How Merch by Amazon Tier-Ups Actually Work
Merch by Amazon tier-ups aren't about total volume — they reward consistent sales spread across designs. Here is the tier ladder, …
Trending designs sell fast and die faster; evergreen niches sell slowly and keep paying for years. Most sellers never split their weekly design output between the two on purpose, and that's the mistake this comparison fixes.
Every seller looking into amazon merch on demand niches eventually hits the same fork: build a catalog of designs that sell steadily for years, or chase whatever phrase, meme, or holiday is spiking in search volume this week. Both are legitimate strategies. Neither is automatically better, and the sellers who struggle are usually the ones trying to do both without a plan.
An evergreen niche is a topic people search for year-round regardless of news cycles: nurse humor, dog mom, camping and RV life, teacher appreciation, retirement jokes, dad jokes about grilling. Demand is flatter but it does not disappear. A design in one of these niches can still be earning a small royalty two or three years after you uploaded it.
A trending niche rides a spike: a viral phrase from social media, a specific holiday date, a cultural moment, a meme format. Search volume can be ten times higher than any evergreen niche for a two- to six-week window, then collapse to near zero. You are not building a catalog here; you are timing a wave.
The confusion in most advice you will read is that it treats "find good niches" as one skill. It is actually two different skills with different research methods, different upload cadences, and different failure modes, which is why a single list of "best niches" from a blog post rarely holds up for more than a season.
Before you decide where to spend your design time, it helps to understand what a sale is actually worth, because Merch by Amazon does not pay a flat percentage. As of September 2026, Amazon groups sellers into three royalty tiers based on how much of your sales volume comes from traffic you personally drove (your own ads, social posts, or off-Amazon promotion) rather than organic Amazon search:
Your group is recalculated monthly from a trailing two-month average, and a group change can take up to two business days to show up in your dashboard. Amazon states these definitions are subject to change, so check merch.amazon.com/resource/201858580 for the current wording before you plan around it.
Here is why this matters for the evergreen-versus-trending decision. Trending designs are naturally easier to push with off-Amazon promotion because there is already a wave of social attention to ride, which makes hitting the 15% or 35% non-organic threshold realistic. Evergreen designs mostly rely on organic Amazon search, so they tend to sit in the Creator group unless you are actively promoting them elsewhere.
| Offer Price | Creator Royalty | Plus Royalty (2x) | Premium Royalty (2.16x) |
|---|---|---|---|
| $15.99 | $0.96 | $1.92 | $2.07 |
| $17.99 | $1.70 | $3.40 | $3.67 |
| $19.99 | $2.44 | $4.88 | $5.27 |
| $21.99 | $3.18 | $6.36 | $6.87 |
| $23.99 | $3.92 | $7.84 | $8.47 |
| $25.99 | $4.66 | $9.32 | $10.07 |
The Creator column is Amazon's own published example for a standard US DTG t-shirt. The Plus and Premium columns apply the stated multipliers to that same figure, which is how the mechanic is described on Amazon's royalty page — treat them as a planning estimate and confirm the live number on your own payments dashboard, since Amazon's cost inputs (materials, fulfillment, returns) can shift the base figure at any time.
An evergreen catalog compounds. Every design you upload keeps a small chance of selling every single day, indefinitely, without you touching it again. After a year of steady uploads, most of your weekly royalty comes from designs you made months ago, not the ones you made this week. That is the entire appeal: your past work keeps working.
Lower research overhead per design, since the demand does not move week to week. Less competition from opportunistic sellers who chase whatever is trending. A design-slot tier that grows steadily as your total catalog racks up small, consistent sales rather than one big spike.
Any individual niche is usually crowded with sellers who found it years ago, so a brand-new "dog mom" design competes against listings with a long sales history and reviews. Growth is slow. You will not see a design take off in its first week the way a well-timed trending piece can.
Evergreen work also suits sellers who cannot check their account daily. If you upload a batch of camping-niche designs and go quiet for a month, they keep selling at roughly the same trickle. A trending batch left unattended for a month has usually already peaked and dropped by the time you check on it.
Trending niches can produce a design that outsells an entire evergreen collection in its first ten days. The catch is that the window is short and unforgiving: miss it by a week and you are uploading into a search term that already peaked.
Fast feedback. You know within days, not months, whether a design is working, which makes it far easier to learn what resonates. It is also the more realistic path to the non-organic traffic percentage that Plus and Premium royalty groups require, since spikes are what off-Amazon promotion actually has something to ride on.
Higher research load every single week, since last week's winning phrase is dead this week. Heavier trademark exposure, because viral phrases are exactly the kind of thing brand owners and law firms scan for. Once the spike passes, that design's sales usually drop to near zero and stay there.
Trend chasing also front-loads risk. A viral phrase is more likely than a generic evergreen phrase to already belong to someone, whether that is a registered trademark, a copyrighted meme format, or a public figure's name and likeness. That risk needs a check before upload, not after a takedown notice.
Rather than treating this as an all-or-nothing choice, most experienced sellers split their weekly design budget between the two and adjust the ratio based on how much research time they actually have. The table below is the comparison to run the decision through before you open your design software.
| Factor | Evergreen Niches | Trending Niches |
|---|---|---|
| Time to first sale | Weeks to months | Days, if you catch the wave early |
| Design shelf life | Months to years | Typically 2–6 weeks |
| Weekly research load | Low once a niche list exists | High, every week is a fresh search |
| Trademark exposure | Lower, if you avoid brand names outright | Higher, viral phrases attract legal scrutiny fast |
| Fit for Plus/Premium royalty groups | Harder without active promotion | Easier, spikes give promotion something to push |
| Failure mode | Slow, quiet catalog with no growth | Sudden dead zone once the spike passes |
| Best suited to | Sellers who upload in batches and check in monthly | Sellers who can research and publish within a day or two of a spike |
A reasonable starting ratio for a seller with no strong preference yet is roughly 70% evergreen, 30% trending. Evergreen carries the account while you learn which trending format actually converts for your specific design style, then you can shift the ratio once you have a few trend cycles of your own data to look at.
This is the step where almost everyone gets stuck, and it is worth naming plainly: Merch by Amazon caps how many products you can have live at once, and that cap rises in tiers as your account performs. This tier system is completely separate from the royalty groups covered above — a Creator-group account and a Premium-group account can sit at the same design tier.
Amazon has never publicly documented the exact sales volume or timeframe needed to move from one tier to the next. Every specific threshold you see in a forum post or a YouTube video is seller-reported consensus, not an official Amazon figure, and it can change without notice. Treat the numbers below as a rough shape of the system, not a guarantee.
| Design-Slot Tier | Commonly Cited Cap | What Sellers Report Moves You Up |
|---|---|---|
| Starting tier | 10 designs | Where every new account begins |
| Tier 25 | 25 designs | A small run of completed sales and clean account standing, no fixed number confirmed by Amazon |
| Tier 100 | 100 designs | Sustained sales velocity over multiple weeks, per seller-reported consensus |
| Tier 500 | 500 designs | Consistent sales across a growing catalog, unofficial and self-reported |
| Tier 1000+ | 1,000 or more | Long-term account health, exact criteria unpublished |
What this means for the evergreen-versus-trending decision is direct: your starting tier only fits ten designs, so your very first batch needs to be the highest-confidence bet you can make, not an experiment. Most sellers get this backwards and burn their first ten slots on speculative trending phrases that die before they even reach tier 25.
The safer opening move is to fill your first ten slots with evergreen niches you already understand from your own hobbies or work, since that catalog needs to earn its way to the next tier before you have room to experiment with anything short-lived.
Say you have design capacity for 20 new pieces this week and an account already past the starting tier. Here is one reasonable way to split that output using the 70/30 ratio from the decision framework above.
Pick two or three niches you already know from personal experience — for this example, camping and RV life, veterinary-tech humor, and grandparent jokes — and spread the 14 designs across variations in wording and layout within those niches rather than starting new topics from scratch.
Run every wording variant through a USPTO trademark search before you finalize artwork, since even a common-sounding phrase like a specific job-title pun can already be registered by someone else.
Pick a single spiking topic from this week's search and social signals rather than spreading six designs across six different trends — concentrating on one theme gives you a clearer read on whether the format works for your style.
Trending research decays fast, so the gap between spotting the spike and having artwork live should be measured in hours and days, not the following week.
Check the trending batch's performance at the 10-day mark to decide whether to make more variants while the wave is still up, and check the evergreen batch at 6–8 weeks, since organic search ranking for a new listing typically takes longer to settle.
Notice that the evergreen batch gets more slots but less urgency, while the trending batch gets fewer slots and a hard deadline. That asymmetry is the actual skill here: knowing which of your two buckets is allowed to wait and which one punishes hesitation.
The research method should match the bucket. Evergreen research is closer to a one-time investment; trending research is a recurring weekly chore that either gets a system or gets skipped.
The niches that hold up longest usually come from a hobby, a profession, or a life stage the seller actually understands, not from a random "top niches" list — that is also what makes the copy and visual jokes land instead of feeling generic.
Browse Best Sellers and Movers & Shakers inside Amazon's clothing and novelty categories to see which phrasing patterns are currently converting, then look for a gap your own niche knowledge can fill.
General search-trend tools show whether a phrase's interest is rising, flat, or already declining — a phrase that peaked two weeks ago on those tools is already too late for a trending upload.
Search the exact wording on the USPTO Trademark Search tool (tmsearch.uspto.gov, the successor to the old TESS system) before you spend art time on it — this applies to evergreen niches too, since common phrases in crowded categories like nursing or teaching are frequently trademarked.
A simple spreadsheet of niche, upload date, and 30-day sales tells you far more about your own audience than any general "best niches" article ever will, because it is specific to your art style and your account's history.
This is not legal advice, and trademark status can change; treat any USPTO search as a snapshot, last checked September 2026, not a permanent clearance. If a phrase is genuinely borderline, the safer move is skipping it rather than defending a takedown later.
One honest limit worth naming: no research tool, including a trademark checker, can tell you whether a niche will actually sell. It can only tell you whether the phrase is legally safe to try and whether search interest exists. The selling part still comes down to design quality and pricing, which no amount of research replaces.
Sellers running Merch by Amazon alongside Redbubble or TeePublic storefronts should also weigh whether a given design idea is better suited to one platform's audience than another — the buyer behavior and royalty structure are different enough that the same artwork does not always perform the same way twice.
There is no fixed permanent list, since demand shifts and trademark status changes over time. Start with two or three evergreen categories tied to a hobby, profession, or life stage you understand personally, since that knowledge shows up in better wording and jokes than a generic niche list can give you.
A common starting ratio is roughly 70% evergreen and 30% trending, adjusted once you have a few trend cycles of your own sales data. Sellers with less time for weekly research should lean further toward evergreen, since it requires far less ongoing attention per design.
Yes, though competition inside popular evergreen categories has grown, since many sellers found the same obvious niches years ago. The advantage is still there for designs built from real personal knowledge of a niche rather than a copied phrase, since that specificity is harder for competitors to replicate quickly.
Most spikes run roughly two to six weeks from first noticeable search interest to the point sales flatten out, though this varies by topic and can end sooner if a bigger trend displaces it. Checking performance around the 10-day mark is usually enough to tell whether a design is worth extending with more variants.
Past that point, tracking which individual designs are earning versus sitting idle matters more than uploading more volume blindly. Screening every new phrase against USPTO records before art time, and checking your royalty group monthly on your dashboard, both prevent the two most common account-health problems sellers run into at this stage.
Octozia's Trademark Checker runs your wording against USPTO records before you commit art time, on a 3-day trial that requires a card and can be canceled before it ends.