Enterprise Product
SCTT
Search Configuration & Testing Toolkit
Know if your search changes made things better or worse — before you ship them.
The Problem
Search tuning is a guessing game.
When you tune search — changing boosts, synonyms, field weights, ranking logic — you have no reliable way to know if things got better or worse. You make a change, push it, and hope. SCTT replaces that guesswork with data.
Change synonym mapping for “oil filter”
Boost in-stock items by field weight
Adjust relevancy ranking for category pages
Deploy to production → did it actually help?
SCTT: Compare BEFORE vs AFTER. Know the answer.
How It Works
A repeatable pipeline for measuring search quality.
Define your test queries. Capture results before and after a config change. Let SCTT score the difference.
1
Build Sample
Define a set of test queries that represent real user searches — “battery”, “oil filter”, “spark plugs”, etc. This becomes your reusable test suite.
SCTT2
Extract — BEFORE Snapshot
Run your test queries against the current search configuration and capture the ranked results as a baseline.
SCTT3
Make Your Config Change
Update synonyms, boosting rules, schema tweaks, field weights — whatever you’re testing. This is your change, in your environment.
Your change4
Extract — AFTER Snapshot
Run the same queries again. SCTT captures the new ranked results for direct comparison.
SCTT5
Summarize
Automatically compute match scores, position deltas, row count changes, and query time differences across every query in your test suite.
SCTT6
Compare
View BEFORE vs AFTER side-by-side with color-coded position changes. Green = same position. Amber = moved. Red = missing. Quantified, not guessed.
SCTTKey Features
Built for search engineers, not marketers.
Multi-Engine Support
Connects to Apache Solr, Elasticsearch, OpenSearch, Typesense, and Meilisearch. One tool for any search stack.
Reusable Script Assets
Build a library of scripts shared across test configurations. Full code editor with syntax highlighting and asset linking.
Real-Time Execution
Script Runner with live terminal output. Run pipeline stages individually or in sequence with dynamic argument configuration.
Position Delta Scoring
Tracks which products moved up, moved down, appeared, or disappeared. Match scores quantify the impact of every change.
CSV Export & Feedback
Export summary results to CSV. Query feedback tracking for human-graded quality scores — build an auditable record over time.
Self-Hosted
Runs locally or on your own servers. No SaaS dependency, no data sharing. Your search data never leaves your infrastructure.
Who It’s For
Built for teams that take search seriously.
Search Engineers
- Validate relevancy tuning before production deploys
- Catch regressions in ranking logic across query sets
- Build reusable test suites that mirror real user behavior
- Compare results across Solr, Elasticsearch, or OpenSearch
E-Commerce & Product Teams
- Quantify search improvements for stakeholders
- Track whether synonym and boost changes improve key queries
- Maintain a quality baseline as catalog and content changes
- Ship with confidence instead of guessing
Screenshots
See it in action.
PIPELINE TESTS
Test configuration with multi-stage pipeline
COMPARE VIEW
BEFORE vs AFTER with color-coded position deltas
SCRIPT ASSETS
Reusable script editor with asset management
SCRIPT RUNNER
Execute with live terminal output and dynamic args
Why SCTT
Measurable. Repeatable. Yours.
5+
Search Engines Supported
A/B
Before vs After Comparison
0
Data Sent Externally
CSV
Export & Audit Trail
Ready to stop guessing?
See how SCTT fits your search workflow. Request a demo or check out the repo.