Custom search checklist

This page provides a checklist of the steps required to create acustom search app. For a checklist to create a website search app, seeWebsite search checklist.

If you're new to Vertex AI Search, consider following theGet startedwith Vertex AI Search tutorial tocreate a sample app.

Set up a Google Cloud project, turn on Vertex AI Search, and set up accesscontrol for your project. You can use an existing Google Cloud project if youhave one already.

Actions

  1. ReviewBefore you begin and confirm that you have completed the steps.

Determine what kind of data you'll make available in your search app and prepareit for importing to Vertex AI Search.

You can base your data store on the following types of data:

  • Unstructured data. Documents (such as PDFs, HTML files, and TXTfiles) that are stored in Cloud Storage. Optionally, you can providemetadata in a JSON file or in a BigQuery table.
  • Structured data. Data provided with a specific schema. For example, youcan provide data in a BigQuery table, as JSON files inCloud Storage, or from third-party connectors such as Jira.

Actions

  1. Review the information about supported data and the relationship between appsand data stores inAbout apps and datastores.

  2. Prepare your data according to the requirements inPrepare data foringestion.

  3. If you need to set up access control to limit the data that users can view inyour search app's results, review the prerequisites and follow theinstructions for your identity provider type and data source inUse datasource access control.

Create a data store and then import your data into it, or set up athird-party connector as your synced data source.

How you import your data depends on where you're importing it from. Forexample, if your data is in Cloud Storage, you can import it using theconsole or the API by providing the bucket location of your data.

Actions

  1. Follow the instructions for your data source inCreate a search datastore.

Create your custom search app and connect it to your new data store.

Actions

  1. Create a search app.

Vertex AI Search for custom search supports two pricing models:

  • General: A pay-as-you-go consumption-based model.
  • Configurable: A mixed-pricing model that lets you to unbundle capabilitiesand create a subscription package that is more closely aligned with yourspecific use cases. This pricing model lets you to pay only for the featuresthat you need and benefit from their powerful capabilities. It offers twosubscriptions and optional add-ons. After you enable this pricing model atproject level, you can select configurable pricing for your apps and datastores.

Actions

By default, a project is set to the pay-as-you-go model. However, to switch to theconfigurable pricing, do the following:

  1. Follow the instructions toSet up configurable pricing for custom search.

Vertex AI Search provides many configuration options. Some optionsdepend on whether you plan to deploy a search widget.

Actions

Depending on your use case and whether you plan to deploy the out-of-the-boxsearch widget or integrate search API calls into your own code,Vertex AI Search provides several options for configuration.

  1. You can embed a search widget into your website. The widget automaticallyprovides a search bar and expandable search interface. If you plan to deploythe search widget, configure the following:

  2. If you plan to integrate search API calls into your server or applicationinstead of using the widget, you can configure your search settings usingthe following options:

    • Field settings. For structured data or for unstructured data withmetadata, update field settings to refine how Vertex AI Searchuses metadata for search. SeeConfigure fields forsearch.

    • Autocomplete. Depending on your data, set up autocomplete suggestionsbased on document content, fields, search history, or user events. SeeConfigure autocomplete.

    • Serving controls. Control when search resultsare boosted, buried, filtered, or redirected, or whether certain queriesare associated with other queries. SeeConfigure servingcontrols.

    • Search tuning (Preview). Tune the search model with your own trainingdata. SeeImprove search results with searchtuning.

    • Custom embeddings (Preview). If you've created your own embeddings,you might prefer to use them instead of those generated byVertex AI Search to enrich your searches with additionalcontext. This feature is available for data stores with structured data orunstructured data with metadata. SeeUse customembeddings.

If you plan to deploy your app by integrating search API calls into your owncode, Vertex AI Search provides additional options for configuring howyour search results are returned.

Actions

  1. Configure your search results with the following options:

You can preview your search results to check if your app configurations areworking as expected.

Actions

  1. To preview your search results, use the console or the API.

    • Console. Use the Vertex AI Search consolePreview page to previewhow search widget configurations affect your results. See theConsoleinstructions inGet searchresults.

    • API. If you're integrating API calls into your application, make APIcalls to preview your search configurations. See theRESTinstructions inGet searchresults.

When you're happy with the preview version of your search app, share it withyour users by deploying it to your website.

Actions

  1. You can deploy your search app in either of the following ways:

    • Embed the search widget into your website. Vertex AI Searchprovides code that you can copy into your website or web application. Thisdeploys the search widget. You can preview search results in the console.SeeAdd the search widget to a webpage.

    • Integrate search API calls into your server or application. For fullcontrol over how your search results are displayed, you can integrate APIcalls into your server or applications. For more information about makingAPI calls, seeGet search results. For client library resources, seeVertex AI Search clientlibraries.

To get personalized search and browse results, you can update the user eventsin your search app. For more information seeAbout user events for custom apps.

Actions

  1. Import historical user events.
  2. Record real-time user events.

You can maintain your app to ensure that latest and necessary data is availablein your data store.

Actions

  1. Refresh your data.

Except as otherwise noted, the content of this page is licensed under theCreative Commons Attribution 4.0 License, and code samples are licensed under theApache 2.0 License. For details, see theGoogle Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2026-02-18 UTC.