This chapter covers SharePoint Search administration, a key topic in the MS-102 exam's tenant management domain (Objective 1.5). Search is how users find content across SharePoint, OneDrive, and Microsoft 365. Approximately 5-10% of exam questions touch on search topology, crawl schedules, result sources, query rules, and managed properties. Understanding these components is critical for ensuring content discoverability and performance.
Jump to a section
Imagine a large public library with millions of books. The library has a card catalog system that lists every book by title, author, and subject. When a patron searches for 'quantum physics,' the catalog returns a list of matching books with their shelf locations. However, if a new book arrives but the catalog isn't updated for a day, that book won't appear in search results until the update. Similarly, SharePoint Search uses a crawl process to index content, and the index is updated on a schedule. The catalog is the search index, the librarian's daily update is the crawl schedule, and the patron's query is the search request. But unlike a simple card catalog, SharePoint Search also ranks results based on relevance, like a librarian recommending the most popular or authoritative books first. If the catalog is poorly maintained (crawl not covering all content), patrons miss books. If the ranking algorithm is flawed, irrelevant books appear first. The analogy breaks down with dynamic content and security trimming, where the library would need to check each patron's borrowing privileges before showing results—SharePoint does this by filtering results based on the user's permissions at query time.
What is SharePoint Search and Why It Exists
SharePoint Search is the engine that indexes content from SharePoint sites, OneDrive, and external sources, and returns relevant results to users. It is built on the SharePoint Search service application, which consists of a crawl component, content processing, index, query processing, and analytics. The search system is essential for users to find documents, list items, pages, and people across the organization. Without proper administration, search can become slow, incomplete, or return irrelevant results.
How Search Works Internally: The Crawl-Index-Query Pipeline
The search process follows a pipeline:
Crawl: The crawl component connects to content sources (SharePoint sites, file shares, etc.) and sends requests to retrieve content. It uses a protocol handler for each type of content (e.g., HTTP for SharePoint, SMB for file shares). The crawler parses the content and extracts metadata and full text.
Content Processing: The crawled content is sent to the content processing component, which transforms it into an indexable format. This includes parsing document formats (PDF, Office), extracting terms, applying language detection, and generating managed property values from crawled properties.
Index: The processed content is written to the search index, which is a compressed, inverted index optimized for fast retrieval. The index is distributed across index partitions and replicas for scale and redundancy.
Query Processing: When a user submits a query, the query processing component parses the query, applies query rules (e.g., synonyms, result blocks), consults the index, and applies ranking algorithms (e.g., relevance based on click-through data, freshness, metadata). Results are then security-trimmed based on the user's permissions and returned.
Key Components, Values, Defaults, and Timers
Crawl Schedules: Two types – continuous crawl (incremental every 1-15 minutes) and incremental/full crawl (scheduled). Default continuous crawl interval is 5 minutes for SharePoint Online. For on-premises, full crawl default is daily, incremental every 15 minutes.
Content Sources: Each content source has a start address (URL), crawl schedule, and crawl rules. Up to 50 content sources per Search Service Application.
Crawl Rules: Define what to include/exclude. Can be based on URL path, content type, or metadata. Up to 100 crawl rules per content source.
Managed Properties: Mapped from crawled properties. There are approximately 1,000 managed properties available. Default managed properties include Title, Author, Created, Modified, ContentType, etc. Custom managed properties can be created.
Result Sources: Define where to search (local SharePoint, remote SharePoint, OpenSearch, etc.). Default result sources: Local SharePoint Results, People Results, Videos, Conversations. Up to 500 result sources.
Query Rules: Allow promoting or demoting results based on conditions (e.g., user segment, query term). Each rule can have multiple conditions and actions (e.g., show a result block, change ranking). Up to 100 query rules per search application.
Search Schema: The mapping between crawled and managed properties. Managed properties can be configured for searchable, retrievable, sortable, refinable, etc.
Analytics: Tracks clicks and impressions to influence ranking. The analytics processing component runs every 24 hours.
Configuration and Verification Commands
For SharePoint Online, administration is via SharePoint Admin Center or PowerShell. Key PowerShell cmdlets:
# Get search configuration
Get-SPEnterpriseSearchServiceApplication
Get-SPEnterpriseSearchCrawlContentSource
Get-SPEnterpriseSearchCrawlRule
Get-SPEnterpriseSearchMetadataManagedProperty
Get-SPEnterpriseSearchQueryRule
Get-SPEnterpriseSearchResultSource
# Create a new content source
New-SPEnterpriseSearchCrawlContentSource -Name "My Content Source" -Type SharePoint -StartAddress "https://contoso.sharepoint.com/sites/mysite" -SearchApplication $ssa
# Start a full crawl
Start-SPEnterpriseSearchCrawlContentSource -Identity "My Content Source" -FullCrawl
# Create a managed property
New-SPEnterpriseSearchMetadataManagedProperty -Name "CustomProperty" -Type Text -SearchApplication $ssa
# Map crawled property to managed property
New-SPEnterpriseSearchMetadataMapping -SearchApplication $ssa -ManagedProperty "CustomProperty" -CrawledProperty "MyCrawledProperty"
# Create a query rule
New-SPEnterpriseSearchQueryRule -SearchApplication $ssa -QueryRuleName "Promote Sales" -QueryCondition "sales" -IsAdvanced $true -RuleType 1 -Context "SPSite" -ContextValue "https://contoso.sharepoint.com/sites/sales"For verification, use:
# Check crawl status
Get-SPEnterpriseSearchCrawlContentSource | Select Name, CrawlStatus, CrawlCompletedTime
# Test search query
$query = New-SPEnterpriseSearchQuery -SearchApplication $ssa -QueryText "test"
$result = $query.Execute()
$result | Select Title, Path, RankHow Search Interacts with Related Technologies
Microsoft Graph: SharePoint Search uses Microsoft Graph for people search and to index content from Exchange and Teams. The unified search index in Microsoft 365 combines results from SharePoint, OneDrive, Exchange, and Teams.
Azure Active Directory: User profiles and permissions from AAD are used for security trimming. If a user doesn't have access to a document, it won't appear in search results even if it matches the query.
OneDrive: OneDrive content is crawled similarly to SharePoint sites. Users can search their own OneDrive and shared content.
Hybrid Search: In hybrid deployments, search results can be federated between on-premises and cloud. This requires configuring a result source pointing to the other environment and setting up cloud search service application (CSSA) for reverse hybrid.
Search Center: A specialized site collection that provides advanced search features like search scopes, search box suggestions, and search verticals (e.g., People, Conversations).
Search Web Parts: Content Search Web Part (CSWP) and Search Results Web Part can display search results on pages. They use a query and display template.
Performance and Scale Considerations
Crawl Impact: Crawling consumes server resources. In SharePoint Online, Microsoft manages the crawl infrastructure. On-premises, you should size crawl components based on content volume (e.g., one crawl component per 10 million items).
Index Size: Each index partition can handle up to 10 million items. For larger farms, multiple partitions are needed. Default is 1 partition per index component.
Query Latency: Target under 1 second for simple queries. Complex queries with many refiners or large result sets can take longer.
Replicas: Index replicas provide redundancy and query throughput. Each replica is a copy of the index partition. For high availability, have at least 2 replicas per partition.
Common Issues and Troubleshooting
Content not appearing in search: Check crawl status, crawl rules (maybe excluding content), security trimming, and index freshness.
Slow search: Review query rules (too many rules), ranking complexity, index fragmentation (rebuild index), or hardware resources.
Irrelevant results: Adjust ranking by boosting managed properties (e.g., Author weight), use query rules to promote important content, or refine the search schema.
Crawl errors: Check crawl logs for access denied, timeout, or protocol errors. Ensure the crawl account has read permissions.
Security Trimming in Detail
Security trimming is the process of filtering search results based on the user's permissions. SharePoint Search achieves this by storing the ACL (access control list) for each item in the index. When a query is executed, the search engine retrieves the ACL of the user and compares it to the ACL of each candidate result. Only items where the user has at least read access are returned. This happens at query time, not crawl time, so changes to permissions are reflected immediately without recrawling. However, if an item's ACL changes and the item is not recrawled, the old ACL remains in the index, which could cause over-trimming (user doesn't see item they should) or under-trimming (user sees item they shouldn't). To avoid under-trimming, SharePoint Online uses a security update mechanism that updates ACLs in the index without a full recrawl. For on-premises, incremental crawls pick up ACL changes.
Search Schema Deep Dive
Managed properties are the backbone of search customization. Each managed property has attributes:
- Type: Text, Integer, Decimal, DateTime, YesNo, etc.
- Searchable: The property's value is included in full-text search.
- Retrievable: The property's value can be returned in search results.
- Sortable: Results can be sorted by this property.
- Refinable: Users can filter results by this property using refiners.
- Multivalued: The property can have multiple values (e.g., multiple authors).
- Queryable: The property can be used in search queries (e.g., author:"John").
Crawled properties are automatically discovered during crawl. They have names like ows_MyField for SharePoint list columns. You map crawled properties to managed properties. Up to 20 crawled properties can be mapped to one managed property. The mapping can include a custom processing instruction (e.g., combine values, extract substring).
Query Rules and Result Blocks
Query rules allow you to promote specific results when certain conditions are met. Conditions can be based on:
Query text (exact match or contains)
User segment (e.g., department, location)
Source (e.g., specific result source)
File type
Search vertical
Actions include: - Promote results: Add specific URLs to the top of results. - Add result block: Display a block of results from a predefined query (e.g., show all documents from the Sales library). - Change ranking: Boost or demote results based on a property. - Change query: Modify the query (e.g., add a filter).
Query rules are evaluated in order of priority (0-100, lower number = higher priority). They can be active or inactive. You can also set a start and end date for time-bound promotions.
Result Sources
Result sources define the scope of a search query. They can point to:
Local SharePoint (default)
Remote SharePoint (another farm)
OpenSearch 1.0/1.1 (external search engines like Elasticsearch)
People (for people search)
Video (for video results)
Conversations (from Yammer/Teams)
You can create custom result sources with specific provider URLs or query transforms. For example, a result source that searches only the Sales site collection:
{?{searchTerms} site:https://contoso.sharepoint.com/sites/sales}Search Vertical
Search verticals are tabs on the search results page that limit results to a specific result source. Out-of-the-box verticals: All, People, Conversations, Videos. You can create custom verticals (e.g., Contracts, Policies). Each vertical is associated with a result source and can have its own display template.
Define Content Sources and Crawl Schedules
Content sources specify which URLs to crawl. In the SharePoint Admin Center, navigate to Search > Content Sources. Click 'New Content Source' and enter a name, start address (e.g., https://yourtenant.sharepoint.com), and crawl schedule. For continuous crawl, set the interval between 1 and 15 minutes. For incremental/full crawl, specify time and recurrence. The crawler will start with the start address and follow links within the same site collection. Ensure the crawl account has read access to all content. After creation, you can start a full crawl immediately or wait for the schedule. Monitor crawl status in the same page.
Configure Crawl Rules
Crawl rules control what content is included or excluded during crawling. From Search Administration, click 'Crawl Rules' then 'New Crawl Rule'. Enter a path (e.g., https://tenant.sharepoint.com/sites/archived) and choose to include or exclude. You can also exclude based on content type (e.g., .mp4 files) or metadata (e.g., items with a specific managed property). Rules are evaluated in order; the first matching rule applies. Use exclusion rules to avoid crawling irrelevant or large binary files. After adding rules, you must start a full crawl for changes to take effect. Crawl rules do not affect content already indexed; a full crawl removes excluded items from the index.
Manage Search Schema and Managed Properties
The search schema maps crawled properties to managed properties. In the SharePoint Admin Center, go to Search > Search Schema. You can view crawled properties, managed properties, and their mappings. To create a managed property, click 'New Managed Property', give it a name, select type (Text, Integer, etc.), and configure searchable, retrievable, sortable, refinable, and queryable flags. Then add a mapping to one or more crawled properties. For example, map the crawled property 'ows_SalesRegion' to a managed property 'SalesRegion'. After mapping, you must recrawl content that contains the crawled property. Managed properties can be used in search queries, refiners, and display templates.
Create and Manage Result Sources
Result sources define where search results come from. Navigate to Search > Result Sources and click 'New Result Source'. Enter a name, select protocol (Local SharePoint, Remote SharePoint, OpenSearch, etc.), and specify the source URL. For Local SharePoint, you can add a query transform to limit scope (e.g., only a specific site). For OpenSearch, provide the provider URL. You can also set the result source as default for a search vertical. After creation, you can associate it with a search vertical or use it in query rules. For example, create a result source 'SalesDocuments' that searches only the Sales site collection with a query transform: {?{searchTerms} site:https://tenant.sharepoint.com/sites/sales}.
Implement Query Rules for Result Promotion
Query rules allow you to promote or demote specific results based on conditions. Go to Search > Query Rules and click 'New Query Rule'. Enter a name, set priority (lower number = higher priority), and define conditions (e.g., query contains 'urgent'). Under actions, you can 'Promote results' by adding URLs, 'Add result block' with a predefined query, or 'Change ranking' by boosting a managed property. For example, create a rule that promotes the company intranet page when users search 'home'. Test the rule by performing a search as the target user segment. Rules can be time-bound with start and end dates. Remember that rules are evaluated in priority order; only the first matching rule applies.
Monitor Crawl and Search Performance
Regular monitoring ensures search is healthy. In SharePoint Admin Center, check crawl status (idle, crawling, paused, error). Review crawl logs for errors (access denied, timeout). Use PowerShell to export crawl errors: Get-SPEnterpriseSearchCrawlContentSource | Get-SPEnterpriseSearchCrawlLog -FullCrawl. For query performance, use search analytics to review top queries, click-through rates, and zero-result queries. In on-premises, monitor search component health via Central Administration. If crawl is slow, consider adjusting crawl schedules or adding crawl components. If queries are slow, check index fragmentation and consider index rebuild. Set up alerts for crawl failures.
Enterprise Scenario 1: Global Law Firm with Strict Security Requirements
A global law firm with 10,000 users across 50 offices uses SharePoint Online for document management. Each client matter is a separate site with unique permissions. The firm needs search to return only documents the user has access to (security trimming). They configure continuous crawl every 5 minutes to keep index fresh. However, they notice that when a new associate joins a matter, it takes up to 5 minutes for documents to appear in search. To reduce this, they set continuous crawl to 1 minute, but Microsoft throttles at 1 minute minimum. They also create a query rule that promotes the most recent documents in the user's primary practice area. The search administrator regularly reviews crawl logs for access denied errors, which often indicate permission misconfigurations. They also use managed properties like 'ClientMatterID' to allow users to refine by client.
Enterprise Scenario 2: Large University with Hybrid Search
A university with 50,000 students and staff runs SharePoint on-premises for administrative systems and SharePoint Online for student collaboration. They need a unified search experience. They configure a cloud hybrid search: on-premises content is crawled by the cloud search service application (CSSA) and indexed in Office 365. The on-premises farm has 10 million items, so they create 2 index partitions with 2 replicas each. They set incremental crawl every 15 minutes and full crawl weekly. They create a result source in SharePoint Online that queries the on-premises index via the cloud hybrid search endpoint. They also create query rules that boost results from the online environment for current student content. A common issue is that on-premises content may not appear if the hybrid connection fails; they monitor the hybrid search topology using PowerShell.
Enterprise Scenario 3: Retail Company with Custom Search Vertical
A retail company with 5,000 employees uses SharePoint Online for product catalogs, sales playbooks, and training materials. They want a search vertical for 'Products' that only searches the product catalog site. They create a result source with query transform: {?{searchTerms} site:https://tenant.sharepoint.com/sites/products}. Then they create a search vertical in the Search Center site and assign the result source. They also create a managed property 'ProductCategory' and map it to a crawled property from a list column. Users can refine by category. To ensure new products appear quickly, they configure continuous crawl for the product site. They also set up a query rule that promotes top-selling products when users search for a category. The search team regularly reviews analytics to see which queries return zero results and adds synonyms for common misspellings.
What MS-102 Tests on SharePoint Search Administration
The exam focuses on Objective 1.5: Manage search administration. Key sub-objectives include:
Configure content sources and crawl schedules
Manage search schema (managed properties, crawled properties, mappings)
Create and manage result sources
Implement query rules
Understand security trimming
Manage search center settings
Common Wrong Answers and Why Candidates Choose Them
Confusing crawl schedules: Candidates often think continuous crawl can be set to less than 1 minute. The minimum is 1 minute. They also confuse incremental vs. full crawl – incremental picks up changes, full crawl rebuilds the entire index.
Believing managed properties are automatically created: Candidates think all crawled properties become managed properties automatically. In reality, only a subset of crawled properties are automatically mapped to existing managed properties; custom mappings require manual creation.
Assuming query rules apply to all users: Query rules can be scoped to user segments (e.g., department). Candidates forget to check the user segment condition and wonder why a rule doesn't fire for a specific user.
Thinking security trimming is done at crawl time: Security trimming happens at query time, not crawl time. Candidates may think recrawling is needed to update permissions; in fact, permission changes are reflected immediately (in Online) without recrawl.
Specific Numbers and Terms That Appear on the Exam
Continuous crawl interval: 1-15 minutes (default 5)
Maximum content sources: 50 per Search Service Application
Maximum crawl rules: 100 per content source
Maximum managed properties: approximately 1,000
Maximum query rules: 100 per search application
Result sources limit: 500
Crawl rule evaluation order: first match applies
Priority for query rules: 0-100, lower number = higher priority
Edge Cases and Exceptions
Hybrid search: The exam may ask about cloud hybrid search (on-premises content crawled by Office 365) vs. hybrid federated search (results from both environments merged).
Search Center site: The Search Center is a special site collection that must be provisioned. It is not created by default in SharePoint Online.
Display templates: Managed properties used in display templates must be marked as retrievable.
Crawl rules and wildcards: Crawl rules support wildcards (*) in paths. For example, https://*sharepoint.com/sites/archive/* excludes all archive sites.
Query transform: In result sources, the query transform uses KQL syntax. The default is {?{searchTerms}}. You can add site, filetype, etc.
How to Eliminate Wrong Answers
If a question asks about making a property searchable, the answer involves setting the searchable flag on the managed property, not re-crawling.
If a question asks about excluding a site from search, the answer is a crawl rule, not a query rule.
If a question asks about promoting results for a specific query, the answer is a query rule, not a result source.
If a question asks about security trimming, the answer is that it happens at query time based on the user's token.
Continuous crawl interval minimum is 1 minute (default 5).
Maximum 50 content sources and 100 crawl rules per content source.
Managed properties must be manually created and mapped from crawled properties.
Query rules are evaluated in priority order (0-100, lower = higher).
Security trimming happens at query time, not crawl time.
Result sources can limit search scope via query transforms.
Full crawl is needed after schema changes or to rebuild index.
Search Center site must be provisioned separately in SharePoint Online.
User segments in query rules are based on Active Directory properties.
These come up on the exam all the time. Here's how to tell them apart.
Continuous Crawl
Runs automatically at intervals of 1-15 minutes.
Picks up changes since the last crawl.
Does not require scheduling; runs continuously.
Lower overhead than full crawl.
Recommended for SharePoint Online.
Incremental Crawl
Runs on a scheduled basis (e.g., every 15 minutes).
Picks up changes since the last incremental or full crawl.
Requires scheduling; not continuous.
Higher overhead than continuous crawl but lower than full crawl.
Used in on-premises or when continuous crawl is disabled.
Managed Property
Used in queries, refiners, and display templates.
Must be created manually or automatically from crawled properties.
Has attributes like searchable, retrievable, sortable, refinable.
Can be mapped from multiple crawled properties.
Up to ~1,000 managed properties per farm.
Crawled Property
Discovered automatically during crawl.
Cannot be used directly in queries or refiners.
Named like 'ows_FieldName' for SharePoint fields.
Must be mapped to a managed property to be useful.
Unlimited number; only a subset are mapped.
Mistake
Continuous crawl can be set to any interval, including 30 seconds.
Correct
The minimum interval for continuous crawl is 1 minute. Values below 1 minute are not allowed.
Mistake
All crawled properties are automatically mapped to managed properties.
Correct
Only a subset of crawled properties are automatically mapped. Most require manual mapping to a managed property.
Mistake
Query rules affect all users equally.
Correct
Query rules can be scoped to user segments (e.g., department, location) and only apply to users matching the condition.
Mistake
Security trimming occurs at crawl time, so recrawling is needed after permission changes.
Correct
Security trimming occurs at query time. Permission changes are reflected immediately without recrawling (in SharePoint Online).
Mistake
A full crawl is required to see new content in search results.
Correct
An incremental crawl or continuous crawl is sufficient to pick up new content. A full crawl is only needed for schema changes or to rebuild the index.
Reveal each answer, then mark whether you got it right. Score 60%+ to unlock the next chapter.
To make a custom column searchable, you need to map its crawled property to a managed property. First, ensure the column has data in some items. Perform a full crawl so the crawled property appears. Then, in Search Schema, find the crawled property (e.g., ows_MyColumn) and create a new managed property or map it to an existing one. Set the managed property's searchable flag to true. After mapping, run a full crawl to populate the index. You can then use the managed property in queries (e.g., MyColumn:"value") or as a refiner.
Possible reasons: (1) The document is in an excluded path defined by a crawl rule. (2) The crawl has not completed or failed (check crawl logs). (3) The document's ACL denies the user read access (security trimming). (4) The document is not crawled because of a protocol handler issue (e.g., unsupported file type). (5) The managed property used in the query is not set as searchable. (6) The document was recently added and continuous crawl hasn't picked it up yet (wait up to 5 minutes). Check each of these systematically.
A result source defines the scope of content to search (e.g., local SharePoint, a specific site collection, or an external provider). It is set on a search vertical or used as a source for search web parts. A query rule modifies the behavior of a query after it is executed, such as promoting specific results, adding result blocks, or changing ranking. Result sources limit where to search; query rules change what appears in the results. They work together: a query rule can be scoped to a specific result source.
By default, SharePoint Online uses continuous crawl with a 5-minute interval. This means the crawler checks for changes every 5 minutes. You can change this interval between 1 and 15 minutes. There is also a full crawl that runs automatically based on Microsoft's schedule (typically weekly). You cannot schedule a full crawl manually in SharePoint Online; you can only trigger it via PowerShell or request it via support.
Yes, you can create a crawl rule to exclude a specific document library. In Search Administration, go to Crawl Rules and create a new rule. Enter the path to the library (e.g., https://tenant.sharepoint.com/sites/mysite/libraryname) and set it to Exclude. You can also exclude by file type (e.g., *.mp4) or by content type. After adding the rule, you must start a full crawl to remove already indexed items from the library.
First, create a result source that searches only the Contracts site collection. Go to Search > Result Sources, click 'New Result Source', name it 'Contracts', choose 'Local SharePoint', and add a query transform: {?{searchTerms} site:https://tenant.sharepoint.com/sites/contracts}. Then, in the Search Center site, edit the Search Results page and add a new search vertical tab. In the tab properties, set the result source to the 'Contracts' source. You can also customize the display template to show specific metadata.
You can have up to 100 query rules per search application. Each rule can have multiple conditions and actions. There is no limit on the number of promoted results within a rule, but performance may degrade with too many. Rules are evaluated in priority order (0-100). Only the first matching rule applies. You can also set start and end dates for time-bound rules.
You've just covered SharePoint Search Administration — now see how well it sticks with free MS-102 practice questions. Full explanations included, no account needed.
Done with this chapter?