Drop-down suggestions
Without query suggestions, users must type a full search term and press Enter before finding out whether results exist. If the term is misspelled or uncommon, they get nothing useful. Drop-down suggestions surface matching query terms as users type, so they can select a validated search with a single click. OpenSearch Industry Algorithm Edition generates these suggestions from actual search traffic on your application, so candidates reflect what your users already search for successfully.
How it works
OpenSearch generates drop-down suggestions through three stages:
Collect — The service aggregates search queries submitted by users of your application over a rolling time window.
Filter — Low-frequency queries, queries that return no results, and any queries on your blocklist are removed from the candidate pool.
Rank — Remaining candidates are sorted by search frequency over the past 30 days, with the most popular suggestions appearing first.
The result is a suggestion list that reflects what your actual users search for, not a static keyword dictionary.
Use cases
Cold-start guidance — Direct new users toward popular or high-converting search terms before they know what to type.
Reducing zero-result searches — Surface validated queries so users avoid searches that return no results.
Accelerating repeat searches — Let returning users re-run frequent searches with a single click instead of retyping.
Enable drop-down suggestions
Before you begin, ensure that you have:
An active OpenSearch Industry Algorithm Edition instance
Search analytics data — drop-down suggestions require a minimum volume of recorded queries to generate meaningful candidates
The required permissions to modify your OpenSearch application configuration
To enable and configure drop-down suggestions:
Log in to the OpenSearch console.
Click the name of your application to open its configuration page.
In the left navigation pane, click Search guidance.
On the Drop-down suggestions tab, toggle the feature to Enabled.
Configure the following settings:
Setting Description Minimum query frequency The minimum number of times a query must appear in search history before it becomes a suggestion candidate. Increase this value to surface only well-established search terms. Suggestion count The maximum number of suggestions displayed in the drop-down list. Blocklist A list of queries to exclude from suggestions regardless of frequency. Click Save.
Changes take effect within a few minutes. New suggestions appear as your analytics data accumulates.
Troubleshooting
No suggestions appear in the drop-down list
The most common cause is insufficient search history. Drop-down suggestions require enough query volume to pass the minimum frequency threshold. If the feature was recently enabled or your application has low traffic, wait until more search data accumulates before expecting suggestions to appear.
If suggestions still do not appear after sufficient traffic has been recorded, check the following:
Confirm the feature is set to Enabled on the Drop-down suggestions tab.
Verify that the minimum query frequency is not set too high for your traffic volume.
Check whether the missing queries are on your blocklist.
Suggestions are irrelevant or outdated
Suggestions reflect the past 30 days of search traffic. If your catalog or user intent has changed recently, the suggestion pool will update naturally as older queries drop out of the ranking window. To remove specific unwanted suggestions immediately, add them to the blocklist.
What's next
Query correction — Automatically correct misspelled queries to improve result relevance.
Personalized ranking — Reorder search results based on individual user behavior.