FlatSearch.Pro Enterprise Documentation

Configure faster, smarter WooCommerce search.

This guide explains how to configure FlatSearch.Pro’s search engine, Fast Index, search accuracy, caching, product discovery, images and commerce actions.

The current architecture supports the deeply optimized Flatsome integration and a theme-independent FlatSearch.Pro search surface for WooCommerce.

Fast Index Engine Universal WooCommerce search Optional Cloudflare edge Optional Redis/object cache Production HAR tested

Current search architecture

Core requirements versus optional acceleration.

WC
WooCommerce Commerce catalogue and product data
Core
IDX
Fast Index Engine Prepared high-speed search data
Built in
UI
Universal Search Surface Theme-independent form / icon option
Built in
CF
Cloudflare Fast Search Shared edge acceleration
Optional
R
Redis / Object Cache Persistent server cache acceleration
Optional
Recommended first setup

Five steps to a production-ready search.

Start with the core search path first. Add optional cache and presentation features after search results are correct.

1

Choose the search surface

Keep the optimized Flatsome integration or use FlatSearch.Pro’s universal WooCommerce search form / icon.

2

Enable Fast Index

Build the prepared product index, then use the rebuild tool once to generate current commerce-search data.

3

Enable search accuracy

Start with Better Word Matching, Typo Fix and the multi-word / first-letter capabilities appropriate for your catalogue.

4

Add optional acceleration

Cloudflare Fast Search and persistent object cache can improve repeat and high-traffic search performance.

5

Validate in Incognito

Test exact, typo, multi-word and repeat searches on desktop and mobile before considering the configuration finished.

Recommended

Change one layer at a time.

Search correctness first. Then measure caching, frontend optimization and optional features separately.

Best practice: build Fast Index → confirm search accuracy → enable cache acceleration → validate frontend UX → capture a HAR only when deeper network analysis is needed.
Current search route

One canonical search contract. Multiple acceleration layers.

FlatSearch.Pro no longer needs a different search meaning for each speed mode. Browser memory, shared cache, Cloudflare and Fast Index accelerate the same commerce-search contract.

LAYER 01

Instant UI

Available local/browser data can paint immediately while the canonical search settles.

LAYER 02

Browser Cache

Repeated queries may be reused locally without another full search round trip.

LAYER 03

Shared / Edge Cache

Public query responses can be reused across visitors when valid.

LAYER 04

Fast Index

The prepared WooCommerce search index answers uncached origin queries.

LAYER 05

Safe Fallback

Compatibility paths remain available when an optional acceleration layer is unavailable.

Canonical browser transport: current validated search traffic uses the FlatSearch.Pro search endpoint such as /fsp-search.js?q=keyword&n=20 rather than requiring a separate request for every typed character.
Network ownership

Settled-query coordination

Transient typing can be coalesced so the browser does not need a full network search for every prefix.

Current runtime

No deliberate aborts

Once a canonical request has started, the current runtime can let it finish while stale-response guards prevent old results from repainting the current query.

Search parity

One search meaning

Speed layers are designed to accelerate search without creating a different relevance definition for cached and uncached requests.

Production HAR evidence

Know what good performance looks like.

These figures are measurements from our production test environment. They are useful diagnostic references, not guaranteed latency for every hosting configuration.

29.5ms Recorded cached edge search Observed during production Incognito HAR testing.
~30–50ms Warm edge-search range Observed across validated cached production queries.
1× Canonical settled query Validated final flows avoid a per-character network request storm.
0 Deliberate search aborts Final validated runtime completed without FlatSearch-owned cancelled search requests.
Cold cache is different: a first Cloudflare MISS can include CDN-to-origin fill time that is outside the internal FlatSearch search pipeline. Always distinguish origin search computation from browser-visible cold-fill latency.
Theme-independent option

FlatSearch.Pro can provide its own WooCommerce search surface.

The original Flatsome integration remains deeply optimized, but the search UI is no longer limited to Flatsome’s native search component.

Search form

Universal Search Form

Renders a FlatSearch-owned WooCommerce search form where required.

Shortcode [flat_search_pro_form]
Search icon

Universal Search Icon

Adds a compact FlatSearch-owned search icon / modal trigger.

Shortcode [flat_search_pro_icon]
Important: theme-independent does not mean every heavily customized theme has been individually certified. Custom headers, builders, CSS, z-index rules and script optimizers can still require presentation-level validation.
Performance

Speed settings and acceleration layers.

These settings reduce repeated PHP, database, network or browser work. Not every store needs every layer.

Core speed engine

Fast Index Engine

Builds prepared WooCommerce search data so uncached searches can avoid expensive general-purpose catalogue work.

  • Recommended for most stores.
  • Rebuild after relevant catalogue/index changes.
  • Especially useful for medium and large catalogues.
Optional edge

Cloudflare Fast Search

Allows public search responses to be reused at Cloudflare where the store’s infrastructure supports it.

  • Optional — not required for core search.
  • Best value comes from repeated public queries.
  • Use DevTools to distinguish HIT from MISS.
Browser acceleration

Browser Quick Search

Reuses suitable previously seen result data in the shopper’s browser so repeated interactions can appear faster.

  • Useful for repeated searches.
  • Does not require Cloudflare.
  • Should preserve the same search result contract.
Persistent cache

Memory Cache Boost

Uses a persistent object cache such as Redis when available to reduce repeated server-side work.

  • Optional.
  • Useful on busy WooCommerce stores.
  • Not required for search correctness.
Product memory

Product List Memory

Keeps useful product-search data ready for low-latency result presentation and compatible fallback behavior.

Instant display

Instant Result

Helps FlatSearch.Pro display useful search results immediately while authoritative search transport remains coordinated behind the interface.

Repeat queries

Repeated Search Cache

Reuses valid results for commonly repeated normalized search terms.

Slow-result recovery

Slow Search Cache

Stores selected expensive search responses so an unusually slow query does not necessarily require the same work every time.

Background preparation

Search Cache Warm-Up

Can prepare selected common search terms in the background.

  • Use selectively.
  • Avoid warming large numbers of random queries.
  • Requires healthy scheduled task execution.
Product navigation

Product Result Preload

Can prepare selected product navigation after search without aggressively waking every product-page resource.

  • Use safe HTML-oriented behavior.
  • Avoid redundant preload from multiple plugins.
  • Do not preload large sets of product images unnecessarily.
Database diagnostics

Database Speed Check

Helps confirm that the store has the expected database/index support for efficient product search.

Results page

Search Results Page Cache

Caches the full submitted search-results page separately from the small live autocomplete request.

Relevance

Help shoppers find what they mean.

These modules improve search interpretation while keeping one consistent relevance contract across supported speed paths.

Words

Better Word Matching

Improves matching for product names containing multiple meaningful words and product-family terms.

  • Useful for long product titles.
  • Exact matches should remain strong.
  • Works with Fast Index search.
Typo

Typo Fix

Helps common misspellings reach the intended product or brand without requiring perfect spelling.

  • Example: civas → Chivas.
  • Useful for difficult brand names.
  • Exact queries should not be weakened by typo logic.
Multi-word

Multi-Word Matching

Coordinates searches containing several words so important terms can contribute together instead of behaving like unrelated queries.

Price intent

Price Search Words

Lets shoppers use natural budget phrases such as:

  • whisky under 300
  • wine above 100
  • between 100 and 250
First letters

Search From First Letter

Can begin useful discovery from short prefixes while the canonical network request remains coordinated.

Content

Page & Blog Search

Allows relevant published pages or posts to appear alongside commerce search where enabled.

  • Useful for guides and help content.
  • Product results remain important for commerce queries.
Search experience

Make fast search look fast too.

Perceived speed depends on result presentation, images and interaction — not only server timing.

Main UI

Search Box Design

Applies FlatSearch.Pro’s result presentation including search rows, prices, popular/recent terms, View More and supported mobile presentation.

Images

Better Search Images

Chooses search-appropriate WooCommerce image URLs rather than unnecessarily using oversized source images.

Protection

Small Image Guard

Helps prevent unsuitable, oversized or broken thumbnail choices from degrading the search dropdown.

Current image system

Viewport Image Admission

The current frontend prioritizes images actually visible inside the dropdown viewport while off-screen result images remain deferred.

  • Visible thumbnails can start together.
  • Off-screen rows do not need immediate image requests.
  • Scrolling admits newly visible rows.
Focus

Background Dimmer

Dims surrounding page content where supported so attention remains on the search experience.

Discovery

Popular & Recent Searches

Shows useful discovery terms before or around search activity and supports the View More search journey.

Mobile

Mobile Keyboard Helper

Helps supported mobile search surfaces focus the input and bring up the keyboard when necessary.

Optional CDN

Cloudflare Image Resizing

Optional image-transformation support for stores that intentionally use Cloudflare Image Resizing.

Avoid applying another transformation to an image that is already correctly sized for the search dropdown.
Commerce

Search can move beyond discovery.

FlatSearch.Pro can expose shopping actions and guided product recommendations directly inside the search experience.

Conversion

Quick Add

Eligible products can expose Add to Cart actions directly from search.

  • Optional module.
  • Static assets load conditionally.
  • Test special product types separately.
Guided discovery

Guided Answer Search

Shows curated answer cards and associated products for selected search topics.

  • Store-controlled content.
  • Can recommend real WooCommerce products.
  • Supports product selection and cart actions.
Full results

Search Page Sorting

Controls how the full WooCommerce search-results page prioritizes available and relevant products after submission.

Guided Answer

Curated answers that lead to real products.

Guided Answer Search is store-controlled content rather than an uncontrolled live AI response.

Store Guided Answer Example products
Drinks Cocktail recipe or gifting guide Spirit, mixer, garnish, glassware
Beauty Beginner skincare routine Cleanser, serum, moisturizer
Electronics Gaming setup guide Keyboard, mouse, headset, monitor
Fashion Travel outfit guide Shirt, pants, shoes, bag
Home New-home starter kit Bedding, towels, kettle, storage
Submitted search

Full results-page settings.

These controls apply after the shopper presses Enter or chooses View More.

Request efficiency

Search Cleanup Mode

Reduces unrelated work during supported search-result requests where it is safe to do so.

Commerce sorting

Search Page Sorting

Helps prioritize suitable and available products on the submitted search-results page.

Result count

Dropdown Result Limit

Controls how many autocomplete results are returned for a settled query.

10–20 is a practical range for many stores. The interface can display more rows while viewport image admission keeps hidden thumbnails deferred.
Recent terms

Recent Search Limit

Keep recent-term presentation compact. Use 0 to disable it or a small number such as 3–8.

Frontend protection

Delay JS Search Protection

Helps preserve search usability when a performance plugin delays JavaScript execution.

  • Relevant to WP Rocket and similar optimizers.
  • Validate after changing JS-delay settings.
  • Do not exclude scripts unnecessarily when everything already works.
Compatibility

Let performance tools accelerate FlatSearch.Pro — not redefine it.

Layer Requirement Guidance
Cloudflare Optional Useful for shared edge search caching. Core Fast Index search does not depend on Cloudflare.
Redis / Object Cache Pro Optional Useful for persistent server caching but not required for search correctness.
WP Rocket Compatible with validation Re-test search after changing JavaScript delay, cache or asset optimization settings.
Other JS optimizers Validate Aggressive script removal, combination or delay can affect any interactive search product.
Server/page cache Generally compatible Purge after relevant FlatSearch.Pro frontend updates to avoid stale asset references.
Operations

Tools, diagnostics and reporting.

Status

Admin Status

Quickly inspect active search modes, index readiness and important FlatSearch.Pro state.

Analytics

Search Analytics

Review popular queries and search behavior to understand customer demand and discovery gaps.

Debug

Debug Report

Collects useful environment, module and search-readiness details for diagnosis.

Cache

Clear Cache

Use cache clearing after relevant search/index changes rather than purging everything repeatedly without reason.

Validation

Test FlatSearch.Pro like a shopper and like an engineer.

The best configuration is the one that produces correct results, clean browser behavior and good perceived speed on the actual store.

Search

Test relevance

  • Full product name
  • Brand or product family
  • SKU where applicable
  • Multi-word query
  • Common typo
  • Price-intent query
Network

Test canonical transport

  • Open browser DevTools → Network.
  • Type a representative query.
  • Look for the canonical /fsp-search.js request.
  • Confirm there is no unnecessary per-character network storm.
  • Confirm there are no red cancelled FlatSearch search rows.
Cache

Test cold versus warm

  • Use Incognito for a clean first-load test.
  • Check Cloudflare HIT / MISS where enabled.
  • Repeat the exact same query.
  • Do not compare a cold MISS directly with a browser-cached repeat.
Images

Test viewport thumbnails

  • Visible search images should appear quickly.
  • Hidden off-screen rows should not require every thumbnail immediately.
  • Scroll and confirm newly visible images load correctly.
  • Watch for duplicate transformations or broken CDN URLs.
Troubleshooting

Diagnose the layer that is actually slow or incorrect.

Search accuracy, PHP origin time, Cloudflare cold fill, browser cache and product images are different layers. Avoid changing the ranking engine to solve an image or CDN problem.

Search shows “No products found” too early.
Check Fast Index readiness and the Search From First Letter / accuracy settings. Clear relevant search cache after changing them, then test from Incognito.
Fast Index results look different from expected search results.
Rebuild the product index, clear relevant caches and verify Better Word Matching, Typo Fix, Price Search Words, multi-word behavior and content-search settings. Speed providers should preserve the same intended search meaning.
I see many /fsp-search.js requests while typing.
The current architecture is designed to coalesce transient typing and give one settled query canonical network ownership. Check for stale JavaScript/page cache, duplicate plugin versions or another search script creating a competing transport.
I see red cancelled FlatSearch requests in DevTools.
Current no-abort coordination should not deliberately cancel a started canonical FlatSearch request. Purge stale frontend assets and verify the browser is loading the current runtime.
Search is approximately 30–50 ms when warm but much slower after cache clear.
Check whether the slow request is a Cloudflare MISS. A cold CDN-to-origin fill can be much slower than the internal Fast Index computation. Measure origin/search timing and edge-fill timing separately before changing search SQL or relevance.
Search images are loading too many requests.
Verify the current viewport-aware image runtime is active. Visible rows should be prioritized while hidden dropdown rows remain deferred. Also check that another plugin is not independently preloading the same product images.
Search images are grey, missing or broken.
Check Better Search Images and Small Image Guard, verify the underlying WooCommerce thumbnail exists, and inspect whether an image CDN or transformation service has rewritten an already valid search thumbnail.
WP Rocket optimization causes search problems.
Test JavaScript delay/defer settings first. Keep Delay JS Search Protection enabled where appropriate. Exclude only the specific FlatSearch asset that is demonstrably affected rather than disabling optimization broadly.
After changing settings, the frontend still behaves like the old version.
Clear the relevant WordPress/page cache, persistent object cache where applicable, Cloudflare cache where applicable, and retest in Incognito. Browser cache can otherwise continue serving old static assets.
WordPress shows a critical error.
Check the server PHP error log or wp-content/debug.log. Use the exact fatal-error file and line to determine whether the problem originates from FlatSearch.Pro, WooCommerce, the theme or another plugin.
When should I capture a HAR?
Use a HAR when the problem involves timing, duplicate requests, cancelled requests, cache status, images, frontend transport or a difference between cold and repeated search behavior. Capture from Incognito where possible.
Maintenance

Clean Uninstall

Decide whether FlatSearch.Pro data should remain available if the plugin is removed.

Destructive option

Remove stored plugin data

Clean Uninstall can remove FlatSearch.Pro options, index data and associated plugin storage during deletion.

Leave Clean Uninstall disabled if there is any chance the plugin will be installed again later.
Need deeper diagnosis?

Send the search evidence, not just the symptom.

A FlatSearch.Pro Debug Report plus an Incognito HAR can show the active search modules, canonical request, cache state, images and browser-visible timing.