For most sellers the feature is not available, which makes the better question what you were trying to solve.
Amazon geo targeting sounds like a lever every advertiser should have. Show ads where the product sells, and stop paying where it doesn’t.
Most Amazon sellers can’t pull it. Geo-targeting exists in Amazon DSP and nowhere else. Sponsored Products has none, and neither does Sponsored Brands. Sponsored Display has a version restricted to non-endemic advertisers, meaning brands that don’t sell on Amazon at all.
So the question splits in two. Does geo-targeting work? It does. Is it available to you? Usually not.
The second half is the useful one, because a seller asking about geo-targeting is almost always trying to solve something else. Work out what that is, measure whether regional variance is costing you anything, and you’ll know whether the strategy makes sense for your business before you price a channel to fix it.
At a Glance
Table of Contents
Five things to take away
- Geo-targeting is a DSP feature. Sponsored Products and Sponsored Brands have none.
- The Sponsored Display version is non-endemic only, meaning brands that don’t sell on Amazon.
- DSP carries a managed-service minimum near $50,000 per month, so the feature has a price attached.
- Amazon Marketing Cloud will show conversion rate by state without any DSP spend.
- Aggregation thresholds mean ZIP-level questions usually return nothing. State level is the practical floor.
Quick Answer:
Geo-targeting is worth it if you already run Amazon DSP and have measured real regional variance in conversion or margin. If you sell on Amazon and don’t run DSP, the feature isn’t available to you, because Sponsored Products and Sponsored Brands have no geographic targeting and the Sponsored Display version is restricted to non-endemic advertisers. The question underneath it, whether some regions convert worse than others, can be answered in Amazon Marketing Cloud without buying anything.
Definition
Geo-targeting delivers ads to an audience based on where they are. On Amazon that means state, city, DMA, ZIP or radius, set at line-item level, and you can include or exclude each one.
Dayparting does the same thing with time instead of place, scheduling ads to run during specific hours or days.
Audience targeting picks people by behavior or attribute, not location. Geo-targeting narrows an audience you’ve already defined, so it sits on top of audience targeting.
All three can run together.
1. Where Amazon geo targeting exists, and where it doesn’t
Amazon DSP is the only surface with full geographic controls.
The levels available. State, city, DMA, ZIP code and radius. Radius targeting runs from 2 to 100 miles around a point, and a single line item accepts roughly 1,000 locations. Every level supports inclusion and exclusion.
Where it’s set. At the line item, not the campaign. Amazon folded location targeting into a single line-item setting in April 2024, pulling what used to be separate region targeting into the same control.
Sponsored Products and Sponsored Brands have nothing. No state, no city, no ZIP, no radius. As of September 2026 that hasn’t changed.
Sponsored Display has a version that isn’t for you. Geo-targeting on Sponsored Display is only open to non-endemic advertisers, meaning brands that don’t sell products on Amazon or that run link-out campaigns away from Amazon.com, like a bank advertising loans, an insurer or a local services business.
Geo-targeting showed up briefly in Sponsored Display for endemic sellers in early 2024. It was a rollout error that reached accounts it wasn’t meant for, and Amazon pulled it within days. It hasn’t been extended since.
What to do
- Check which surface you’re asking about. The answer is completely different for DSP and everything else.
- Price the feature in full. Geo-targeting through DSP means buying DSP, and DSP carries a spend commitment.
What to avoid
- Assuming a feature you saw in a screenshot or a webinar applies to your account type. Most coverage of this topic never mentions the non-endemic distinction.
- Waiting for Amazon to extend it. Two and a half years of no movement tells you something.
2. What you were probably trying to solve
A seller asking about geo-targeting usually has one of three problems underneath the question, and the most common one is the one you can check for free.
Regional inventory imbalance. Stock sits in one fulfillment region and sells in another, or a product ships badly to some states. That’s a fulfillment and catalog problem, and geo-targeting would only suppress demand without fixing the imbalance. Suppressing demand costs rank.
Suppressing demand to match your stock position is a decision with a cost attached, and the cost lands on rank rather than on the advertising line. A region you stop selling into stops contributing to velocity, and velocity is what holds the position you will want back when the stock arrives. Amazon account management is where the fulfilment position and the advertising position get read as one decision.
A promotion that only runs in some markets. A regional retail partnership, a market-limited launch, a price test in one state. That’s a real geo use case and it does need DSP. It’s also rare below enterprise scale.
A suspicion that some regions convert badly. This is the common one, and it’s the only one you can test without spending anything.
What to do
- Name the problem before you name the tool. Write down what you expect geo-targeting to change, and by how much.
- Check whether the problem is upstream. Inventory placement and shipping cost produce much of what looks like regional performance variance.
What to avoid
- Reaching for a channel to solve an operations problem. It costs more and works less well.
- Assuming variance exists because it feels like it should. Section 3 is how you find out.
3. Measuring regional variance before you buy anything
Amazon Marketing Cloud answers the question that sends people to geo-targeting, and you don’t need DSP spend to run the query.
What AMC shows. Unique reach, total conversions and conversion rate by geography. You can compare performance across states, countries and media markets, and read it against time of day if dayparting is also in play.
|
+1,119%
Profit Growth
Levtex Home
 |
+586%
Profit Growth
Karat
 |
+833%
Revenue Growth
Crazy Dog
 |
The constraint is aggregation thresholds. Amazon holds back results below a minimum user count, so individuals can’t be identified from the data. The threshold is a fixed number, not a percentage, so a query that returns clean results at state level can return nothing at ZIP level.
That sets the practical floor. For most catalogs, state is the level where the data comes back. If you need finer detail, the answer is usually a bigger data set or a longer time window, not a narrower query.
What the answer tells you. If conversion rate is flat across regions, geo-targeting was never the answer and you’ve saved the money. If the variance is large and holds across periods, you have a case for DSP and a number to justify the commitment with.
What to do
- Run the query at state level first. Narrow it only if the numbers hold up and the threshold allows it.
- Read across at least two periods. A single quarter can show regional variance that seasonality explains.
- Turn the variance into money before deciding. A five-point conversion gap in a state carrying 2% of volume isn’t a business case.
What to avoid
- Starting at ZIP level. It’s the query most likely to return nothing, and the one most likely to be misread as no variance at all.
- Treating a withheld result as a zero. Suppressed and absent aren’t the same thing.
4. Running it well, if you have DSP
If you’re already at DSP scale with confirmed variance, a few execution rules do most of the work.
Start broad, then refine. Begin at state level and narrow only where performance data supports it. Over-segmenting limits reach and raises delivery cost without improving the result.
Exclusion is often the higher-value move. Removing regions where shipping cost or delivery time wipes out margin does more for the P&L than adding a region where you already perform.
Match the creative to the location where it earns it. Local imagery, regional language or a market-specific offer. Geographic targeting without geographic creative is just a smaller audience seeing the same ad.
Combine with dayparting deliberately. Layering location and time sharpens precision, and it multiplies over-segmentation too. Use it when the data shows variance on both dimensions, not because both controls are there.
Monitor and reallocate. Shift budget toward regions that perform and pause the ones that don’t, on a set schedule.
What to do
- Set a floor for how narrow you’ll go. Decide it before launch, not once delivery starts falling.
- Review location performance against the same window you used to justify it. Otherwise you’re comparing a targeted period to an untargeted one.
What to avoid
- Segmenting just because the control exists.
- Running geographic creative variants without enough volume in each to read them separately.
Decision grid
The problems that send sellers to geo-targeting, and where each one gets resolved.
If the first column doesn’t describe your situation, geo-targeting isn’t the thing to price.
How Adverio Helps
Adverio is an agency that runs Amazon, Walmart and Target for consumer brands, measured on contribution margin against a baseline set before the work starts.
On a geo-targeting question, that means running the variance check before any channel conversation. Usually performance is flat enough across regions that the feature wouldn’t pay for itself, and telling you that is worth more than selling you a DSP engagement that won’t pay back.
Where the variance is real, it becomes the number that justifies the DSP commitment, not a hope attached to it.
Frequently asked questions
Can I geo-target Sponsored Products campaigns?
No. Sponsored Products has no geographic targeting of any kind, and neither does Sponsored Brands, and Amazon hasn’t signaled any plan to change that.
Geo-targeting on Amazon means Amazon DSP, with one exception covered below.
What about the Sponsored Display geo-targeting I’ve seen mentioned?
It exists, and it’s restricted to non-endemic advertisers, meaning brands that don’t sell products on Amazon or that run link-out campaigns away from Amazon.com.
The version endemic sellers saw in early 2024 was a rollout error, and it hasn’t come back.
How granular can DSP geo-targeting get?
State, city, DMA, ZIP and radius, with radius running from 2 to 100 miles and a line item accepting roughly 1,000 locations. Everything is set at line-item level.
Being able to target a ZIP code doesn’t mean you can measure one, because aggregation thresholds usually hide ZIP-level results.
Do I need DSP to find out whether geo-targeting would help?
No, and that’s the useful part. Amazon Marketing Cloud will show conversion rate by geography without any DSP spend attached to the query.
If the variance comes back small at state level, the question is answered and it cost you nothing.
Closing
Run the state-level query before anyone quotes you a DSP budget.
References
Platform terms move. Figures were checked on 3 September 2026 and should be confirmed before you budget.
- Amazon Ads, DSP ad group targeting, geo locations. https://advertising.amazon.com/API/docs/en-us/dsp-ad-group-targeting-geo
- Amazon Ads, Sponsored Display locations guide for non-Amazon sellers. https://advertising.amazon.com/API/docs/en-us/guides/sponsored-display/non-amazon-sellers/locations
- Amazon Ads, location targeting line item simplification, April 2024. https://advertising.amazon.com/resources/whats-new/location-targeting-line-item-page-simplification-for-self-service