Skip to main content
Getting started with Analytics

Get to know key metrics and benefits for Search and Category Merchandising Analytics

Paulius Grigaitis avatar
Written by Paulius Grigaitis
Updated over a month ago

Search Analytics lets you know what your users search, how they behave in search, and general metrics to feel the pulse of the value coming from Nosto’s Search.

In this article, we take a closer look at different sections within the analytics and what certain metrics stand for.

Analytics structure

Enhanced analytics provide with the clear breakdown of performance data in three separate sections:

  1. Search;

  2. Search Autocomplete;

  3. Category Merchandising.

Key Metrics: Search

Feature-level statistics

  • Total Searches - The total count of search impressions. Actions contributing to search impressions include:

    1. Submitting a Search Term: Each time a search term is submitted, it counts as a separate search impression. This includes repeated submissions of the same term within the same session. For example, submitting the same search term five times in a session will count as five search impressions.

    2. Applying a Filter: Each filter applied adds a search impression. For instance, submitting a search term like "Shorts" and then filtering by "Size M" would result in two search impressions.

    3. Pagination: Navigating to another page, such as moving from page 1 to page 2, generates an additional search impression. For example, submitting "Shorts" as a search term and then moving to the second page would count as two search impressions. Adding a filter on top of this would increase the total to three search impressions.

    4. Changing the Sorting Option: Selecting a different sorting option is also counted as a search impression. For example, submitting "Shorts" and then sorting by "Price: Low to High" would result in two search impressions.

  • Orders - Total number of orders generated from searches, including Autocomplete interactions

  • Sales - Total sales generated through search, including Autocomplete interactions;

  • CTR % (Click-through rate) - The percentage of searches that resulted in clicks in the search results page;

  • CR % (Conversion rate) - The percentage of visits that led to orders.

All search terms

  • CTR % (Click-through rate) -The percentage of searches that resulted in clicks for the specific search term;

  • Orders - The number of orders generated from searches with this term;

  • CR % (Conversion rate) - The percentage of visits that resulted in orders from the specific search term.

  • Share % - the percentage of searches for a specific term compared to the total number of all searches. In other words, it represents this search term’s share of all searches.

Top clicked products

  • Clicks - Total number of clicks received by the product in search results;

  • CR % - The percentage of visits that led to orders for the specific product in search results.

  • Share % - the percentage of total product clicks that the product received, compared to clicks on all products. In other words, it represents this product's share of all tracked product clicks in search.

Top purchased products

  • Orders - Total number of orders placed for the product through search results;

  • Sales - Total number of sales products generated through search results.

Organic vs. Non-Organic Searches, excluding Autocomplete interactions

  • Organic - The percentage of searches initiated by directly visiting the site and using search;

  • Non-organic - The percentage of searches that were driven by promotions or external influences and led to the search results page (e.g. clicks on banners or Google links leading directly to certain search results pages).

Device type

  • Device - The percentage of the type of device used for performing searches (e.g., mobile, desktop).

Key Metrics: Categories

Feature-level statistics

  • Total Category Views - The total number of category views (impressions). Actions that contribute to category views include:

    1. Accessing a Category: Each time a category is viewed, it is counted as a separate category view (impression), even if the same category is revisited within the same session. For example, navigating to the same category five times in a session would result in five category views.

    2. Applying a Filter: Each filter applied within a category view adds another category view. For instance, navigating to a category like "Shorts" and then applying a filter for "Size M" would result in two category views.

    3. Pagination: Moving to another page within a category view, such as navigating from page 1 to page 2, generates an additional category view. For example, accessing the "Shorts" category and then going to the second page of results would count as two category views. Adding a filter on top of this would increase the total to three category views.

    4. Changing the Sorting Option: Selecting a different sorting option within a category also counts as a category view. For example, accessing the "Shorts" category and then selecting a sort option like "Price: Low to High" would result in two category views.

  • Orders - Total number of orders made through category views;

  • Sales - Total sales generated through categories;

  • CTR % (Click-through rate) - The percentage of category views that resulted in clicks;

  • CR % (Conversion rate) - The percentage of visits to categories that led to orders.

All categories

  • Views - Total number of views for the category;

  • CTR % (Click-through rate) - The percentage of category views that led to clicks;

  • Orders - The number of orders placed through the specific category;

  • CR % (Conversion rate) - The percentage of visits to the category that resulted in orders;

  • Share % - the percentage of views for a specific category relative to the total views across all categories. In other words, it represents this category's portion of all category views.

Top clicked products

  • Clicks - Total number of clicks received by the product in categories;

  • CR % - The percentage of visits that led to orders for the product in categories;

  • Share % - the percentage of total product clicks that the product received, compared to clicks on all products. In other words, it represents this product's share of all tracked product clicks in categories.

Top purchased products

  • Orders - Total number of orders placed for the product through categories;

  • Sales - Total number of sales products generated through categories.

Device type

  • Device - The percentage of the type of device used to view categories (e.g., mobile, desktop).

Key Metrics: Autocomplete

Feature-level statistics

  • Product clicks - Total clicks on autocomplete suggestions that led to product views;

  • Keyword clicks - Total impressions from keyword searches in autocomplete;

  • Usage - The percentage of total impressions from both keyword searches and product clicks in autocomplete within total search usage.

All keywords

  • Clicks - Total number of clicks generated from searches using specific autocomplete keyword;

  • Orders - Number of orders generated from searches using specific keyword;

  • CR % (Conversion rate) - The percentage of visits that resulted in orders from the unique keyword.

  • Share % - the percentage of clicks for a specific autocomplete keyword suggestion relative to the total number of autocomplete clicks. In other words, it represents this keyword suggestion's share of all autocomplete clicks.

Top clicked products

  • Clicks - Total number of clicks received by the product from autocomplete suggestions;

  • CR % - The percentage of visits to the product from autocomplete that resulted in orders.

  • Share % - the percentage of total product clicks that the product received, compared to clicks on all products. In other words, it represents this product's share of all tracked product clicks within autocomplete.

Data Export

Nosto analytics provide a possibility to export different data points allowing you to perform data analysis out of the platform. Before exporting make sure relevant data ranges are selected and export relevant tables as .csv files. Once ready, the link to download the export will be shared to the email.

Note: Export link is available only for a short period of time.

We recommend merchants have a monthly routine to go through these and improve search based on that. Remember that many relevance challenges can be easily tackled by improving Searchable Fields or with merchandising rules targeting specific or all search queries.

Data Attribution

Some important points to understand on how certain data points and metrics are calculated:

Same search terms appearing in both "All search terms" and "No results search terms" tables: For example, if the search term “Shirt” appears in the "All search terms" table with a count of 20 and in the "No results search terms" table with a count of 10, this indicates that in 50% of cases, the search term did not return any results. There could be multiple reasons for search terms appearing in both tables, including:

  • Fluctuating availability of the product;

  • The "Filter" functionality in merchandising rules may result in no search results during certain periods. For example, if a rule with a filter is applied, it may cause a search term to return no results if the filter eliminates all matching products. This can occur during A/B tests or when specific segments are applied, affecting only certain visitor groups rather than the entire audience;

  • Changes to the product catalog, such as product discontinuation.

Difference in the sum of orders from the "All search terms" table and the "Statistics" table: This difference occurs for several reasons:

  • A single order can be attributed to multiple search terms within the same session, causing the sum of orders from the "All search terms" table to be higher than the order count in the "Statistics" table.

  • The "All search terms" table represents the top 500 searched terms, meaning that rarer search terms are not included, and their orders, if any, are not considered.

  • As noted in the tooltip, orders in the "Statistics" table include orders generated through both Search and Autocomplete, whereas the "All search terms" table attributes orders specifically to Search.

  • Pagination, Filtering, and Sorting Options: Each action—pagination, filtering, and sorting—is counted as a separate search query or category view. For example, searching for "Pants" or navigating to the "Pants" category, then applying a filter such as "Type=Shorts," would count as two search queries or category views. Similarly, searching for "Pants" or accessing the "Pants" category and then moving to the second page of results would also be counted as two queries or views. Selecting an additional sorting option, such as "Price: Low to High," would bring the total to three search queries or category views.

  • Number of search terms presented: Search analytics return the top 500 search terms based on count. Therefore, any sorting in the "All search terms" table applies specifically to the 500 most frequently searched terms.

Did this answer your question?