Somewhere on your phone there is a photograph you took for no reason at all: a market stall in good light, a colleague laughing over tea, a street corner just after rain. It earned you nothing, and it never will while it sits in your camera roll. This guide explains how to sell stock photos online through the large stock agencies, what those agencies actually pay, and why the honest version of this business looks quite different from the way it is usually described.
The misunderstanding that wastes most beginners’ first month is simple. You do not sell the photograph. You keep it, you keep the copyright, and you rent out permission to use it, one licence at a time, potentially for years. Once that single point is clear, every other part of the process, from pricing to rejection letters, starts to make sense.
You License the Photo. You Do Not Sell It
A stock agency is a marketplace with a particular shape. On one side are buyers: businesses, bloggers, publishers, designers and marketing teams who need an image for a website, an advert, a book cover or a presentation, and who would rather pay a few dollars for an existing photograph than commission a shoot. On the other side are contributors, who upload photographs and describe them carefully. The agency sits in the middle, sets the price, handles the payment, and passes a percentage of each sale to the contributor. That percentage is the royalty, and it is the only money in this system that reaches you.
Most licences sold today are royalty-free. The term confuses people, because the buyer does pay: royalty-free means the buyer pays once for the licence and can then use the image repeatedly within the licence terms, without paying a fresh fee for every use. The older model, rights-managed, prices each licence according to the exact use: how large, how long, in which countries. Rights-managed sales are larger and rarer, and they survive mostly at agencies such as Alamy. For a beginner, royalty-free subscription sales will make up nearly everything, and they pay small amounts per download. The compensation for those small amounts is repetition. A single accepted photograph can be licensed hundreds of times over its life, by different buyers, without you doing anything further.
Two consequences follow from this structure. First, you do not set your prices at any of these agencies; the agency does, and your only real decisions are which agency to join and which photographs to upload. Second, because the same file can be licensed at Adobe Stock, Shutterstock and Alamy at the same time under non-exclusive terms, one good photograph is an asset that works in several shops at once.
This is also where stock photography parts company with selling your own products. If you create a template or an ebook and sell it through your own shop, you set the price and keep most of the revenue, but you also carry the whole burden of finding buyers. (I covered that model in my guide to selling digital products online, and it suits a different kind of work.) Stock agencies reverse the arrangement: they bring the buyers and set the prices, and in exchange they keep the larger share of each sale. Neither model is better in the abstract. They reward different things, and confusing one for the other leads to bad expectations in both.
Start With One Agency, Not Three
Beginners often open accounts everywhere at once, upload the same folder five times, and burn out on the repetitive admin before any pattern in their sales can emerge. My recommendation is to begin with one agency, learn its review standards and its keywording habits properly, and add a second only after the first portfolio is running. The three agencies worth considering first are these.
| Agency | Royalty for photos | How the rate moves | Payout threshold | AI-generated images | Best suited to |
|---|---|---|---|---|---|
| Adobe Stock | Flat 33% (35% for video) | No tiers, no annual reset | $25 | Accepted, if labelled as generative AI | Beginners who want simple, predictable rules |
| Shutterstock | 15% to 40% across six levels | Levels reset to 15% every 1 January | $25 | Not accepted from contributors | Volume: a very large buyer pool |
| Alamy | 20% for new contributors (Silver) | Bronze 15%, Silver 20%, Gold 40%, Platinum 50% | $75 in cleared funds | Not accepted | Editorial and specific subjects; higher price per sale |
Adobe Stock pays a flat 33% of the net price on photos, vectors and illustrations, and 35% on video. There are no tiers and no reset, which makes it the easiest structure to understand. Adobe’s own example explains the scale honestly: on a plan costing $29.99 a month for ten photo licences, the contributor receives about $0.99 per licensed photo. On the largest subscription plans the minimum royalty per download is $0.33 for most contributors, rising slightly with lifetime licence counts. Payouts start at $25. The full rate card is published on Adobe’s contributor royalty page, and it is worth reading before you upload anything.
Shutterstock runs the largest shop window of the three, and the most complicated pay structure. Contributors climb six earnings levels within a calendar year, from 15% on the first 100 image licences up to 40% beyond 25,000 licences, with a separate ladder for video. The catch is stated plainly in its help pages: every contributor drops back to Level 1 on 1 January, whatever they achieved the year before. A download can also pay as little as $0.10 on the biggest subscription packs, at any level. None of this makes Shutterstock a bad choice; its buyer volume is real, and frequent small downloads teach you quickly what the market wants. It does mean the headline 40% figure applies to very few people, and almost nobody in their first year.
Alamy requires a warning, because most older guides describe a company that no longer exists. Until recently Alamy was famous for paying contributors 40% as a matter of course. Its revised contract took effect on 1 September 2026 and replaced that with four tiers based on gross sales in a revenue year running from September to August: Bronze at 15% below $250, Silver at 20% between $250 and $3,000, Gold at 40% above $3,000, and Platinum at 50% only for exclusive content above $25,000. Every new contributor starts at Silver. So a beginner at Alamy earns 20%, not 40%, and the higher tiers are a reward for substantial sales rather than a starting condition. Alamy also pays out only once cleared funds pass $75. What Alamy still offers in return is a different kind of buyer: newspapers, publishers and picture desks looking for specific, editorial and hard-to-find subjects, where a single licence can be worth far more than a subscription download. If your camera roll is full of distinctive local places, events and everyday scenes that the big libraries lack, Alamy deserves a place in your plans, just with current numbers rather than the ones still circulating in old articles. Its contributor contract help page sets out the tiers in its own words.
My order of operations would be Adobe Stock first, for the flat rate and the clearest rules; Shutterstock second, once keywording feels routine, for volume; Alamy third, and mainly for editorial material. One agency, properly worked, beats three agencies filled in a hurry.
What Buyers Actually Pay For
Stock photography has a buyer’s logic, and it is not the logic of a photography competition. Buyers are usually illustrating something specific: a small business owner at work, a delivery rider in a real street, a family eating food that looks like food people actually eat, a nurse, a mechanic, a student with a laptop in natural light. The common thread is use. An image sells when a designer can picture the headline that goes with it.
Subjects with a steady commercial appetite include everyday work and small business scenes, food prepared and eaten in real settings, health and education, technology in ordinary hands, and people doing recognisable tasks with natural expressions. Local and regional scenes have genuine value too, precisely because the big libraries are saturated with the same few cities. A well-lit photograph of a bazaar, a rickshaw or a tea stall, keyworded properly, competes in a much smaller pool than another photograph of a Manhattan skyline.
What rarely sells is equally predictable. Sunsets, generic flowers, landscapes with no human or commercial hook, and holiday photographs taken for your own memories have almost no buyer behind them. They are pleasant, and the libraries already hold them in the millions.
On equipment, the honest answer is that a modern phone is enough to begin. Agencies judge the file, not the camera badge: sharp focus where it matters, clean exposure, no visible noise in the shadows, no heavy filters. A phone in good light produces files that pass review at all three agencies named here. A dedicated camera helps in low light and gives you more control, and it becomes worthwhile if the portfolio starts earning. Spending money on gear before your first hundred uploads, on the other hand, puts the cost before the evidence.
Getting Accepted, and Staying Accepted
Every upload passes a technical review before it can be sold, and rejections are a normal part of the process rather than a verdict on you as a photographer. The reasons repeat so often that they are worth memorising.
- Noise and grain, usually from shooting in poor light at a high ISO. Fix it by adding light, not by applying aggressive noise reduction afterwards, which reviewers can spot.
- Soft focus. The main subject must be sharp at full size. Check your files at 100% zoom before uploading; a photograph that looks fine on a phone may fail this test immediately.
- Logos and trademarks in frame. A branded shirt, a recognisable car badge or a shop sign can get a commercial photograph rejected, or pushed into editorial-only use. Remove the branding, change the angle, or accept the editorial route where the agency offers one.
- Over-editing. Heavy saturation, strong filters and artificial sharpening all read as manipulation at review size. A clean, natural edit survives review far better than a dramatic one.
- Similar images. Uploading ten near-identical frames from one burst invites rejection of the whole set. Choose the strongest one or two.
The practical habit that saves the most time is a two-minute inspection of every file at full size before it leaves your computer. Most rejections are visible at 100% zoom if you are willing to look.
Releases: the Paperwork That Protects the Sale
A release is a signed permission form, and two kinds matter here. A model release is required when a recognisable person appears in a photograph you want to sell for commercial use, meaning advertising and marketing. If a stranger’s face is identifiable and you have no release, the agency will either reject the file or restrict it to editorial use, where the image can illustrate news and factual content but cannot advertise anything. A property release covers recognisable private property in some situations: a distinctive home, a private interior, certain famous buildings. Public streets and general city scenes are normally fine without one.
This sounds more forbidding than it is. Each agency publishes its own release forms in the contributor area, and the process is usually a one-page document signed by the model, plus a witness signature at some agencies. The working rule is simple: photograph people you know, ask first, and keep the signed forms with the original files. Photographing strangers for commercial stock without a release is the one shortcut that can unravel later, because a buyer who uses an unreleased face in an advert carries a problem that leads back to you.
Keywording: the Unglamorous Craft That Decides Everything
Here is the part of this business that nobody puts in the highlight reel, and the part that decides whether a portfolio earns anything at all. A buyer finds your photograph by typing words into a search box. If your words do not match the buyer’s words, the photograph does not exist, however good it is.
The discipline is to describe the photograph the way a buyer searches, not the way a photographer admires it. A buyer types “small business owner counting cash register evening”, not “golden hour vibes at my uncle’s shop”. Useful keywords name the subject, the action, the setting, the number of people, the mood where it is genuine, and the concepts the image could illustrate, such as teamwork, deadline or savings. Most agencies accept up to 50 keywords; somewhere between 25 and 40 accurate ones is a sensible target. Accuracy matters more than volume. Keyword stuffing, adding popular terms that are not in the image, is against agency rules, damages your ranking when buyers ignore the file, and can end an account if it is done deliberately with misleading terms.
Titles deserve the same care. Write a plain, factual sentence that a search engine and a buyer both understand. Adobe Stock also uses your title and keywords to categorise the file automatically, so precision there does double work. Allow proper time for this stage. In my judgement, keywording is roughly half the labour of stock photography, and the contributors who treat it as an afterthought are the ones who conclude, wrongly, that nothing sells.
One related channel is worth a mention while you build. Photographers who also share their work publicly, on a blog or a visual platform, give their portfolio a second route to attention, and prints or commissions sometimes follow. If you are starting from zero on that side, my guide to Pinterest for beginners covers how visual search traffic works and how to set up properly.
Where AI-Generated Images Fit, and Where They Do Not
The rules differ sharply by agency, and getting this wrong wastes uploads at best. Adobe Stock accepts AI-generated images, provided each file is submitted with the ‘Created using generative AI tools’ option ticked, and provided prompts, titles and keywords do not name real artists, real people, fictional characters or brands. Shutterstock’s contributor policy page states that it will not allow AI-generated content to be submitted by contributors for licensing, even though Shutterstock sells its own AI image tool to customers; the two things are separate, and contributor uploads of AI work are refused. Alamy does not accept AI-generated images either. Getty Images and iStock take the same position as Shutterstock.
Policies in this area change faster than anywhere else in the industry, so check the contributor page of the agency in question on the day you upload rather than relying on any article, including this one. The rule that does not change is honesty: never submit AI work as a camera photograph anywhere. Detection is good, accounts are closed for it, and a closed account takes the whole portfolio with it.
Honest Expectations, Since Nobody Else Will Give You Any
Stock photography is a slow business at the start, and I would rather state that plainly than dress it up. On subscription plans, a single download commonly pays somewhere between a few cents and a dollar or two. Some licences pay more, particularly at Alamy, but they arrive less often. A new portfolio of twenty photographs may sell nothing for weeks, not because the photographs are poor, but because twenty files in a library of hundreds of millions is a very small signal. The realistic shape of this work is a portfolio that compounds: each accepted, well-keyworded photograph adds a small permanent chance of income, and the total becomes noticeable only when the portfolio reaches the low hundreds of files and has had months to be found.
It follows that this is a poor choice for anyone who needs money this month, and a reasonable one for anyone who already takes photographs and can treat uploading as a steady habit alongside other work. Many contributors pair it with faster-earning uses of the same skills while the portfolio builds: offering photography or design services directly, for example, through the route I described in my Fiverr beginner’s guide, pays for the same skills on a much shorter timeline. The two sit comfortably together, because the portfolio grows in the background while services pay in the foreground.
Judge the experiment properly when you run it. Ten uploads prove nothing in either direction. A hundred properly keyworded photographs, given a few months, will tell you whether your subjects have a market.
What I Would Do in the First Week
If I were starting this today, I would keep the first week deliberately small. Open one Adobe Stock contributor account. Go through the phone and the hard drive and select the twenty strongest photographs with a clear subject and a plausible buyer, not the twenty prettiest. Inspect each at full size, fix the fixable problems, and reject your own weak files before a reviewer does it for you. Upload, then spend as long on the titles and keywords as you spent on the selection. After that, upload in small regular batches and build towards one hundred files before drawing any conclusion.
Who should try this: anyone who already photographs regularly, enjoys the editing and describing as much as the shooting, and can wait months for modest returns that grow slowly. Who should skip it: anyone looking for quick income, anyone who dislikes repetitive desk work, and anyone whose camera roll is mostly private family photographs they would not want licensed to strangers. For the right person, the appeal is real. The photographs already exist. The work is turning them into properly described, properly licensed assets, once, and then letting the library do the selling.