
In short
- CTR (clicks ÷ impressions) shows whether shoppers want to click your ad, and conversion rate (orders ÷ clicks) whether they buy after clicking.
- Conversion rate affects ACOS directly: ACOS is CPC divided by price times conversion rate.
- Read them together and compare them with your own products, terms and history, not with a borrowed benchmark.
- Before calling a keyword a loser, wait for two to three times the clicks an order usually takes.
Click-through rate tells you whether shoppers want to click your ad, and conversion rate tells you whether they buy once they have clicked. Read together, they point to where a campaign is failing: the search term, the main image and price, or the product page.
Each one alone can send you the wrong way. This guide shows how to calculate both, how they connect to ACOS, and how many clicks you need before you can trust what they say.
What are CTR and conversion rate on Amazon?
CTR is clicks divided by impressions. Conversion rate is orders divided by clicks. Both are percentages, and they measure two consecutive steps of the same shopper's path.
- CTR = clicks ÷ impressions × 100. It reflects what the shopper sees in the results: main image, title, price, rating and how well all that fits the search.
- Conversion rate = orders ÷ clicks × 100. It reflects what happens on the product page: content, reviews, price compared with the alternatives, delivery and availability.
Here is an invented example for one keyword: 20,000 impressions, 100 clicks and 8 orders. CTR is 0.5% and conversion rate is 8%.
Some reports show units or attributed orders over different windows, so use the same definition every time you compare. Otherwise you will see a change that only exists in the report.
How do CTR and conversion rate affect ACOS?
Conversion rate affects ACOS directly, through a simple formula. CTR affects it only indirectly.
Formula
ACOS = CPC ÷ (price × conversion rate)
Example
This comes straight from the definition Amazon Ads uses for ACOS: spend divided by attributed sales. Continuing the example, with a $1.00 CPC and a $40 price:
- Spend: 100 clicks × $1.00 = $100.
- Attributed sales: 8 orders × $40 = $320.
- ACOS: $100 ÷ $320 = 31.25%. The formula gives the same: $1.00 ÷ ($40 × 0.08).
Now raise conversion rate from 8% to 10% with nothing else changed, and ACOS falls to 25%. That is why improving a product page can do as much for ACOS as cutting a bid, and without the loss of traffic.
CTR does not appear in the formula, because you pay per click and not per impression. It matters because it decides how much traffic you can get from a search term, and a poor CTR on a relevant term is often the first sign that your offer looks weak next to the others on the page.
How do you read CTR and conversion rate together?
Put each search term or product in one of four boxes. Each one leads to a different first check, and none of them is solved by the same action.
| CTR | Conversion rate | Likely reading | What to check first |
|---|---|---|---|
| Good | Good | Relevant term and a solid offer | Whether you can take more volume within your target ACOS |
| Good | Low | Shoppers are interested, but the page or the term disappoints them | Whether the term matches the product, then price, reviews and content |
| Low | Good | Few click, but those who do buy | Main image, price shown in the results, and placement |
| Low | Low | The term or the offer is not a good fit | Relevance of the term before anything else; consider a negative |
"Good" and "low" should mean good or low for you, compared with your own products, terms and history. Search terms vary widely by category and intent, so a borrowed benchmark says little. A branded term and a generic one can differ enormously inside the same campaign, and both can be perfectly healthy.
How many clicks do you need before judging a keyword?
You need enough clicks that having no orders would actually be surprising. With too few, conversion rate is mostly chance.
Common mistake
Suppose a product usually converts 8% of clicks. A keyword with 12 clicks and no orders looks bad, but at that rate there is about a 37% chance of seeing exactly that by pure luck: 0.92 multiplied by itself 12 times. Pausing it now would mean judging on a coin flip.
A practical way to set your threshold:
- Take the product's usual conversion rate and work out how many clicks one order normally takes. At 8% that is 12.5 clicks.
- Wait for two to three times that number before calling a keyword a loser, which here means roughly 25 to 40 clicks.
- Check the cost of waiting. If those clicks would cost more than you are willing to risk, lower the bid instead of pausing, so that the test costs less.
- For a keyword that seems to be winning, be just as careful. Two orders from 10 clicks is encouraging, but it is not a 20% conversion rate you can plan around.
Tip
Group data while you wait. An ad group or product has more clicks than any single keyword, so its conversion rate is steadier and a better starting assumption.
What should you change when CTR is low?
Check relevance first, then the elements shoppers see before they click. Raising the bid is often the last thing to try, because it buys more visibility for an offer shoppers are already skipping.
- Relevance: read the actual search terms in your report. Terms that do not describe the product belong in your negatives.
- Main image: compare it with the competing results for that search on a phone screen, which is how many shoppers will see it.
- Price and rating: a visibly higher price or weaker rating will lower CTR, and no bid can fix that.
- Placement: the same keyword can behave very differently at the top of search and on product pages. Read the placement report before drawing a conclusion from the total.
What should you change when conversion rate is low?
Separate search terms from the product. If conversion is low only on some terms, the problem lies in those terms. If it is low everywhere, the problem lies in the product page or the offer.
- Low on some terms: these shoppers wanted something else. Reduce the bid or add a negative, and keep your budget on the terms that fit.
- Low on all terms: look at price compared with alternatives, the number and recency of reviews, the first images, delivery promise and stock.
- Sudden drop: look for what changed. A new competitor, a lost Buy Box, a price rise, a poor recent review or a delivery change are more common causes than the campaign itself.
- Drop after a bid increase: you may have moved into placements or broader queries that convert less. Compare by placement and search term.
Change one thing at a time and give it time. Selloop evaluates each applied change from day 8 and again from day 15, and you can use those same two checkpoints by hand. Without a date and a starting value, you will not know which change did what.
Frequently asked questions
What is a good CTR on Amazon?
There is no universal figure, because it depends on category, placement and how specific the search is. Compare a term with your own similar terms, and with its own history.
What is a good conversion rate on Amazon?
The useful question is whether it is high enough to reach your target ACOS at the CPC you pay. Use the formula: ACOS equals CPC divided by price times conversion rate.
Is a high CTR always good?
No. Clicks cost money, so a high CTR with a low conversion rate can be the most expensive combination you have. Read CTR next to conversion rate.
Why is my CTR low even though my ad ranks at the top?
The term may not be as relevant as it looks, or your image, price or rating may compare badly with the results around it. Check the real search terms and the results page.
Should I pause a keyword with clicks but no orders?
Only after enough clicks. Work out how many clicks an order normally takes and wait for two to three times that number. Lowering the bid first is often cheaper than losing a good term.
Selloop syncs your Sponsored Products campaigns, proposes changes, and you apply them in Amazon with one click. Start your 7-day free trial.