Measurement, testing, and Preview controls
Use this guide when you need evidence, a controlled comparison, or a safe reproduction of search behavior without creating normal storefront activity.
A/B Test compares eligible VIBE and Shopify native-search sessions.
Search Preview is the safe place to reproduce ranking and Explore behavior.
Analytics
Section titled “Analytics”Select 7, 30, or 90 days. Each value compares with the immediately preceding period of the same length.
Overview
Section titled “Overview”- Sessions usage.
- Total searches.
- Click-through rate.
- Conversion rate.
- Revenue from search.
- No-result rate.
- Daily search volume on advanced plans.
Engagement
Section titled “Engagement”- Viewed products.
- Searches per session.
- Unique products per search and per session.
- Average time to product view.
- Discovery activity by surface.
- Average click position and time to click.
Attribution
Section titled “Attribution”- Direct VIBE revenue and orders.
- Cart-add-confirmed revenue.
- Click and view revenue.
- Shopify or control revenue where present.
- Conversion funnel.
- Revenue by attribution signal.
- Top attributed queries.
Search insights
Section titled “Search insights”- Top queries and products.
- Query-product winners.
- Product opportunities.
- Weak performers.
- Hidden gems.
- Revenue efficiency.
- No-result queries.
- Filter usage.
Analytics reports evidence. It does not prove that one setting caused a change. Compare metrics with catalog, theme, rule, campaign, seasonality, and traffic timing.
Read Analytics.
A/B Test
Section titled “A/B Test”An A/B test splits eligible storefront sessions between VIBE and Shopify native search.
- Start only after setup, index, theme, and major rules are stable.
- While running, review elapsed days, sessions, and both variant metric tables.
- Stop early only for an operational reason. Early results can be noisy.
- The test ends at the configured completion condition or maximum duration.
- A cooldown applies before another test.
- Control-group sessions are not billed as VIBE sessions while the test is active.
Do not start a test while changing catalog structure, theme integration, or major merchandising rules. The comparison becomes difficult to interpret.
Read A/B testing.
Search Preview
Section titled “Search Preview”Search tab
Section titled “Search tab”Runs a diagnostic query against the active index and shows:
- Ranked results.
- Overall and semantic scores.
- Indexed product data.
- Sync state.
- Query understanding.
- Product-level diagnostic detail.
Browse tab
Section titled “Browse tab”Runs Explore-style requests with active controls and starting collections.
Preview requests do not add normal storefront analytics or billable shopper sessions. Storefront results can still differ because of language, market, filters, control positions, rules, and presentation mode.
Read Search Preview.
Which tool to open first
Section titled “Which tool to open first”| Question | Start with |
|---|---|
| Why is this product missing or low? | Search Preview |
| Which queries produce no results? | Analytics > Search insights |
| Are shoppers clicking useful results? | Analytics > Engagement |
| Which searches contribute to orders? | Analytics > Attribution |
| Is VIBE outperforming the native baseline? | A/B Test |
| Did a targeted change help? | Preview first, then an equal Analytics period |
Verification checklist
Section titled “Verification checklist”- Use the same date-range length for before and after comparisons.
- Record the exact release or setting-change date.
- Reproduce important queries in Preview.
- Check whether traffic source, campaign, or product availability changed.
- Separate correlation from causation in conclusions.
- Confirm storefront tracking before relying on downstream reporting.
Next: review plan limits, support paths, system checks, and scheduling a call.
Account and support