Amazon Search Frequency Rank is a ranking of search terms by how often shoppers use them, and rank 1
is the most searched term in the marketplace. Lower number, higher demand. That inversion catches people
out constantly, and a seller who reads it the wrong way will systematically chase the terms nobody
searches for.
Two more things worth knowing before you use it. It is a rank, not a volume, so the gap between rank 200
and rank 400 is not proportional to anything. And it lives in Brand Analytics, which means you need Brand
Registry to see it at all.
Below: where to find it, how to read it against your own performance, and the four things it genuinely
cannot tell you.
What the number actually means
A rank, ordered by search volume, refreshed on a schedule and specific to a marketplace.
Rank 1 is the single most searched term in that marketplace for the period. Rank 100,000 is searched
far less. I’d write that on a sticky note.
It is ordinal, not cardinal. The distance between rank 1 and rank 2 is enormous; the distance between
rank 50,000 and rank 50,100 is trivial. Treating rank as a proxy for volume is the second most common
mistake here, after reading the direction backwards.
It is per marketplace. A term’s rank in the US tells you nothing reliable about the UK, and in our experience sellers expanding internationally assume otherwise.
It moves. Seasonal terms swing hard, and a rank read once is a snapshot rather than a trend. I’d pull
it repeatedly rather than once.
It describes the market, not you. Your position on that term is a separate question entirely, and this
is where most SFR analysis goes wrong.
Where to find it
Brand Analytics, inside Seller Central, in the search terms report. Brand Registry is the gate. No
registry, no access, and no workaround I’d recommend.
The report gives you more than the rank. Alongside the term you get the top clicked ASINs and their
click and conversion share, which is the part that makes the rank useful rather than decorative.
Search Query Performance is the companion report and the one I’d actually spend time in, because it
tells you your own impression, click, and purchase share on terms relevant to your brand. Our guide to
Amazon Brand Analytics covers
reading both properly.
Set a cadence. Weekly for a fast category, monthly otherwise. In our experience sellers pull this once,
build a keyword list, and never look again, which defeats the point of a metric that moves.
Reading it against your own numbers
This is the whole technique, and it takes four columns.
Column one: the term’s rank. How much demand exists.
Column two: the top clicked ASINs. Whether that demand goes to products like yours or to something
adjacent. A high-demand term whose clicks all go to a different product type is not your opportunity,
and I’ve found this single check kills about a third of a naive keyword list.
Column three: your own click share on that term, from Search Query Performance.
Column four: your conversion share. Because ranking on a term you convert badly on costs money.
The four combinations that matter:
- Good rank, you have share, you convert. Defend it. Bid to hold position.
- Good rank, you have share, you convert badly. A listing problem, not an advertising one. Our guide to
improving your product detail page
covers diagnosing which element loses the sale. - Good rank, no share. The opportunity, if the top clicked ASINs resemble your product. Otherwise
noise. - Poor rank, you have share. Fine, and do not spend to grow it. There is nothing there.
I’d build that table before touching a bid. When I have done it on accounts with long keyword lists, the
useful rows were always fewer than expected.
A worked read, start to finish
Abstract advice about metrics is easy to nod along to and hard to apply. So here is the sequence on one
term.
Say the term ranks well. Genuinely high demand. Encouraging, and by itself worth nothing.
Look at the top clicked ASINs next. Suppose they are all a slightly different product: a bigger
capacity, a different material, a professional rather than domestic version. That demand is not
addressable by your listing, and no bid fixes a mismatch between what shoppers wanted and what you sell.
I’d drop the term here and feel good about it.
Now suppose the top ASINs do resemble your product. Check your own click share. If it is near zero,
you have found something: real demand, relevant competitors, and no presence.
Then check conversion share before spending. If you already receive clicks on this term and convert
poorly, advertising harder buys more of an existing failure. The fix is the listing, and I’d fix it first.
Finally, price the opportunity. Take your margin, your conversion rate, and derive break-even cost per
click. Compare that against what the term plausibly costs. A term you cannot afford is not an
opportunity either, and this is the step sellers skip most often.
Five checks. Perhaps twenty minutes per term. In our experience the discipline pays for itself the first
time it stops a campaign that would have run for a quarter.
When the report looks empty or useless
Three situations that come up, and none of them mean the metric is broken.
No Brand Registry, so no report. The honest answer is that this metric is unavailable to you. Work
from your own search term report in the advertising console instead, which shows what shoppers actually
typed before clicking your ads. Less market context. Better intent data.
A brand-new product with no share anywhere. Expected. The report is telling you about the market, and
you have not entered it yet. Use it to choose which terms to target rather than to measure yourself.
A niche so small the ranks are all poor. Also fine, and genuinely useful information. A category where
every relevant term ranks badly is a category with thin demand, and I’d rather learn that from a report
than from a year of advertising.
What SFR cannot tell you
Four limits, and pretending otherwise is how sellers waste quarters.
It does not give you volume. No absolute number of searches whatsoever. Anyone quoting one has estimated it, and I’d treat the estimate as marketing.
It does not tell you profitability. A heavily searched term in a heavily advertised category can be
unaffordable at your margin. Break-even cost per click decides that, not demand, and our guide to
Amazon bid types and strategies covers the
arithmetic.
It does not tell you intent quality. “Cheap running shoes” and “waterproof trail running shoes size
11” can rank similarly and behave nothing alike. When I have compared the two side by side, the conversion gap was the whole story.
It does not tell you whether you can win. Rank plus incumbent click share does. A term dominated by an
entrenched brand with thousands of reviews is a rank you can see and a position you cannot take.
Using it without over-fitting
Five habits that keep this useful.
Start from your own converting terms and work outward. Not from the highest-ranked term in the
category. In our experience the top-ranked head terms are the worst place for a small account to begin.
Group terms by the attribute they name, then check whether your listing actually names that attribute.
This is where SFR turns into listing work rather than bidding work.
Use it to kill keywords, not only to add them. A term you are bidding on that ranks poorly and
converts poorly should go. I’d do a subtraction pass every quarter.
Feed the good terms into structure, not just into bids. One ASIN per ad group, terms grouped by
attribute. Our guide to
Amazon PPC campaign structure covers the shape.
Re-read it seasonally. A term that ranked poorly in March can rank well in October, and I’d rather
find that in the report than in a competitor’s advertising.
The mistakes worth naming
Reading the direction backwards. Chasing high rank numbers because higher sounds better. It is the
error this page exists to prevent, and in my experience it survives a surprising amount of experience.
Treating rank as volume. It is not, and the difference compounds when you rank-order a budget.
Ignoring the top clicked ASINs column. The rank tells you demand exists. That column tells you whose
demand it is.
Stuffing high-demand terms into a title. A title written for the index reads as spam to the shopper who
is deciding. Our guide to
Amazon search term optimization covers where
keywords belong instead.
Pulling it once. The metric moves. A one-time keyword project stays a one-time keyword project, and I’d schedule the next pull before closing the report.
If you would rather have Brand Analytics read and turned into a campaign structure against your margin,
that sits inside our
Amazon PPC service. We are an Amazon Ads
partner and an Amazon SPN Verified Partner.
FAQ
What is Amazon Search Frequency Rank?
A ranking of search terms by how often shoppers in a marketplace use them, where rank 1 is the most
searched term. It appears in the Brand Analytics search terms report in Seller Central and requires Brand
Registry to access.
Is a lower Search Frequency Rank better?
Yes. Rank 1 is the most searched term and higher numbers mean less demand, which is the reverse of how
most metrics read and the most common source of error when sellers first use it.
Does Search Frequency Rank show search volume?
No. It is an ordinal rank, not a volume figure, so the gap between rank 1 and rank 2 is far larger than
the gap between rank 50,000 and rank 50,100. Any absolute search number attached to it has been estimated
by a third party.
Do I need Brand Registry to see Search Frequency Rank?
Yes. The metric lives in Brand Analytics, which is available to Brand Registry enrolled sellers. Without
registry you cannot access the report.
How should I use Search Frequency Rank?
Pair it with the top clicked ASINs from the same report and with your own click and conversion share from
Search Query Performance. Demand alone is not an opportunity; demand that flows to products resembling
yours, where you have low share, is.
What are the limits of Search Frequency Rank?
It gives no volume figure, no indication of profitability at your margin, no read on intent quality, and no
sense of whether an entrenched competitor makes a term unwinnable. It describes the market rather than your
position in it.
Last updated: August 31, 2026. Brand Analytics report names, available columns, refresh cadence, and
eligibility change over time; Seller Central carries the current version for your account.
