The Short Answer
GPM is the GMV a video produces per 1,000 views. Calculate it as orders value ÷ views × 1,000. If a video did 40,000 views and $620 in attributed sales, its GPM is $15.50.
The number that matters is not the number itself. It is whether that GPM covers what you spent to make and distribute the video. A video with a $40 GPM is excellent for a $12 product and unremarkable for a $90 one. So the useful form of the question is not “what is a good GPM” but “what GPM do I need,” and that is a function of your margin and your cost per video.
This article is for the operator running TikTok Shop across more than one market. Usually that means a US shop plus two or three Southeast Asian ones. You carry 50 to 200 SKUs and a mix of brand-made and creator-made video. If you are currently ranking videos by views or by completion rate, GPM is the metric that will change which videos you make next.
It is a decision guide, not a benchmark study. The ranges below are working ranges observed across seller accounts, not official platform figures, and the method for calibrating them to your own catalog matters more than the numbers themselves.
Why GPM Beats the Metrics Teams Usually Rank By
Every other video metric measures a step in the process. GPM measures the outcome, which is why it changes decisions that the other metrics cannot.
Views tell you reach and nothing about whether the reach was commercial. A video can do 200,000 views and produce four orders.
Completion rate tells you attention held. It says nothing about whether the viewer wanted the product, and as covered elsewhere, raising it can lower orders.
Likes and saves tell you the content landed. Saves in particular indicate postponement rather than purchase.
GPM contains all of it. It is the only common metric already weighted by whether people bought. That makes it the only one you can compare across videos of different lengths, different products, and different markets.
The limitation is worth stating plainly: GPM is an average, and averages hide the distribution. A video with a $30 GPM driven by one large order is not the same asset as a video with a $30 GPM driven by thirty small ones. Check the order count next to the GPM before you act on it.
The Ranges We See, and How Far to Trust Them
These are working ranges by average order value, observed across multi-market seller accounts in 2025 and early 2026. They are not platform-published benchmarks — TikTok Shop does not publish GPM figures by category — and you should treat them as a starting hypothesis to calibrate against your own data, not as a target.
| Average order value | Typical working GPM | Read it as | Action |
|---|---|---|---|
| Under $15 | $8 – $20 | Volume-dependent, thin margin | Push for volume; small GPM moves matter a lot |
| $15 – $40 | $20 – $60 | Normal range for most catalogs | Compare within your own catalog only |
| $40 – $90 | $50 – $140 | Considered purchase, slower | Judge over 14 days, not 7 |
| Above $90 | $120+ | Rare and unstable | Check order count; likely one or two large orders |
The honest caveats on this table are significant. Market matters as much as category. The same product at the same price typically produces a lower GPM in Indonesia or the Philippines than in the US. Average order values are lower there and conversion behaviour differs. Creator-made video typically produces a higher GPM than brand-made video for the same product, because the recommendation carries. And a new account’s GPM is not comparable to an established one’s, because distribution quality differs.
Calibrate Against Your Own Median
Calibrate by computing GPM across your last 30 videos, ranking them, and taking your own median as the baseline. The external ranges above are for the first week, before you have that number.
“The published ranges were useless to us for the first two months,” a Shenzhen operator running a phone accessories shop across the US and Indonesia told me. “Our Indonesia median was about a third of our US median on identical products. Once we ranked videos inside each market separately, the next decisions became obvious.”
The Decision Line GPM Actually Supports
GPM becomes useful when it is compared against a threshold you set yourself. That threshold is the cost of the video divided by the views you expect it to earn.
Work out your cost per video. For a creator video, that is the sample cost plus commission plus any flat fee. For a brand-made video, it is production time and any promotion spend. Then work out the GPM at which the video pays for itself: if a creator video costs $60 all-in and you expect 8,000 views, the break-even GPM is $7.50.
Once you have that number, every video falls into one of three buckets, and each has a clear action.
GPM above break-even by a comfortable margin. Make more of this structure with this product, or put more distribution behind it. This is the only case where adding spend is clearly justified.
GPM near break-even. The video is not losing money and not compounding. Leave it running, do not add spend, and do not re-make it. Most of a catalog’s video sits here, and that is normal.
GPM below break-even. Stop putting distribution behind it. Whether to remove it is a separate question — a low-GPM video still contributes reach — but it should not receive spend or a re-make on the same structure.
The limitation: this line assumes the views you expect are the views you get. A video’s GPM usually moves in the first week as distribution shifts. Set the line against a 7-day GPM rather than a 48-hour one.
Three Ways GPM Misleads
GPM is the best single metric available and it still gets misread in three specific ways, all of which are easy to guard against.
One large order inflates it. A single $400 order on a 20,000-view video moves GPM from $18 to $38. Always look at order count next to GPM; a GPM carried by one or two orders is not a repeatable asset.
Attribution windows distort it. A video credited with sales it merely assisted will show a GPM it did not earn, and a video that starts a purchase completed days later may be credited nothing. Read GPM over a consistent window — 7 days is the practical default — and do not compare a 3-day GPM to a 14-day one.
Market mixing hides the real picture. Averaging GPM across a US shop and an Indonesian shop produces a number that describes neither. Keep GPM per market, and only blend when you are reporting blended margin.
The DAMI store data view keeps GPM readable per shop rather than blended, which is the main practical requirement. The DAMI team overview is where a multi-store operator can see whether one shop’s GPM is drifting while the others hold. What neither of them can do is tell you the break-even line; that has to come from your own cost per video.
How to Raise GPM Without More Views
The useful property of GPM is that it can be improved without improving reach, and these are the three levers that do it, in order of reliability.
Raise the order value on the same views. Bundles, multi-buy, and a higher-value variant in the same video move GPM directly. This is usually the largest and most reliable lever available.
Match the product to the video’s audience. A video that earned its views from a specific demographic will convert better on a product that demographic buys. Check which product the video’s viewers actually bought and make the next video on that product rather than the one you planned.
Move the product link earlier. This is the smallest of the three levers and it only applies when the drop analysis shows viewers leaving before the link appears. It typically moves GPM by a modest percentage, not a multiple.
What does not raise GPM reliably is making more video. Volume increases total GMV without changing per-view efficiency. Teams often confuse the two. A rising GMV with a flat GPM means you are buying growth with output, and that stops working the moment output stops.
What Your GPM Tells You About the Commission You Can Afford
GPM has a second use that most sellers miss: it sets the ceiling on the commission rate you can offer a creator without losing money on the video.
The arithmetic is direct. If a video’s GPM is $30 and your gross margin is 45%, that video produces about $13.50 of gross profit per 1,000 views. Any commission above roughly 13% of the order value starts eating the profit, and any commission above 45% makes the video loss-making regardless of how well it performs on reach.
That is why a fixed commission rate across a catalog quietly destroys margin. A 20% commission is comfortable for a product whose video runs at a $90 GPM. It is unaffordable for one running at $18. Worse, the second product is usually the one getting the most creator attention, because it is the one that needs help.
Tier Commissions by What the Video Earns
The practical version is to tier commission by the GPM the product’s video actually produces. Products with proven high-GPM video can carry a higher rate, because the video pays for it. Products with low-GPM video either need a lower rate or a fix to the video first — offering a high commission on a video that does not convert just moves the loss from your content budget to your commission budget.
A Hangzhou seller managing commissions for a 90-SKU home living shop learned this the expensive way: “We paid a flat 20% across everything for a quarter. The high-GPM products were fine. The low-GPM ones ate the difference, and we only saw it clearly when one of them went viral at exactly the wrong commission.”
The caveat: GPM per product is noisy until a product has been in several videos. Do not set a commission tier from one video’s GPM. Wait until you have three or four videos on the same product and use the median.
Frequently Asked Questions
How do I calculate GPM? Attributed GMV divided by views, times 1,000. Use a consistent attribution window — 7 days is the practical default — and compute it per market rather than blended.
What GPM should I target? Whatever covers your cost per video. Work out your all-in cost and your expected views, divide, and that is your break-even GPM. External ranges are a first-week hypothesis, not a target.
Is a high GPM always good? Not without the order count. A $120 GPM from one order on 8,000 views is not repeatable. Check orders and GPM together every time.
Why is my GPM lower in Southeast Asia than in the US? Lower average order value and different conversion behaviour, mostly. The same product at the same price generally produces a lower GPM in Indonesia or the Philippines. Judge each market against its own baseline.
Does GPM apply to live selling too? The concept transfers but the inputs differ — live sessions have concurrent viewers rather than cumulative ones, so the comparable metric is GMV per session or per viewer-hour. Do not compare live and video GPM directly.
Set Your Own Line Before You Chase Anyone Else’s
The useful output of this article is not a number to hit. It is your own break-even GPM, computed from your own cost per video and your own expected views, applied per market. Everything else — the ranges, the comparisons, the category norms — is context for that one line.
Compute GPM across your last 30 videos this week and take your own median. That median, not any published benchmark, is the number your next video decision should be measured against. If you want the attribution method behind the GMV side of the calculation, this ROI walkthrough covers how shared sales get counted. See how DAMI keeps GPM readable per shop on the product page, or create your account to run the 30-video GPM read on your own catalog.
