Etsy Conversion Rate Labels: Why Very High Misleads

Etsy puts a coloured label beside every search term inside Marketplace insights: Very low, Low, Typical, High, Very high. It is the fastest signal on the page and it was the one I trusted most. Then I crossed it against the median purchase price on the same twenty keywords, and the single Very high term in my dataset turned out to be the worst business on the list.

Written by Mouhcine El Aboudi at SellerAIReady. Last updated 7 September 2026.

The label I screenshotted and did not interrogate

If you spend any time in Etsy seller communities you have seen the screenshot. Somebody searches a keyword inside Marketplace insights, the page returns a green badge that says Very high, and the screenshot gets posted with a caption along the lines of found my next product. I have seen that exact post dozens of times. I have also written the seller-side equivalent of it in my own notes.

The label is genuinely valuable. It is first-party data, it comes directly from Etsy, and no third-party keyword tool can manufacture it. That is precisely why it deserves more scrutiny than I gave it. A number that cannot be checked against a second source tends to be accepted without question, and a label is even easier to accept than a number because it already looks like a verdict.

Here is what happened to me. I ran twenty seller-facing keywords through Marketplace insights and published the full result in the twenty-keyword conversion study. Exactly one of the twenty came back Very high: etsy listing template, at 241 searches over thirty days and climbing. I wrote a three-product roadmap around it that same evening. Twenty-four hours later I cancelled the entire roadmap, and the thing that killed it was the median purchase price sitting two scrolls below the badge.

The label was not wrong. My reading of it was. I had treated an ordinal ranking of likelihood as if it were a statement about money, and those are not the same claim. This article is the correction, written out properly, with the arithmetic exposed so you can disagree with it if you want to.

Where the label lives, and what Etsy actually says about it

The path has not changed since I first documented it in the Marketplace insights walkthrough: Shop Manager then Stats then Marketplace insights, then type a phrase into the search bar. The result page returns the term you searched, a conversion badge next to it, a Last 30 days block containing Searches with a percentage change, a Search results count, a daily chart with a seven-day average, a paginated table of similar search terms, and a Search term analysis section holding the Listing price range panel.

The conversion badge is the second thing on the page and the first thing your eye lands on, because it is the only element that is colour-coded. Everything else is grey text. That design choice matters more than it sounds: the page is visually organised so that the least quantified field is the most prominent one.

That disclaimer sits under the price panel, and it is worth reading twice. Etsy is telling you three things in one sentence: the figures are a sample, they are aggregated, and the responsibility for any decision built on them is entirely yours. None of that makes the data less useful. It does mean the data is a starting point for a calculation rather than the end of one.

What Etsy publishes, and what it withholds

It is worth being precise about the boundary of this dataset, because the bad decisions built on it come from assuming the boundary is somewhere else.

• Etsy publishes: the band name only, from a fixed set of five values, for the phrase exactly as you typed it, over a rolling thirty-day window.
• Etsy withholds: the numeric conversion percentage, the sample size behind it, the denominator it is measured against, whether the measurement runs search-to-purchase or search-to-listing-view-to-purchase, and how wide each band is.

The practical consequence is that this label is ordinal, not cardinal. You can legitimately say that a High term converts better than a Low term. You cannot say it converts twice as well, or five times as well, and you cannot multiply the label by anything. Every conversion figure that appears later in this article is an assumption I am making openly, not a number Etsy gave me.

The five bands, and how our twenty terms fell across them

Across the twenty seller-facing terms I checked, the distribution was not a bell curve. It was heavily weighted to the bottom of the scale.

Look at the right-hand column and the pattern is already visible before any arithmetic. The two bands at the bottom of the scale contain terms with low competing-listing counts. The three bands at the top contain terms with 19,700 to 23,000 competing listings. In this dataset, conversion quality and supply pressure move together, and they move in opposite directions for your profit.

That is not a coincidence and it is not unique to seller keywords. It is what an efficient marketplace looks like. Etsy has roughly five point seven million active sellers, and a substantial share of them have access to the same label you do. A phrase that publicly announces itself as high-converting does not stay under-supplied for long. I covered the seller-to-buyer structure behind this in the saturation analysis, and the conversion label is that same dynamic expressed one keyword at a time.

A conversion label is a rate, not a revenue figure

A conversion rate is a fraction. It tells you what proportion of searches turn into purchases. It deliberately says nothing about the three quantities that decide whether a keyword is worth building for.

1. How many searches exist. A perfect conversion rate on thirty-one monthly searches is still thirty-one monthly searches. etsy seo guide is the extreme case in my dataset, and it happens to be Very low as well, which is the worst of both.
2. How much money moves per sale. A rate is dimensionless. It has no currency attached. Two keywords with identical labels can differ by a factor of five in what a buyer actually pays.
3. How many listings share the result. The label describes the whole search, not your slice of it. Nineteen thousand seven hundred listings splitting a keyword is a different business from three thousand nine hundred splitting one.

This is the same discipline as the five-gate framework in the demand validation article, compressed into a single line. The gates exist because each field on its own is capable of pointing you in the wrong direction, and the conversion label is the field that misled me, precisely because it is the most confident-looking thing on the page.

The comparison that broke my assumption about "Very high"

In my whole dataset there are exactly two search terms where Etsy returned both a conversion label and a numeric median purchase price range. Everything else came back Not enough data on price. Two data points is a small sample and I am not going to pretend otherwise, but these two happen to sit at opposite ends of the conversion scale, which makes the comparison unusually clean.

One is badge reel, the consumer craft keyword I used as a scale reference in the twenty-keyword study: Very low conversion, enormous volume. The other is etsy listing template, the seller keyword that was going to be my product line: Very high conversion, small volume. I put both through the formula above, and to keep myself honest I loaded the assumption against the conclusion I ended up with, giving the Very high term a conversion rate five times better than the Very low one.

The bottom row is the entire article. The keyword Etsy labels Very low makes roughly sixty-four times more money available per competing listing than the keyword Etsy labels Very high. I handed the Very high term a five-fold advantage on the one variable I was allowed to assume, and it still lost by a factor of sixty-four.

The sensitivity check, in case my assumption is wrong

A conclusion that depends on a guessed input is only worth publishing if it survives the guess being wrong. So I moved the assumption to two more extreme positions, both of them even more generous to the Very high term.

• Very low at 0.5 per cent, Very high at 10 per cent — a twenty-fold advantage. Result: USD 0.0394 against USD 0.0024, so badge reel still wins by about sixteen times.
• Very low at 0.25 per cent, Very high at 15 per cent — a sixty-fold advantage, well past anything plausible for a marketplace search. Result: USD 0.0197 against USD 0.0037, so badge reel still wins by more than five times.

The ranking does not flip. It cannot flip, because the gap being measured is not really a conversion gap at all: it is a 900-fold volume gap multiplied by a five-fold price gap, against a fourteen-fold supply advantage. Conversion rate is the smallest of the four forces in play, and it is the only one the badge tells you about.

What actually makes a keyword convert: three mechanics

Once you accept that the label is a rate and not a verdict, a better question opens up: what causes a phrase to convert at all? Across the twenty-two terms I have now examined, three mechanics explain almost every label I have seen, and none of them are about product quality.

1. Decision friction, which is mostly price. A two-dollar digital download is not a purchase decision, it is an impulse. A two-dollar file does not invite comparison shopping. In my own data that is what sits under a Very high label more than anything else, and it is also the reason a Very high label so often sits on top of a terrible margin.
2. Intent specificity. A phrase that names an exact deliverable converts. A phrase that names a topic does not. This is the mechanic with the biggest teaching value, and it gets its own section below.
3. Category familiarity. If the searcher already knows what the product looks like, what it costs and what it does, the listing has almost no explaining to do. If the searcher is still learning the category, the listing has to educate first and sell second, and education-stage traffic converts badly no matter how good the listing is.

Price is the friction, not the quality

This is worth separating out because it is where I quietly substituted a flattering explanation for the real one. When a keyword converts well, the instinct is to read it as buyers on this term know what they want. Sometimes that is true. More often the term simply sits at a price point below the threshold where the purchase needs thinking about.

You can test which explanation applies without spending a lookup. Open the Listing price range panel under Search term analysis and read the median purchase price band. If the band is low, the conversion label is mostly measuring cheapness. If the band is healthy and the label is still strong, the conversion label is measuring genuine intent, and that is the combination worth building for. I walked through this panel field by field in the median purchase price breakdown, including what the number does to your margin once Etsy's fees come out and you are left holding USD 1.36 on a two-dollar sale.

One word changed the label across three bands

This is the finding from the whole dataset that I would keep if I had to throw the rest away. The conversion label is extraordinarily sensitive to phrasing, and the size of the swing has almost nothing to do with the size of the market.

Read the first two rows. Same product category. Adding the single word custom removes ninety-six per cent of the search volume and moves the label from the bottom of the scale to the top. Somebody typing badge reel is browsing a category. Somebody typing custom badge reel has already decided to buy and is now specifying. Those are two different humans at two different moments, and Etsy's label is detecting the difference.

Now read rows three, four and five. etsy shop templates is Very low. Drop the plural and etsy shop template becomes Typical. Swap one noun and etsy listing template becomes Very high. Three bands across three phrases that a human would describe as the same search, with search volumes of 210, 167 and 241 respectively. The volumes are effectively identical. The labels are not remotely identical.

The lesson is not use singular nouns. The lesson is that the label is attached to the exact string, not to the concept, so checking the concept is not the same as checking the phrase. If you validate etsy shop templates, conclude the niche is dead and walk away, you have just walked away from a Very high term that differs by two words. This is also why title and tag work is genuinely load-bearing rather than cosmetic, a point I made structurally in the tags versus attributes breakdown: the string you occupy is the string you get judged on.

Why ten of twenty seller keywords came back "Very low"

Half of my sample landed in the worst band. That is not bad luck and it is not a coincidence of which twenty terms I picked. It is structural, and understanding why it is structural is more useful than any individual number in this article.

Seller-facing searches are overwhelmingly research searches. Somebody typing etsy seller guide or etsy shop ideas or etsy seo guide is trying to learn something. And information about selling on Etsy has an abundant, high-quality, permanently free substitute: blog posts, help documentation, videos, forum threads. The searcher can satisfy the underlying need without ever completing a transaction, and a meaningful proportion of them do exactly that. When the need behind a search has a free substitute, the conversion rate on that search is capped by something no listing can influence.

Compare that with a consumer craft search. Somebody typing custom badge reel cannot download a badge reel. There is no free substitute for a physical object, so the only way to satisfy the need is to buy. That single asymmetry explains the gap between the seller-tool cluster and the consumer cluster on Etsy, and it explains why the etsy [x] segment is simultaneously the most crowded and the least profitable segment I have measured.

There is an uncomfortable conclusion in that for a shop selling to Etsy sellers, and it applies to my own. It says the seller-education market is a genuinely large market where the demand is real but the willingness to pay inside Etsy search is structurally weak. It also says something more useful: that demand does not disappear, it simply gets satisfied somewhere other than a listing page. Which is precisely why I put the analysis in the demand validation framework and then, after running it against my own shop, chose to stop producing listings and keep producing data instead. The same conclusion, reached from the pricing side rather than the intent side, is written up in the median price article.

The four quadrants: label crossed with median price

Two free fields, four combinations. This is the whole decision framework, and once you have seen it you cannot un-see the trap in the top-left corner.

• Very high label, low median price — the impulse trap. Where etsy listing template lives. Buyers convert readily and pay almost nothing. Feasible only at extreme volume, which requires the reviews and the ranking you do not yet have. This quadrant looks like the best one on the page and is the worst one to enter.
• Very high label, healthy median price — the only genuinely good quadrant. Where custom badge reel lives: high intent, real money per transaction, volume small enough that the label alone has not crowded the term. It is also the hardest quadrant to enter, because it requires either a genuine craft capability or accumulated trust. If you find a term here, stop reading and go build.
• Very low label, healthy median price — the volume and trust play. Where badge reel lives. Enormous traffic, weak conversion, real prices. Viable for an established shop with reviews and photography that stops the scroll. Not viable as a cold start.
• Very low label, low median price — dead. Where etsy seo guide lives at thirty-one monthly searches. There is no listing, no price, no photograph and no tag strategy that rescues this quadrant. Leave it.

The uncomfortable part of this map is that quadrant two is the only one worth entering, and it is the one you cannot reach by reading data alone. Data finds it. Capability and trust are what let you occupy it. Keyword research alone does not produce a business, and advice that stops at the data is skipping that step.

What the label tells you to fix, and what it tells you to abandon

The most practical use of the label has nothing to do with picking products. It is diagnostic. It tells you whether a disappointing listing is a fixable problem or a category error, and those two situations need opposite responses.

• Typical or High, and your listing is not selling: this is a fixable problem. Buyers on that phrase do purchase; they are simply not purchasing from you. Photograph, title, first line of the description, price position against the median, and review count are the levers, in roughly that order. I broke the full diagnostic sequence down in the views-but-no-sales article.
• Very low, and your listing is not selling: this is not a fixable problem. You are competing for attention inside a search that was not heading towards a purchase in the first place. No amount of optimisation converts a reader into a buyer. Change the phrase you occupy, not the listing.
• Very high with a low median, and your listing is not selling: check your price against the band before touching anything else. If you are priced at two to six times the median with no reviews, the label was not describing your listing in the first place.

That third case is the one I got wrong, and it cost me nothing only because the median price field caught it before I built anything. If you want to run this diagnostic across a whole shop rather than one listing, the third-party tools are genuinely useful for the historical view Etsy does not give you — EverBee for estimated sales history on competing listings and eRank for tag and title auditing at scale. I compared what each one is actually good at, and where both of them are guessing, in the eRank versus EverBee comparison. Use them for the history and the breadth. Use Etsy's own label for the truth about intent, because that is the one number they cannot model.

The label is measured on the search, not on your shop

This is the misreading that has cost me the most, and it took me longer than it should have to state it plainly to myself. The conversion label describes the aggregate behaviour of the buyers who searched that phrase, across every listing that appeared. It is a property of the search. It is not a property of your listing, your shop or your product.

Which means a Very high label and a zero-sales listing are perfectly compatible, and there is no contradiction to resolve. The label is telling you that money changed hands on that phrase. It is not telling you that any of it came near you. On a phrase with 19,700 competing listings, the overwhelming majority of participating shops recorded zero sales in the same thirty-day window that produced the Very high badge. The badge and their zero are the same dataset.

Two consequences follow, and both of them are practical.

• Do not use the label to grade your own performance. Your conversion rate is your own number, visible in your own shop stats, and it has to be judged against your own history. A Very high market label with a flat shop line means the market works and your position in it does not.
• Do not treat the label as a forecast. It is a description of the last thirty days for a whole search, not a projection for a new listing entering that search from position nothing with zero reviews. A new listing's realistic first-month conversion on any phrase is closer to zero than to the market average, whatever the badge says.

If your listings are getting impressions and views without orders, the label tells you which of the two possible causes you are looking at, and then the actual repair work is a listing-level exercise rather than a keyword-level one. That sequence — diagnose with the label, repair at the listing — is the one I would follow, and the repair half is laid out step by step in the views-but-no-sales breakdown.

When "Very low" is still the right keyword

Everything above argues for avoiding Very low terms. There are three situations where that advice is wrong, and I would rather name them than publish a rule that pretends to be universal.

1. You are the destination, not a candidate. If buyers search a phrase that effectively means your shop — your brand name, a product name you originated, a phrase your returning customers use — the aggregate label is irrelevant. You are not competing inside that search, you are the answer to it. Aggregate conversion on a phrase dominated by browsers says nothing about conversion on a phrase dominated by people looking for you specifically.
2. You own the scroll-stopper. On a huge, weakly-converting consumer term, a small conversion rate over enormous volume is still a real business — provided your photograph is the one that stops the scroll. badge reel is Very low and still moves an estimated USD 21,560 a month in my illustration. Somebody is collecting that. But this requires established photography, reviews and ranking, which is exactly what a cold-start shop does not have.
3. You are targeting the phrase with content, not a listing. This is the one that changed my own strategy. A research-intent phrase converts badly on a listing page and well on an article, because an article is what the searcher actually wanted. The same Very low label that disqualifies a product can qualify a blog post, a pin or an email — the difference is what you put in front of the intent. That reframing is the foundation of how I now approach discovery, including the shift toward AI-assisted shopping surfaces described in the AI shopping SEO primer.

Note what all three have in common: each one changes something about you, not something about the keyword. The label is fixed. Your position relative to it is the only variable you control.

Your twenty-minute conversion-label audit

Here is the routine I now run before any product, listing rewrite or article decision. It costs nothing, it uses no third-party tool, and if a term fails at any step the remaining steps are unnecessary.

The exact click path

1. Open Shop Manager, then Stats, then Marketplace insights.
2. Type the phrase exactly as a buyer would, in lower case, with no brand words and no plural you would not naturally use. The string is what gets judged.
3. Read the conversion label. If it is Very low and you are validating a product, stop — unless one of the three exceptions above genuinely applies to you.
4. Read Searches for the last thirty days and the percentage change. Sub-fifty searches is not a market, whatever the label says.
5. Read Search results and divide searches by listings. This is your supply ratio, and I would treat anything worse than fifty competing listings per monthly search as a warning.
6. Scroll to Search term analysis and open Listing price range. Note the median purchase price band, or note Not enough data — which is itself an answer, not a blank.
7. Run the one formula: searches, times your assumed rate, times the median price, divided by competing listings. Write the result down.
8. Page through the similar-terms table underneath and repeat step two on every variant that looks close. This is free, and it is where the singular-plural and one-word swings hide.
9. Only after all of that, decide. And write the assumption you used next to the decision, so that in a month you can tell whether the plan failed or the assumption did.

Nine steps, twenty minutes, zero cost. The step people skip is the eighth one, and it is the step that found the highest-converting phrase in my entire dataset.

Frequently asked questions

What does the Etsy conversion rate label mean?

It is Etsy's own rating of how often searches for a specific phrase result in a purchase, shown inside Marketplace insights as one of five bands: Very low, Low, Typical, High or Very high. It is measured on the exact string you typed, over the last thirty days, across all listings that appeared for that search. Etsy publishes the band name only — never a numeric percentage, a sample size or a denominator — so the label lets you rank two phrases against each other but does not let you calculate with them.

Where do I find the conversion rate for an Etsy search term?

Open Shop Manager, then Stats, then Marketplace insights, then type the phrase into the search bar. The conversion badge appears immediately beside the term at the top of the result page, above the thirty-day search count and the competing-listing count. The tool is free with any active Etsy shop and requires no minimum sales, but new lookups are metered, so re-opening a term you have already searched and paging through its similar-terms table costs you nothing.

Is a "Very high" conversion rate keyword automatically a good keyword?

No, and in my own data this was the misreading that cost me the most. A Very high label frequently means the price point is low enough that buying requires no thought. In my own data the single Very high term carried a median purchase price of USD 1.80 to USD 2.20, which leaves roughly USD 1.36 after Etsy's fees, against 19,700 competing listings. Crossed against a Very low term with real prices and real volume, the Very high keyword made about sixty-four times less money available per competing listing. Read the label together with the median price and the listing count.

Why do so many Etsy seller keywords show "Very low" conversion?

Because the seller-facing searches in my sample were research searches, and the information behind them has an abundant free substitute in blog posts, help pages, videos and forum threads. Somebody searching etsy seller guide or etsy seo guide can satisfy the underlying need without buying anything, and many do. Ten of the twenty seller-facing terms I measured came back Very low for this structural reason. Consumer craft searches behave differently because there is no free substitute for a physical object.

Does the Etsy conversion label reflect my own shop's conversion rate?

No. The label is a property of the search, aggregated across every buyer and every listing that appeared for it. Your own conversion rate is a separate number in your own shop stats and must be judged against your own history. A Very high market label and a zero-sales listing are entirely compatible: on a phrase with nearly twenty thousand competing listings, a listing can sit inside that search for the whole window and record no sales at all. Use the label to judge the market and your shop stats to judge your listing.

The conversion label is one of the most useful free signals Etsy hands sellers, and it is also the easiest one to misread, for the same reason: it arrives already looking like a decision. It is not a decision. It is one of four free fields, and the other three are on the same page, one scroll down. Read all four, write your assumption in the open, and the badge stops being a verdict and starts being what it actually is — evidence.