Search-as-you-type field in Elasticsearch - Step-by-Step Execution

Concept Flow - Search-as-you-type field

User types a letter

↓

Send partial input to Elasticsearch

↓

Elasticsearch matches partial input

↓

Return matching suggestions

↓

Display suggestions to user

↓

User continues typing or selects suggestion

↓

User types a letter

The search-as-you-type field sends partial user input to Elasticsearch, which returns matching suggestions instantly as the user types.

Execution Sample

Elasticsearch

PUT my_index
{
  "mappings": {
    "properties": {
      "title": {
        "type": "search_as_you_type"
      }
    }
  }
}

This code creates an index with a search-as-you-type field called 'title' that supports fast partial matching.

Execution Table

Step	User Input	Query Sent to Elasticsearch	Matching Documents Found	Suggestions Displayed
1	"a"	{"match": {"title": "a"}}	['apple', 'apricot']	['apple', 'apricot']
2	"ap"	{"match": {"title": "ap"}}	['apple', 'apricot']	['apple', 'apricot']
3	"app"	{"match": {"title": "app"}}	['apple', 'application']	['apple', 'application']
4	"appl"	{"match": {"title": "appl"}}	['apple', 'application']	['apple', 'application']
5	"apple"	{"match": {"title": "apple"}}	['apple']	['apple']
6	"apple " (user stops typing)	No new query sent	No change	Suggestions remain ['apple']

💡 User stops typing or selects a suggestion, ending the search-as-you-type interaction.

Variable Tracker

Variable	Start	After 1	After 2	After 3	After 4	After 5	Final
User Input	""	"a"	"ap"	"app"	"appl"	"apple"	"apple"
Query Sent	None	{"match": {"title": "a"}}	{"match": {"title": "ap"}}	{"match": {"title": "app"}}	{"match": {"title": "appl"}}	{"match": {"title": "apple"}}	No new query
Suggestions	[]	['apple', 'apricot']	['apple', 'apricot']	['apple', 'application']	['apple', 'application']	['apple']	['apple']

Key Moments - 3 Insights

Why does Elasticsearch send a new query on every letter typed?

Because the search-as-you-type field is designed to update suggestions instantly as the user types, so each new letter triggers a new query (see execution_table steps 1-5).

What happens if the user stops typing?

No new query is sent and the suggestions stay the same until the user types again or selects a suggestion (see execution_table step 6).

Why do suggestions change as more letters are typed?

Because the query becomes more specific, filtering matches to better fit the typed input (see execution_table rows 1 to 5).

Visual Quiz - 3 Questions

Look at the execution_table at step 3, what suggestions are displayed?

A ['apple', 'application']

B ['apple', 'apricot']

C ['apricot', 'application']

D ['apple']

💡 Hint

Check the 'Suggestions Displayed' column at step 3 in the execution_table.

At which step does the query sent to Elasticsearch become the most specific?

A Step 1

B Step 5

C Step 3

D Step 6

💡 Hint

Look at the 'Query Sent to Elasticsearch' column and find the longest user input.

If the user types 'appl' instead of 'app', how would the suggestions change at step 4?

A Suggestions would include 'apricot' only

B Suggestions would be empty

C Suggestions would include 'apple' and 'application'

D Suggestions would include 'apple' only

💡 Hint

Refer to the suggestions shown at step 4 in the execution_table.

Concept Snapshot

Search-as-you-type field in Elasticsearch:
- Use type 'search_as_you_type' in mapping.
- Sends partial input queries as user types.
- Returns matching suggestions instantly.
- Updates suggestions with each new letter.
- Stops querying when user stops typing or selects.

Full Transcript

This visual execution shows how a search-as-you-type field works in Elasticsearch. When the user types each letter, a query is sent to Elasticsearch with the current partial input. Elasticsearch returns matching documents that contain words starting with that input. Suggestions update instantly as the user types more letters, becoming more specific. If the user stops typing, no new queries are sent and suggestions remain. This helps users find matches quickly by showing results as they type.