gate that comes first, and the step where batches go quiet.
Pet and animal stock lives or dies on breed and species naming, and almost no AI tool gets that right unaided. Here is what each of the main tools does for this category, where each one stops, and which combination fits which contributor.
Dreamstime takes files over plain FTP, then parks them in an Unfinished section until you finish the metadata. Here is the whole sequence, the connection settings, the CSV ordering rule and the limits the platform actually publishes.
In agriculture the metadata problem is identification rather than description: the crop has a name, the machine has a trademark, and a generic model gets both wrong. A comparison of the tools contributors use for farming content, and where each one stops.
In fashion and beauty, the logo on the jacket and the label on the bottle decide whether a file is commercial, editorial or rejected. How to build a keyword set that sells the mood without naming a brand.
In automotive and transport work the badge, the plate and the logo decide whether a file is commercial or editorial. A comparison of the tools contributors use to keyword and deliver these files, and where each one stops.
Depositphotos allows 50 keywords per file and recommends using half that. It also splits phrases into separate terms, which changes how you should write them. A walkthrough of the platform's own keywording guidance.
Fashion and beauty is the category where release paperwork, near-identical frames and colour-critical retouching collide. A comparison of the tools contributors actually use, including where each one stops being the right answer.
Construction and industry images sell to a narrow, well-funded set of buyers who search with trade vocabulary the rest of the stock market never uses. This is how to keyword the category so those searches actually land on your files.
Construction and industrial shoots produce huge, repetitive batches. The tool that helps is the one that keeps machine names, safety terms and release status consistent across a whole folder, then delivers everywhere you sell.
FTP on Freepik is not a setting you switch on. It opens after 500 published files, and a monthly ceiling still applies on top of it. Here is the workflow in the order the platform enforces it.
Education is the one stock niche where the legal layer decides whether the keyword layer gets a chance. Buyers search in institutional language, releases block more files than technical faults do, and the difference between preschool and kindergarten is a real sales difference.
Freepik publishes a keyword range, a sorting rule, a singular-form rule and a list of practices it treats as spam. Most contributors know one of the four. This is the enforced version, taken from the contributor documentation rather than from forum habit.
Education stock is gated by release paperwork and searched by age band, setting and activity. Six tools compared on what they actually do - Xpiks, StockSubmitter, Wirestock, PhotoTag.ai, Bridge and Lightroom, and Rastock AI - plus the keyword axes buyers use.
Most Freepik rejections are decided by an automatic pass before a reviewer ever sees the file. Here is every published threshold - format, dimensions, size, duplicates, metadata, AI tagging and video specs - and the export-side fix for each one.
A server rack photographed well and keyworded literally will sit unsold next to a worse image that named the concept. Technology is the niche where the gap between what is in the frame and what the buyer is searching for is widest, and where keyword sets date fastest.
Freepik, now running its AI products under the Magnific name, requires titles and keywords to live inside the file rather than in an upload form. Here is the full field reference, the CSV alternative, and the metadata mistakes that cost contributors approvals.
Technology shoots break metadata tools in a specific way: the subject is visually generic and the buyer searches for an abstraction. Here is what each of the main tools does about that.
Dreamstime attaches the earnings level to the file rather than to the contributor, which is why two photographers with identical portfolios can be paid very differently on the same day.
Buyers search lifestyle and family images by relationship, activity and feeling, not by object. Here is the keyword structure that matches how they actually look.
Depositphotos sets royalties by contributor level, and the level is decided by download count with subscriptions counted at one third. Here is what each tier actually pays.
Almost every usable lifestyle or family frame contains a recognisable person, which makes release handling as decisive as keyword quality. Here is what each of the main contributor tools does well, where each one stops, and how to choose between them.
Buyers licensing sports and fitness images search for endurance, recovery and routine, not for dumbbells. Here is how to build a keyword set that covers the activity, the body, the setting and the concept without tripping editorial or release rules.
Architecture is the stock niche where the keyword that makes an image findable is often the same keyword that makes it editorial. Here is how to handle that.
Alamy accepts uploads over FTP or through its web tool, with specific file specs and a folder structure that decides how your images are handled. Here is the route end to end.
A single sports shoot produces files that belong in two different licensing buckets. The tool you use has to keep them apart, and most of them handle that differently.
Alamy splits metadata into a caption and a tag list, and its search engine treats them very differently. Here is what each field does and where contributors lose visibility.
Nature is the most crowded category in stock, which means a landscape file is not competing on beauty but on specificity. Here is the four-layer method, the accuracy problems that sink these files, and what each tool can actually do.
Alamy runs a four-tier commission system that recalculates every September, and the tier you land in changes your rate by more than three times. Here is how the tiers, the revenue year and the clearing delay actually work.
Travel is the one category where the most valuable keyword is something no AI can see in the pixels: the name of the place. Here is the four-layer method, the geography mistakes that get files removed, and what each tool can and cannot do.
Alamy names its rejection reasons precisely, checks a random sample of every submission, and fails the whole batch when one file falls short. Here is the full list and what each one actually means.
Healthcare files need the clinical term and the human one in the same keyword set. Here is the four-layer method, the words that get medical images rejected, and an honest comparison of the tools that can do it at volume.
Alamy only makes five fields mandatory, but those five carry all the search weight. Here is the full field reference, the 50-tag and 10-supertag rules, and how to send everything through IPTC instead.
Business and office is the most saturated category in microstock, which makes keywording the main lever a contributor still controls. A layered structure, the terms that bury a file, and an honest look at the tools.
A field by field reference for Getty Images submissions: what each metadata field is for, how embedded IPTC relates to what you type in the portal, and the technical requirements that sit alongside it.
Food is one of the most competitive stock categories, and generic keywords bury good images. Here is the five-layer approach buyers actually search with, how Adobe and Shutterstock differ on limits, and which keywording method fits libraries of different sizes.
Rastock AI has a stand at Baku ID, 17–18 September at SABAH.city — the regional startup and investment festival for the Caucasus, Central Asia and MENA. Bring your own files and watch a real batch go from bare images to described, keyworded and delivered; investors, both founders are there for two full days.
Adobe Stock offers three genuinely different upload paths, and the one you pick decides how big a batch you can handle. Here is how the portal, Creative Cloud publishing and SFTP differ, plus the file rules, CSV limits and submission caps behind them.
Architecture shoots break stock tooling at a different point than most genres. The frame is usually clean; the problem is rights, vocabulary and near-identical sets. Here is how six tools handle that, and where each one stops.
Depositphotos states four decline categories and its inspection team published twenty-two specific notes. Together they explain almost every refusal, and most are avoidable before you upload.
Nature work breaks metadata tools in a specific way: the hard part is not describing the frame but naming the species and the place correctly. Here is how six tools handle that, and where each one stops.
Dreamstime validates title length, keyword count, categories and description before a reviewer ever opens your image. Here is every field in one place, plus the CSV and FTP sequence that silently breaks large batches.
Travel shoots break keywording tools in a specific way: the hard part is not describing the frame but getting the place, the country and the licensing status right. Here is how six tools handle that, and where each one stops.
Depositphotos spreads its metadata rules across six separate help pages, so most contributors meet the limits only after a rejection. Here is every field in one place, with the numbers taken from Depositphotos' own documentation.
Healthcare imagery is judged on clinical plausibility and terminology long before it is judged on light. This compares the metadata and delivery tools contributors actually use, and is honest about the part none of them solves.
Dreamstime has no single rejection-reasons page, but its own photo editors have published the verbatim refusal texts. Here they are, grouped by cause, with what each one actually means and how to fix it before resubmitting.
Business and office imagery fails moderation for trademark and release reasons more often than for technical ones. This is an honest comparison of the tools contributors actually use for corporate content, including what each one does badly.
Getty Images does not offer contributors FTP or SFTP. This is what the real submission path looks like: ESP batches, the mobile app, file and pixel limits, keyword rules that lock after acceptance, and the queue behaviour that quietly delays reviews.
Food is the stock category most likely to break agency rules on brand names and packaging. Here is how the main metadata and delivery tools handle it, with their real pricing and limits.
Getty Images and iStock pay a share of what the customer paid, but the share depends on the site, your agreement, the file type and the license model. Here is how the 2026 numbers work.
Bridge and Lightroom Classic both write IPTC metadata, but they suit different stages of a stock workflow - and neither one delivers files to agencies. An honest comparison, including the tools that pick up where both stop.
A portfolio that passes review but never earns is failing at one of four specific points. Here is how to tell which one, and why fixing them in the wrong order wastes months.
StockSubmitter is Windows-only and priced by submissions. Here are five alternatives compared on what each one actually automates, how each charges, and where each one hands the work back to you.
Neither Adobe nor Shutterstock publishes a review time. Here is what their own contributor documentation says, why two contributors wait very differently, and the one action that turns a slow review into a dead file.
Four pricing models, six tools, and the fees that only appear once you are at volume: per-agency counting, renewals behind one-time payments, and tier-gated file limits.
Getty publishes plenty of contributor advice, but only a few metadata rules carry rejection, suspension or account closure. Here is what those rules actually say.
Generating good keywords and knowing what a particular agency will accept are two different problems. Here is which tools actually encode per-agency rules, and which leave that judgement to you.
Most Getty Images rejection reasons are not reviewer taste. They come from published specifications that a file either meets or does not, and almost all of them can be checked before you upload.
Both tools write titles, descriptions and keywords. They part company on what happens next: one embeds the metadata and hands your files back, the other carries them through to the agencies. Compared honestly, alongside the rest of the category.
Shutterstock pays a percentage that climbs through six levels, but the price the customer paid decides what that percentage is worth. Here is how the 2026 rates, the $0.10 floor and the monthly cycle actually fit together.
Contributors keep asking whether they can A/B test titles and keywords on Shutterstock or Adobe Stock. No agency offers it. What you can run is a matched before-and-after test, and most of the difficulty is in the controls rather than the tooling.
Shutterstock documents its keywording rules in public, yet files still get rejected for metadata every day. Here is which rules are checked by machine, which are judged by a reviewer, and which are enforced months after publication.
Most contributors decide on the sticker price and then wonder why the maths never worked. Here is the actual calculation, plus how the pricing models of the main stock workflow tools differ in ways that change the answer.
Adobe publishes its keywording rules openly, yet contributors lose files to them every week. Here is which rules a moderator enforces, which ones the search engine enforces silently, and how to tag so you survive both.
Editing keywords across thousands of published files is allowed almost everywhere. The risk is not permission, it is doing it in a way you cannot undo or measure.
Freepik gates almost everything behind one hard rule: your first submission has to be 150 to 200 files in the same format. Here is what the program asks for, level by level.
Both promise fewer hours on uploads, but they sit on opposite sides of one line: who owns the agency relationship, and who takes a cut of the royalty.
Selling the photo is the easy half. Thresholds, waiting periods, tax forms and transfer fees decide when the money actually lands, and how much of it survives the trip.
A side-by-side look at Xpiks and Rastock AI: a licensed desktop application versus a hosted metadata and delivery service. What each one is built to do, where the pricing models diverge, which other tools belong in the same conversation, and the cases where each is the wrong answer.
A field-by-field reference for Shutterstock metadata: what the description box expects and why two different character limits exist, where the keyword floor and ceiling sit, how categories work for photos versus video, and the two extra elements editorial submissions require.
Accuracy is the wrong first question for confidential files. Here are the seven data-handling questions that decide whether an AI metadata tool belongs anywhere near unreleased work, and how the main tool categories compare on each.
123RF is non-exclusive, free to join and pays 30% to 60% depending on your rolling twelve-month sales. Here is the application process, the file and keyword rules it actually enforces, and how the eight commission levels work.
StockSubmitter and Rastock AI both move finished files to stock agencies, but they solve different halves of the job. A side-by-side look at coverage, pricing models, credential storage and metadata.
Dreamstime allows up to 80 keywords per file and gives you no way to mark which ones matter. Here is what its published rules actually require, what it refuses, and how to keyword for it at volume.
An overnight batch of a few thousand files fails for boring reasons. Here is what each agency's own documentation actually says about limits and ingestion, and which six tools hold up when the run is large.
Vecteezy pays two different ways depending on the licence attached to each file: a flat rate per thousand free downloads, or a share of subscriber revenue. Here is how the programme works, what it refuses, and how to apply.
An honest look at what a contributor with a few hundred images actually needs. Most of the tools marketed at this stage solve a problem you do not have yet, and the free options are genuinely good.
Canva runs three separate creator programs with three different entry routes, and one of them is currently closed to new signups. Here is what each actually requires and how the royalty pool pays.
Every stock agency wants the same information in a different shape. This is how contributors keep one source of truth and map it onto Adobe, Shutterstock, Getty and the rest without retyping anything.
Getty Images reviews your portfolio before it gives you an upload account. Here is what the application checks, the file specs enforced at upload, and the generative AI rule that catches most applicants off guard.
An honest comparison of the tools contributors use to re-optimize an old stock portfolio, how to decide which files are worth the effort, and the agency rules that limit what you can change after approval.
A working reference to the reasons Shutterstock declines files, which rejections you can correct and resubmit, which ones are final, and how to stop the same reason repeating across an entire catalogue.
Deduplication is a one-time repair and normalization is a standing rule. Most tools do the first well and the second badly, which is why libraries drift back within a year.
Shutterstock's FTPS settings, the file rules that quietly fail files after transfer, and why the metadata step is still waiting for you when the upload finishes.
A stock tool can reach your agency accounts three different ways, and the difference matters. Here is what an agency-issued SFTP credential actually grants, why a full account login is a different level of exposure, and how the main contributor tools compare.
A field-by-field reference for Adobe Stock metadata: what the title field expects, how many keywords you get, why the first ten matter most, how categories and language work, and which technical limits sit behind it all.
Some contributor tools charge a licence fee. Others take a percentage of every sale for as long as the file exists. A comparison built from each vendor's own published terms, including what rights you actually grant.
The programme allocates $100,000 in Microsoft Azure resources. It is cloud capacity rather than cash, and it goes into the compute that reads files and writes metadata.
The filename is the only piece of metadata that follows your file into every agency's CSV, release matcher and FTP directory. Here is a naming convention that holds up across ten agencies.
Cross-agency distribution is allowed and expected. Re-sending a file to the same agency, or shipping near-identical frames in bulk, is not. Here is how to keep the two apart.
Exposure, white balance and setup choices decide whether a commercial stock photo passes technical review — and whether a designer can actually use it.
Old portfolios accumulate keyword debt: copy-pasted sets, terms that meant something in 2019, translations that now hurt. Here is what the agencies actually let you edit after publication, which tools do bulk cleanup, and how to sequence the work.
Adobe counts megapixels, Shutterstock counts megapixels but caps megabytes separately, and Alamy ignores both in favour of uncompressed file size. A practical look at the current numbers and the single export window that satisfies all of them.
Google Cloud Vision and Amazon Rekognition label images accurately and cheaply, but neither produces an agency submission. Here is what separates a general-purpose vision API from a tool built for stock contributors, and when each one is the right choice.
A field-by-field reference to every meta tag that can attach to a picture, and which of them stock agencies and search engines actually read.
Between 5,000 and 20,000 files, the tool that wins is rarely the one with the best features. It is the one whose pricing model does not punish your volume. Here is the arithmetic.
A print on demand marketplace sells a physical product with your image on it. A stock agency licenses the image itself. The two pay on completely different logic, and the work each rewards is different too.
Nothing on the market scans your drives and delivers to agencies in one pass. Here is how local indexing tools, AI culling tools and stock workflow tools actually divide the job, and where each one stops.
Adobe Stock pays every contributor the same flat rate. Shutterstock pays a rate that climbs all year and then resets to the bottom each January. Which one pays more depends on your volume, not the headline percentage.
Indonesia has one of the largest and most active stock photography communities in the world. We are opening one exclusive ambassador seat for it — 10% recurring commission, monthly payouts, and a direct line to the team building the tool.
AI content tagging is the automated process of reading an asset and writing the metadata record that makes it findable. Here is how the pipeline actually works across a mixed photo, video and vector library, where it breaks, and what governance you need before you let it run on thousands of files.
Most photographers do not fail at taking pictures. They fail at building more than one way to get paid for them. Here is how the income layers stack in 2026 - and which one you can start today.
Most digital asset management software was designed for agencies and brand teams, not for someone shipping 300 files a week to ten marketplaces. Here is what each of the main options actually does, where it stops, and how to pick.
Image tagging decides whether a good photo earns money or disappears. Here is how AI tagging actually works, the four failure modes worth checking, and a sixty-second review pass that keeps quality high at volume.
Adobe wants a title under 70 characters. Shutterstock allows 2048 and requires at least seven keywords. Here is how to keep one metadata source of truth and let software reshape it per agency, plus an honest comparison of the tools that do it.
The Magnific rebrand rattled a lot of contributors. Here is what actually changed in the agreement, what the license lets the platform do with your files, how the per-download math works, and how to stay safe on any agency.
Vertical video is a separate deliverable, not a crop. Here are the exact resolutions Adobe Stock and Shutterstock accept, the frame rate and codec limits, the keywording exception almost nobody knows about, and an honest read on where buyer demand actually sits.
A small studio moving thousands of files a month needs metadata consistency across shooters and delivery that does not eat evenings. Here is how the real options — integrated platforms, desktop uploaders, managed distributors and keyword-only AI — actually compare.
True stock is licensed for commercial use and needs releases; editorial documents real life and cannot sell products. Here is how the two licenses differ, when a photo can cross over, and what each means for your metadata and earnings.
Microstock is the volume end of the stock photography market: low prices, royalty-free licensing, and earnings built on downloads rather than exclusivity. Here is how the model actually works, what contributors earn, and how to start without the usual false starts.
Hand-keywording stops working somewhere around a few hundred files. Here is what full-time contributors actually run to keyword thousands of images: the batch tools compared honestly, when each one fits, and the pipeline that holds up at scale.
A photo tag is only as useful as the metadata standard carrying it — this guide breaks down IPTC, XMP and EXIF and shows exactly how Adobe Stock, Shutterstock and Getty/iStock read them. It also covers the export settings that quietly strip keywords before a file ever reaches an agency.
Tagged keywords only work if they're structured in the order and format each agency's search engine expects. This guide breaks down keyword order, limits, and common structuring mistakes across Adobe Stock, Shutterstock, Getty, Dreamstime, and Vecteezy.
No stock agency currently publishes true keyword-to-download attribution, so contributors have to piece the answer together from dashboards, exports and workflow tools. This guide compares StockSubmitter, Xpiks, Wirestock, Microstock+, PhotoTag.ai, Shutterstock's own dashboard and Rastock on exactly what each one can and cannot tell you.
A practical, agency-by-agency walkthrough of how to sell photography as stock: what you need before you apply, how each platform reviews you, and how to keyword and upload without losing hours to spreadsheets.
Choosing good site keywords for photography is less about finding words and more about deciding which pages deserve to exist. Here are the three tests every keyword should pass, the four layers a photography site needs, and how to map them without competing against yourself.
Automatic upload to multiple stock agencies works by generating metadata once and pushing the same files via FTP/SFTP to each agency's contributor account in parallel. This guide compares StockSubmitter, Xpiks, Wirestock, PhotoTag.ai, Microstock+, and Rastock honestly, including where each one falls short.
Most AI keywording tools read a single frame and describe a photo. Footage needs motion, duration and camera movement in the metadata. Here is which tools actually cover both, and where each one stops.
A realistic map of where remote photography jobs actually exist in 2026 — from job boards and freelance editing gigs to stock agency royalties — with honest earnings numbers, not hype.
Contributors who store their portfolio on a NAS or server drive need software that can read, tag, and upload directly from network storage without forcing a local copy first. Here is a neutral, tool-by-tool breakdown of what actually works.
A practical, source-checked walkthrough of how to sell stock photography in 2026 — from setting up contributor accounts to understanding royalty rates and avoiding the metadata mistakes that keep beginner portfolios invisible.
An odd number of subjects reads as more natural than an even number — and buyers search for it that way too. Here is how the rule of odds works, when to ignore it, and how to carry it into your metadata.
Generative AI images have to be labeled twice: once in the agency's submission form and once inside the file itself. Here is what each major agency requires, what the IPTC standard says, and which tools handle it across a whole batch.
Keyword advice for artists is usually written for one marketplace and applied everywhere. This breaks down what works on handmade marketplaces, print-on-demand sites and stock agencies - and why the same tag list cannot serve all three.
No stock tool retrains an AI on your personal portfolio. Here is what "learning from past uploads" actually means across Xpiks, StockSubmitter, PhotoTag.ai, Microstock+, Wirestock and Rastock AI - and the one feedback loop that genuinely works.
Several tools now read an image and write a submission-ready stock title. This compares what each actually produces, how much a title influences agency ranking versus keywords, and where generated titles still fail moderation.
Photographers get keyword advice written for e-commerce and wonder why it does not work. This guide separates the three keyword problems a photographer actually has - portfolio site, stock portfolio, local business - and gives a workable method for each.
Metadata rejections are avoidable and mostly mechanical: trademarks, duplicate keywords, mixed languages, missing editorial dates. Here are the published rules at Adobe Stock and Shutterstock, and an honest look at which tools enforce them before you submit.
Image keywords for SEO are not the same thing as stock agency keywords. Google reads alt text, filenames and page context; agencies read an embedded keyword field. Here is what each system actually uses, and how to keep one set of metadata serving both.
Categories and signed releases are the two parts of stock submission that resist automation. Here is which tools actually carry them through a bulk pipeline, which only write metadata, and how to split the work.
The rules of composition are not laws — they are descriptions of how human vision behaves. Here is what each one does, when breaking it is the better call, and the composition habits that decide whether a stock file licenses or sits unseen.
Very few products do both upload and sales tracking properly. A neutral comparison of StockSubmitter, Microstock+, Microstockr, Wirestock, Xpiks and Rastock AI - what each covers, what it costs, and the point at which you need a second tool.
Full-time contributors do not use one app - they run a pipeline: ingest, metadata, review, multi-agency delivery, sales tracking. A fair comparison of the six tools that cover those stages, what each is best at, and where each falls short.
Shooting is the easy part. This guide covers what the major agencies actually pay in 2026, why files get rejected, and how to build a metadata and multi-agency upload workflow that does not eat your week.
Six tools can take an entire folder of photos and auto-generate titles, descriptions and keywords in one batch. A fair comparison of Rastock AI, PhotoTag.ai, Wirestock, StockSubmitter, Xpiks and Microstock+ — what each is best for, how it charges, and where it falls short.
A free tag generator for stock photos turns hours of manual keywording into seconds of automated work. Here's how the AI reads your images, produces titles and keyword tags, exports a ready-to-use CSV, and where the free versions help — or fall short.
The hub for our workflow guides. Every stage from selection to delivery, what breaks at each one, and which parts are worth automating once volume grows.
The hub for our agency guides. Every major stock platform compared on royalties, review standards, metadata requirements and the AI policies that diverged sharply in 2026.
The tools stock contributors actually use to generate keywords, what each is genuinely good at, and where each one stops working. No single tool wins every case — the right choice depends on your volume, your media types and whether you need distribution as well as metadata.
Adobe Stock pays contributors a flat 33% on photos and 35% on video, but the real dollars-per-download swing with the buyer's plan. Here are the 2026 rates, what contributors actually earn, and how to lift your average.
The Shutterstock CSV template lets you attach metadata to hundreds of files at once instead of typing it into the web form. Here are the exact columns, the formatting rules, and the errors that get files rejected.
Manual keywording costs 5–20 minutes per image and gets worse as you tire. An AI metadata generator for stock photos does the same job in seconds per file — here's how the technology works and where it still needs your judgment.
A data-backed guide to choosing the best keywords for stock photos in 2026: buyer-intent tagging, how many keywords to use, what to avoid, and the trends driving downloads now.
A breakdown of every common Adobe Stock rejection reason — from focus and noise to IP, metadata, and AI content — with a clear fix for each so you can raise your acceptance rate.
A free metadata generator for Adobe Stock can turn hours of manual tagging into seconds of automated work. Here's exactly how these tools analyze your images, produce titles, descriptions, and keywords, export a ready-to-use CSV, and where the free versions help — or fall short.
Metadata is what makes your stock photos findable — and sellable. This complete 2026 guide covers how to write titles, descriptions, and keywords that rank in agency search, how many keywords to use, IPTC embedding, and how AI tools can automate the entire process.
Becoming a VectorStock contributor is free — passing review is the hard part. Here is the signup process, the EPS pre-flight check that prevents structural rejections, and how to keyword illustration work so it sells.
Vecteezy is unusual in running a free tier and a paid tier at the same time. Understanding how each pays is the difference between building an audience and giving work away.
Uploading through a browser is fine until it isn't. Here is how to prepare and deliver Vecteezy batches properly, and why metadata has to be right before transfer rather than after.
Storyblocks runs on unlimited subscriptions, so contributors are paid from a shared pool rather than per licence. That single fact changes which content is worth uploading.
When you shoot a sequence, buyers often want several clips from it — but inconsistent metadata means they find one and never see the rest. Here is how to tag a series so it holds together.
Freepik is one of the few large agencies that accepts AI content from any generator. The catch is the labelling requirement and a pay-per-download model where volume is the only lever that matters.
Shutterstock's category assignment is the field contributors most often get wrong, and it is one of the few that cannot be fixed with a CSV afterwards. Here is how the taxonomy, keyword limit and editorial rules fit together.
Pond5 lets contributors set their own prices, which makes it unusual among stock marketplaces. It also means your metadata has to reach the right buyer before pricing matters at all.
PIXTA's buyers have visual expectations that differ meaningfully from Western stock. Understanding those conventions matters more than any keyword tactic.
MotionArray deals in templates and project files as much as footage, which introduces a metadata problem stock libraries do not have: software compatibility, version and editability all determine whether a buyer can even use your asset.
Most keyword advice optimises for the wrong thing. Buyers rarely search for what an image shows — they search for what they need it to do.
The Adobe Stock web uploader is fine for ten files and painful for a thousand. FTP is the supported route for bulk submission — here are the credentials, the settings, and the errors that make files disappear.
Getty and iStock run the strictest content rules in stock photography and the most complex royalty structure. Here is how the 2026 rate card works, why external AI is banned outright, and what metadata the review team expects.
EyeEm built its audience around photography that does not look like stock. Understanding what buyers come there for is the difference between licensing regularly and uploading into silence.
Phone cameras are good enough for stock now, but the failure points are different from a dedicated camera — computational processing, missing metadata and file handling all catch people out.
Envato Elements pays contributors from a subscription pool rather than per sale, so item usage drives everything. That rewards a specific kind of asset — and penalises another.
Dreamstime's review is stricter than most microstock agencies and its community-driven standards catch things automated checks miss. Here is what actually gets rejected and how to raise your approval rate.
Depositphotos sells across dozens of countries, which makes its search behave differently from single-market agencies. Here is how royalties and review work, and why translation-aware metadata matters here more than elsewhere.
A designer searching for assets is matching an existing brief — style, mood, format, compatibility. Tagging by subject alone misses most of how they actually search.
Creative Market is a marketplace of independent shops rather than a stock library. That means presentation, bundling and pricing carry weight that keyword optimisation alone never will.
Canva pays on usage rather than downloads, and its users search inside a design editor rather than a stock library. Both facts change what good metadata looks like.
Manual keywording degrades measurably after about twenty images. Here is the batch workflow that keeps quality flat across thousands of files, and where a human still has to intervene.
Alamy's amended contributor contract takes effect on 1 September 2026. AI-generated images are banned outright, the commission tiers have been restructured so Gold now requires $3,000 in annual sales, and the payout threshold rises to $75.
Metadata is three different standards doing three different jobs, and stock agencies only read some of them. Here is what each one holds and which fields decide whether your work is found.
123RF has a stronger Asia-Pacific buyer base than most Western-focused agencies, which changes which images sell and which keywords find them. Here is how royalties, review and search work.