'%20clip-path='url(%23canvas-clip)'/%3e%3cg%20clip-path='url(%23canvas-clip)'%3e%3cg%20transform='rotate(0,%20256,%20256)'%3e%3cg%20transform='translate(87%2087)%20scale(14.083333333333334%2014.083333333333334)%20translate(0%200)'%3e%3cpath%20fill='none'%20stroke='%23FFFFFF'%20stroke-linecap='round'%20stroke-linejoin='round'%20stroke-width='2'%20d='m18.064%2010.877l-4.89-7.26c-.42-.625-1.287-.803-1.936-.397a1.4%201.4%200%200%200-.41.397l-4.893%207.26C4.24%2013.715%204.9%2017.318%207.502%2019.423a7.13%207.13%200%200%200%204.168%201.572M15%2018a3%203%200%201%200%206%200a3%203%200%201%200-6%200m5.2%202.2L22%2022'/%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/svg%3e)
ParticleSearch Revenue Attribution: Direct, Assisted, and Verified Order Evidence
Search revenue attribution should answer a narrow question: which completed Shopify orders can be responsibly associated with a recorded search journey? It should not turn every search before checkout into a causal revenue claim.
ParticleSearch keeps direct product evidence separate from broader session assistance, verifies order facts when Shopify access is available, and leaves unmatched orders visible. This guide explains how to use that evidence for prioritization, evaluation, and follow-up.
Attribution behavior checked July 28, 2026. This article was verified against the current ParticleSearch dashboard, storefront event contract, Shopify order verification path, and attribution builders. It intentionally describes merchant-visible behavior, not private implementation details.
The short answer
Use attribution to prioritize work, not to claim causality
ParticleSearch revenue attribution answers a practical question: which completed orders had a captured search touchpoint, and how strong is the product or session evidence? That is valuable for finding high-value queries, products, and repair paths. It is not an experiment, and it cannot prove that search created an order or that one rule generated incremental lift.
Direct match
A purchased product matches a product opened or added from search. Strongest inspection context.
Assisted match
Search appears in the captured path, but the purchased product was not directly tied to a product action. Useful, weaker context.
Unattributed
The order is visible, but the available evidence cannot connect it to search. Keep it in the denominator.
Chapter 1 · The attribution job
Connect the journey without claiming more than the evidence supports
A storefront can record searches and product actions. Shopify owns the order. Useful attribution connects those two evidence sets, checks the order facts, and communicates the strength of the match.
This is different from incrementality. Attribution describes an observed association under a defined window. Incrementality asks whether the order would have happened without the search experience. That second question needs a controlled comparison.
Search journey
Queries, result interactions, filters, product opens, and direct adds create context.
Checkout completed
A Shopify checkout event provides a candidate order connection.
Order verified
When access is available, ParticleSearch verifies order totals and line items against Shopify.
Confidence assigned
The order is classified as direct, assisted, or unattributed under the available evidence.
Chapter 2 · Confidence levels
Direct, assisted, and unattributed are not interchangeable
Direct product match
Evidence
A purchased product matches a product that the recorded search journey opened or added.
Use it for
Inspect which queries and products appear closest to an order and prioritize high-value failure paths.
Boundary
Strong association does not prove search caused the purchase or that a particular rule created lift.
Assisted session match
Evidence
Search occurred earlier in the captured session, but the purchased product was not directly matched to a recorded product action.
Use it for
Understand journeys where search may have helped discovery before the shopper continued elsewhere.
Boundary
This is weaker context. It should not be reported as a direct product-level conversion.
Unattributed order
Evidence
The order is visible to the reporting window, but the available search evidence does not support an association.
Use it for
Keep the commercial denominator honest and investigate coverage or normal non-search journeys.
Boundary
Unattributed does not mean search had no influence. It means the captured evidence cannot support the claim.
Missing evidence is not evidence of zero value. Consent choices, browser state, cross-device journeys, unavailable Shopify permissions, reporting windows, and untracked surfaces can all reduce visible attribution. Report the coverage boundary beside the result.
Chapter 3 · Read a complete journey
The same order can carry different evidence depending on the path before it
Attribution becomes useful when a merchant can read the chain, not just the final label. The following examples use ordinary shopper paths and deliberately avoid a made-up revenue amount. The point is to show what the evidence supports and where the claim stops.
Direct product match
Observed path
A shopper searches for a replacement filter, opens the matching product, selects a variant, adds it, and later completes an order containing that product.
Responsible reading
The order can be reported as directly associated with the captured search journey. That is strong operational evidence for prioritization, not proof that search caused the order or that one rule created incremental lift.
Assisted session match
Observed path
A shopper searches for a material guide, reads an article, returns through a collection, and later orders a product without a recorded product action from the original search.
Responsible reading
The search may have helped the journey, but the evidence is weaker than a direct product match. Keep the classification separate so a useful discovery path does not inflate product-level conversion.
Unattributed order
Observed path
An order is visible in Shopify, but no complete search chain is available for the selected window or the shopper never used a captured search surface.
Responsible reading
Keep the order in the denominator and call it unattributed. Do not turn missing evidence into zero influence, and do not assign the order to search merely because a query was recorded elsewhere.
Chapter 4 · Read the dashboard
Use the funnel and attribution views for different questions
The funnel shows observed stages: search sessions, searches, product action sessions, and orders. Attribution describes the order associations supported by the available evidence. Neither view should silently fill a missing stage.
A sharp drop between stages is a place to inspect, not an automatic diagnosis. Product fit, buying cycle, variant selection, price, inventory, checkout, and measurement coverage can all affect the path.
Tracked orders and revenue
Shopify order value available to the selected reporting context.
Provides the commercial denominator for attributed and unattributed outcomes.
Search-attributed orders and revenue
Orders and value associated with a captured search journey under the active evidence rules.
Use the direct and assisted split before summarizing the total.
Revenue per search session
Associated revenue divided by observed search sessions in the reporting window.
Useful for comparable periods, not as a universal benchmark across stores.
Query revenue
Associated value grouped by the recorded query context.
Use it to inspect high-value paths, not to award all order value to a word.
Product revenue
Associated value grouped by products connected to search activity.
Compare with exposure, clicks, direct adds, availability, and margin context.
Rule impact evidence
Before-and-after behavior around a published search control, normalized to observed search volume.
Directional evidence only. Other changes can affect the same period.
Attribution answers “what can we connect?”
It joins recorded search context to an order under an explicit window and evidence rule. That is valuable for deciding which query, product, or experience deserves attention next.
An experiment answers “what changed because of it?”
It compares eligible control and variant experiences. Use it when two plausible choices need a causal comparison, and keep attribution as supporting context rather than a substitute for the comparison.
Chapter 5 · Merchant decisions
Use revenue context to prioritize, protect, and investigate
Which search failures deserve attention first?
Combine repeated query problems with nearby revenue evidence. A rare curiosity should not displace a common high-value failure.
Which successful paths should be protected?
Identify queries that repeatedly lead to product actions and verified orders, then add them to your acceptance tests before large changes.
Where does the journey lose commercial continuity?
Compare search sessions, searches, product action sessions, and attributed orders to locate the stage that needs inspection.
Did a published control improve the path?
Read comparable before-and-after evidence, confirm the rule scope, and check protected queries. Treat the result as directional unless a controlled experiment isolates the change.
Does search behave differently by market or locale?
Inspect context splits when the data is sufficient, then verify translation, catalog availability, currency, and query wording on the storefront.
Chapter 6 · Data quality
Check the evidence boundary before sharing the revenue number
Coverage
Are the intended storefront search surfaces and checkout evidence connected?
Verification
Can order totals and line items be confirmed against Shopify?
Window
Does the selected period fit the available analytics history and buying cycle?
Currency
Are values being compared within compatible currency contexts?
Consent and storage
Could privacy choices or browser state reduce the visible journey?
Volume
Is there enough direct or assisted evidence to support the intended conclusion?
Release context
Did pricing, campaigns, inventory, theme code, or search settings change too?
A credible summary names the method. Report the period, tracked order coverage, direct and assisted split, currency boundary, and any known instrumentation change. Then describe the decision the evidence supports.
For query and product behavior before the order, use the ParticleSearch search analytics metrics guide. For causal comparison between product experiences, use a controlled ParticleSearch experiment.
Chapter 7 · Evaluate attribution during a trial
A useful trial proves the chain, not a flattering percentage
Trace a direct path
Use a controlled query, open or add a known product, complete a test order, and verify the product connection and Shopify order facts.
Trace an assisted path
Search, continue through another route, and verify that the dashboard keeps the weaker classification separate from a direct product match.
Protect the boundary
Confirm that unmatched orders remain visible, missing revenue is not silently treated as zero, and the report states what the evidence cannot prove.