Key takeaways
- The 10,000-subscriber gate is gone. Since March 2026 the Shopping affiliate programme is open to every creator in the Partner Programme, which now includes the 500-subscriber tier.
- It is not available everywhere. The programme runs in roughly fifteen markets, and the retailer list is different in each one — the UK got six retailers in August 2026, Amazon arrived in the United States the same month.
- There is no YouTube commission rate. Each seller sets its own rate and its own attribution window per item, and returns reverse the commission after the fact.
- A product tag is a packaging change, not just a revenue setting. It puts a shelf under your player and, on a Short, a movable sticker over your composition.
- Auto-tagging writes tags on your behalf using a model that reads your title, description and captions, and converts description links from participating sellers into tags. Treat what it produces as a draft to review, not a result.
- Commission sits at the end of a five-term chain that starts with an impression and a click. The two terms your thumbnail controls are the two that multiply everything after them.
For three years the Shopping affiliate programme was something other creators had. It sat behind a 10,000-subscriber wall, which meant the channels most in need of a revenue line that is not advertising were the ones locked out of it. That wall came down in March 2026, and almost nobody noticed, because the announcement went out as a Help Centre community post rather than as a keynote.
What replaced it is more interesting than a lower threshold. Over six months YouTube added two Latin American markets through a single retailer, opened the United Kingdom with six, and put Amazon's catalogue in front of creators in the United States. It also started letting a model tag your products for you, and started applying disclosure labels to videos where it thinks you forgot.
Most of what has been written about this treats it as a monetisation switch: turn it on, tag things, collect money. That framing hides the part that decides whether any of it works. A product tag changes what your video looks like, changes what it has to promise, and pays out at the end of a chain of multiplied percentages where the first two terms have nothing to do with commerce at all.
What the programme actually is
The Shopping affiliate programme lets you attach a specific product listing to a specific piece of your content — a long-form video, a Short, or a live stream. A viewer who taps that tag lands on the retailer's own site. If they buy within that retailer's attribution window, you earn a commission on the sale.
Two things follow from that description and are worth stating plainly. The first is that you are not selling anything: there is no inventory, no fulfilment and no customer service, because the transaction happens on somebody else's site. The second is that you are not paid for the click. You are paid for a purchase that happens elsewhere, later, by somebody you cannot see, and which can be undone by a return weeks after it lands in your dashboard.
That makes affiliate income structurally different from every other line on a YouTube channel. Advertising pays on impressions you can count. Memberships pay a fixed monthly figure. Affiliate revenue is the only line where the paying event happens off-platform and can be reversed.
What changed in 2026, in order
The programme has been running since 2022, but 2026 is the year it stopped being a large-channel feature. The sequence matters, because each step changed a different constraint.
| When | What changed |
|---|---|
| 27 March 2026 | Eligibility untied from a subscriber count and tied to Partner Programme membership instead, opening it to the 500-subscriber tier. The previous 10,000-subscriber requirement was removed. |
| 18 June 2026 | Mexico and Argentina added through a single retail partner, Mercado Libre, taking the programme to fourteen countries. |
| 6 August 2026 | The United Kingdom opened as the fifteenth market, with six retailers — Wayfair, Currys, Debenhams, Boots, Marks & Spencer and Etsy — and Awin as the affiliate platform partner. |
| 27 August 2026 | Amazon joined in the United States. Eligible US creators can tag Amazon products in Shorts, long-form videos and live streams. |
Alongside those, YouTube began testing something quieter: showing the same creator-tagged product sourced from a different eligible local retailer when the viewer is in another country. More on that below, because it has consequences most creators would not predict.
Who can switch it on
Three conditions have to hold at once, and creators routinely satisfy one and assume the rest.
You are in the Partner Programme. That is the gate now. The lower Partner Programme tier — the fan-funding tier — asks for 500 subscribers plus either 3,000 valid public watch hours in the previous twelve months or three million valid Shorts views in the previous ninety days, along with three valid public uploads in the last ninety days and a channel in good standing. If you are already taking Super Thanks or running memberships, you are on the right side of this line. Note that the higher tier's requirements are changing in 2027, which is covered in the piece on the 2027 monetisation requirements; the affiliate programme sits on the near side of that change, not the far side.
Your channel is in a supported country. This is the condition that stops most people, and it is not negotiable by appeal. The programme runs in the United States, the United Kingdom, Japan, Korea, India, Indonesia, Malaysia, the Philippines, Singapore, Taiwan, Thailand, Vietnam, Brazil, Mexico and Argentina. Europe outside the UK, Canada and Australia are not on that list. YouTube's overview and eligibility page is the authority on the current set, and it changes several times a year.
Your channel has no hate speech strike. YouTube names this one explicitly and separately from the general good-standing requirement.
Amazon adds a fourth condition on top in the United States: an active Amazon Influencer or Associates account, linked to the channel. Amazon tagging is not something the Shopping tab grants you by itself.
What a product tag does to the video
Here is the part the monetisation guides skip, and the reason this is a packaging decision rather than an accounting one.
On a long-form video, tagging produces a shopping affordance below the player — a label that expands into the list of everything you tagged — and, on eligible videos, a scrollable row of products underneath. None of that touches the thumbnail. It touches what happens after the click, which is where your retention lives.
On a Short, it is different, and considerably more intrusive. The tagged product appears as a product sticker rendered over the video itself. By default it sits in the bottom-left corner, it shows the first product in your tag list, and it can be moved and resized in the YouTube mobile app. That is not a detail. The bottom-left of a Short is contested territory: it is where the channel handle, the caption and the sound attribution already live, and it is inside the region a viewer's thumb passes through. A sticker dropped there without thought will land on a face, a hand, or the one word of on-screen text the Short depends on.
The sticker is a composition problem you can solve once
If you tag products on Shorts regularly, decide where the sticker goes before you shoot, not after you upload. Frame the subject to leave the lower-left third genuinely empty and the sticker becomes a design element instead of an obstruction. The same reasoning that governs where a subject sits in a 16:9 frame applies here — see thumbnail composition for why edges belong to the interface rather than to you, and the Shorts thumbnail guide for what the vertical frame already has to carry.
YouTube says its own testing found Shorts product stickers drive materially higher click-through than a conventional shopping button, which is exactly what you would expect from a control that sits inside the frame rather than beside it. The corollary is that a sticker is doing that by taking attention from something. What it takes it from is your decision.
The chain that ends in a commission
Almost every guide to this subject starts at the product click, which is the wrong place to start. A commission is the last term in a chain, and every term multiplies the ones before it.
| Term | What moves it |
|---|---|
| Impressions | The algorithm, your topic, your catalogue's momentum |
| × Click-through rate | Thumbnail and title |
| × Watching to the product | Hook, structure, whether the product actually appears |
| × Product click rate | Verbal mention, timestamp, sticker placement, buying intent |
| × Purchase rate | Price, retailer, the viewer's day |
| × (1 − returns) | Category — apparel and electronics are the worst offenders |
Two of those six terms are packaging and they sit at the front, which means they scale everything downstream. A video with a 3% click-through rate and a 4% product click rate earns the same as a video with a 6% click-through rate and a 2% product click rate — but the first one is a fixable thumbnail and the second is a fixable placement, and only one of those two problems is visible in your Shopping analytics. If you do not know where your channel sits on the first term, what a good click-through rate actually looks like is the place to calibrate before you touch anything commerce-shaped.
The practical consequence is unglamorous: for most channels, the highest-leverage change to affiliate revenue is not tagging more products. It is making the video that already has good tags get clicked more often.
The commission is not a rate you can look up
Search for "YouTube Shopping commission rate" and you will find confident tables of percentages by category. Treat them the way you would treat a table of average membership conversion rates: somebody's estimate, dressed as a benchmark.
There is no YouTube commission rate, because YouTube is not the party paying it. Each seller sets its own rate and its own attribution window, per item. That information is visible to you inside Studio in the affiliate sellers and offers list, which is the only figure that means anything for a product you are actually about to tag. Two items on the same retailer's site can carry different rates, and the same item can carry a different rate in a different month.
Two further mechanics deserve more attention than they get. First, attribution windows are short and vary by retailer, so a viewer who bookmarks the tab and buys next month may earn you nothing. Second, commissions are held before payout and reversed when a product is returned — which means a strong month in a high-return category is provisional in a way ad revenue never is. Anyone planning around affiliate income should treat the dashboard figure as an upper bound, in the same spirit as the gap between Studio's estimate and the bank in the piece on what YouTube actually pays per 1,000 views.
Auto-tagging: what you are agreeing to
Auto-tagging is the change with the largest gap between how it is described and what it does. Enabled from Studio's advanced channel settings, it identifies products in your videos and tags them on your behalf. Three mechanisms sit under that sentence.
- A model reads your video's title, description and captions and matches what it finds against the affiliate product catalogue.
- Link-based detection converts description URLs from participating sellers into native tags.
- Automatic timestamps identify where in the video a tagged product is referenced and mark that point.
The first of those is the one to think about. A model reading your captions does not know the difference between the camera you recommend and the camera you spent four minutes explaining you would not buy. It matches mentions, not verdicts. On a comparison video — the format most likely to carry tags in the first place — that distinction is the entire editorial value of the video.
None of this is an argument against the feature. Link-based detection alone is worth it for a channel with a back catalogue full of description links, since YouTube's own January 2025 experiment in the United States found that videos with product tags and timestamps enabled saw around 43% more product clicks than videos relying on description links alone. It is an argument for treating auto-tagging as a draft. Studio surfaces recently auto-tagged videos under the Shopping section of the Earn tab, and that review queue is the whole safety mechanism.
Disclosure got partly automatic
An affiliate commission is a material connection between you and the seller under the FTC's endorsement guides, and disclosing it is not optional because the tag looks self-evident. It does not look self-evident to a viewer who has never seen a product tag before.
YouTube handles the platform side through the paid product placement, sponsorship and endorsement setting, which produces the "Includes paid promotion" notice at the start of playback. YouTube updated the terminology around that policy in 2026 and added something new: if its systems detect branded content in a newly uploaded video that you did not disclose, it may apply the disclosure label on your behalf. The label arrives whether or not you asked for it.
The instinct is to see that as a cost — a banner over the opening seconds of a video you spent a week on. The research does not obviously support the fear. A study of sponsorship disclosure in short-form video, published in Computers in Human Behavior, found that a distinct disclosure lasting six seconds at the start of a sponsored video raised viewers' sponsorship literacy and their attitudes toward both the brand and the video. Comprehension, in other words, was worth more than concealment. That finding is about sponsored short-form video rather than affiliate tags specifically, and it should not be stretched further than it goes — but it argues against the reflex of hiding the relationship, and it argues for putting the disclosure where a viewer will actually parse it rather than burying it in line nine of the description.
YouTube's documentation on paid product placements is the reference for which arrangements require the setting. The FTC's requirements are separate, apply regardless of what YouTube's toggle does, and are not satisfied by the toggle alone.
Your tag may not point where you think
The localisation experiment is the most consequential thing in this article that has not yet fully shipped.
When a viewer in another country meets a tagged product, YouTube has been testing substituting the same product sourced from a different eligible local retailer — chosen on relevance to what you originally tagged, reliability of the retailer, whether local pricing is consistent with regional expectations, and whether the item is in stock and shippable to that viewer. For a channel with a genuinely international audience, this is straightforwardly good: a viewer in Jakarta is not going to complete a purchase from a US retailer that will not ship to them.
It also means the destination of your tag is not fully yours any more. Notably, the test excludes videos where the Paid Product Placement setting is enabled — which is the correct carve-out, because a sponsored placement that silently redirected to a competitor's stockist would be a contractual problem, not a convenience. If you run sponsored content and affiliate content on the same channel, that distinction is now doing real work and is worth understanding before it applies to you.
The broader pattern here is familiar to anyone who has watched YouTube automate the parts of a video that scale. Audio is dubbed automatically now; product destinations are being localised automatically. The one asset that never localises itself is the artwork, which is the argument made at length in the piece on auto-dubbing and localised thumbnails.
What YouTube's published numbers do and do not say
Two figures are circulating and both come from YouTube's own experiments, which is worth stating before either is used to justify a decision.
In an experiment run between 14 and 29 July 2026, YouTube reported that videos using Shopping product tags alone recorded more than 110% more product clicks than videos relying only on description links. The January 2025 US experiment mentioned above produced the 43% figure for tags plus timestamps against description links alone.
Neither result has published methodology — no sample size, no category breakdown, no statement of how videos were matched. Both compare a native in-player control against a text link buried below the fold, which is close to the easiest comparison a platform could construct in its own favour. The direction is almost certainly right; the magnitudes should not be treated as forecasts for your channel.
What genuinely has no source is the claim, repeated across a dozen guides, that tagging products improves your recommendations. YouTube does not document a ranking effect for Shopping features in either direction, and no first-party data supports one. Tag products because a viewer wants to buy the thing you are holding, not because you have been told the algorithm rewards it.
Which videos should carry tags, and which should not
The temptation after switching this on is to tag everything, which auto-tagging makes trivially easy. Resist it on three grounds.
- Intent has to already exist. A viewer watching a review is shopping. A viewer watching a vlog is not. Tagging a hoodie in a vlog produces a shelf nobody opens and a sticker over your face.
- Tagged products date faster than videos do. An evergreen tutorial with fifteen tags is fifteen listings that will go out of stock, get discontinued or shift price while the video keeps earning views. A dead tag is worse than no tag.
- Every tag is a promise the video has to keep. If the product is tagged, it should appear, be used, and be talked about. The tag is a claim about the content, and mismatches between the shelf and the video are exactly the sort of thing that gets said in comments.
The formats that repay tagging are the obvious ones and they repay it heavily: reviews, comparisons, hauls, setup tours, "what I use" videos, tutorials where a specific tool is required, and unboxings. What those share is not a category — it is a viewer who arrived already deciding whether to buy something.
A packaging checklist for a shoppable video
Everything above collapses into a short list of decisions taken before publish rather than after.
- Decide what the thumbnail promises. A shoppable video still lives or dies on the first term of the chain. If the thumbnail's job is to establish that this is a real test of a real product, the product should be legible in it at feed size — which is a size worth checking rather than assuming, in the thumbnail preview tool or anywhere that renders it small.
- Say the product out loud. A tagged product that is never verbally referenced relies entirely on the viewer noticing a shelf. Naming it is free.
- Use timestamps. The whole point of a native tag over a description link is arriving at the moment the viewer is looking at the thing.
- Order the tag list deliberately. On a Short, the first product in the list is the one that becomes the sticker. That is a design decision disguised as a sort order.
- Place the sticker before you shoot. Leave the lower-left of the vertical frame empty.
- Review what auto-tagging did. Especially on comparison videos, where a model matching mentions can tag the product you argued against.
- Disclose clearly and early. The evidence points towards comprehension helping rather than hurting.
The part that does not change
The Shopping affiliate programme is a genuine improvement to the economics of a small channel. It unlocks at the same tier as memberships, it requires no inventory, and after March 2026 it is available to a class of creator who previously had advertising or nothing. If you are in one of the fifteen markets and you make anything product-adjacent, it is worth switching on this week.
But it does not change the shape of the problem. Commission sits at the end of a chain whose first two terms are an impression and a click, and no amount of tagging improves either of them. The channels that will earn from this are not the ones that tag the most products; they are the ones whose videos get opened in the first place, and which then happen to have a well-placed tag waiting inside. Commerce is a multiplier on attention you already have.
Which is why the work that pays here is the same work that paid before it existed. If the thumbnail is the term you want to move, Thumblore generates and iterates on them quickly enough that testing two real options costs an afternoon rather than a week. And if you are about to attach a commercial layer to a Short, read the Shorts thumbnail guide first — the vertical frame was already crowded before anyone added a sticker to it.