Collect the matching inputs.
Log module visibility, placement, session and product-click events. Define a consistent viewability rule and count each exposed session once.
← All ecommerce KPIs
Product discovery · Recommendation CTRRecommendation click-through rate.
Recommendation click-through rate.
Measure the suggestions that earn attention.
This guide uses session-based recommendation CTR: the share of sessions exposed to a viewable recommendation module that click at least one recommendation.
Start with your numbers.
Recommendation CTR calculator
01 / The definitionWhat is
What is
recommendation click-through rate?
This guide uses session-based recommendation CTR: the share of sessions exposed to a viewable recommendation module that click at least one recommendation.
An impression-level widget CTR has a different denominator. Moving a module above the fold changes exposure and audience composition.
02 / The measurementA clear formula.
A clear formula.
A useful comparison.
Recommendation CTR =Exposed sessions with a recommendation click ÷ Sessions with a viewable recommendation module × 100
Read the result in context.
Compare the same placement, device and viewability definition. Product-page alternatives and checkout accessories serve different intentions.
Document the comparison.
Keep the reporting dates, population, exclusions and calculation with every result. Show underlying counts as well as the average or rate, so a small sample does not look more conclusive than it is.
Illustrative example
From inputs to insight.
- Exposed sessions with a recommendation click
- 300
- Sessions with a viewable recommendation module
- 5,000
- Recommendation CTR
- 6%
6% of exposed sessions clicked a recommendation. This is not clicks divided by individual product impressions.
Try your numbers ↑03 / Industry benchmarksContext first.
Context first.
Targets second.
Compare the same placement, device and viewability definition. Product-page alternatives and checkout accessories serve different intentions.
We have not verified a public industry benchmark that matches the exact definition used on this page. A precise-looking generic range would hide important differences between businesses.
Build your own benchmark
Start with a comparable baseline.
01
Match the population.
Choose the category, channel or customer cohort you can compare consistently.
02
Let the data mature.
Use complete periods and allow the purchase, attribution or returns window in your definition to close.
03
Test a specific opportunity.
Set a target from your economics and observed variation. Keep conversion and contribution in view.
04 / Clerk customer evidenceReal stories.
Real stories.
Clear measurement limits.
We have not verified a published Clerk customer case with a quantified recommendation click-through rate result matching this page’s definition.
What would count as useful evidence?
A case reporting exposed sessions with a recommendation click and sessions with a viewable recommendation module, with the time window and comparison method. General revenue growth or a related engagement metric does not establish this result.
Customer stories describe individual implementations. They are not industry benchmarks, guarantees or standardized causal tests.
05 / Put it into practice
Measure the suggestions that earn attention.
Match suggestions to the moment: alternatives for comparison and complementary items after the shopper chooses a product.
06 / Common questionsA little more
A little more
clarity.
How do I calculate recommendation click-through rate?
Exposed sessions with a recommendation click ÷ Sessions with a viewable recommendation module × 100. 6% of exposed sessions clicked a recommendation. This is not clicks divided by individual product impressions.
What should I check before comparing results?
An impression-level widget CTR has a different denominator. Moving a module above the fold changes exposure and audience composition.
What is a good recommendation click-through rate?
Compare the same placement, device and viewability definition. Product-page alternatives and checkout accessories serve different intentions.
Does a higher attributed result prove incremental growth?
No. Attribution connects an interaction with an outcome under a reporting rule. To estimate what a change added, compare randomly assigned treatment and control groups using the same eligible population and a predefined measurement window.
Keep learning
Connect the numbers.
Your next opportunityMake product discovery
Make product discovery
work harder for your store.
Explore how Clerk helps shoppers find relevant products throughout their journey.