Promotion Click-Through Rate

Also known as: Internal Promotion CTR, Promo CTR, Onsite Banner CTR

EngagementkpiE-commerce: Product Discovery & Merchandising (Search/Filters/PDP)E-commerce

Internal promotion click-through rate is the share of viewable on-site internal promotion impressions (hero banners, carousels, category tiles, PDP and cart cross-sell slots) that a shopper clicks, measured as select_promotion over view_promotion for each promotion_id, creative_slot and location_id. It covers only promotions the retailer serves on its own storefront, not paid-media, email, affiliate or marketplace ad CTR, so Merchandising can judge which placements and creatives earn attention before downstream conversion is considered.

Business Use Case

1️⃣ Site Merchandising: at the Monday trading review, any homepage or category creative_slot whose CTR has been more than 20% below its trailing 8-week slot median for 7 consecutive days goes back to the content team for a creative swap or reprioritisation in the promotion calendar.,2️⃣ Conversion Rate Optimisation (Product): when testing hero-banner layouts or tile copy, the team ships the variant with the higher CTR at 95% confidence only if Merchandising Conversion Rate for the same slot does not decline, so clickbait creatives that attract clicks without sales are rejected.,3️⃣ Trade Marketing: for vendor-funded or co-op placements, the team includes slot-level impressions, clicks and CTR in the end-of-flight brand-partner recap and uses the benchmark against the slot median to price placement renewals.,4️⃣ Digital Product (UX): when mobile CTR for a location_id is less than 60% of desktop CTR for the same promotion over a full week, the UX team audits tap-target size, carousel swipe behaviour and above-the-fold placement on mobile templates.

Business Criticality

High

Journey / Navigation Area

Discovery

Owning Function

Merchandising

Scope

Event

Formula

(Deduplicated select_promotion paired to a viewable impression in the same session / Deduplicated viewable view_promotion) × 100, capped at 100%, where each event is deduplicated to one per session × promotion_id × creative_slot × location_id

Calculation Notes

1️⃣ Denominator counts view_promotion only when the creative is viewable (at least 50% of its pixels in the viewport for at least 1 continuous second, per the IAB/MRC display standard). Slides in a carousel that never render, lazy-loaded slots below the fold that are never scrolled to, and page-load pushes fired before render are excluded, because they inflate impressions and understate CTR.,2️⃣ Both events are deduplicated to one per session × promotion_id × creative_slot × location_id. Carousel auto-rotation, back-button restores and SPA re-renders therefore do not create extra impressions, and double-taps or repeat clicks on the same tile count as a single selection.,3️⃣ A select_promotion with no matching viewable view_promotion in the same session (for example from cached pages, deep links that replay the click, or consent granted after the impression) is excluded from the numerator. The slot-level ratio is capped at 100%, and any slot above 100% before capping is sent to the tagging owner as an instrumentation defect.,4️⃣ Scope is on-site internal promotions only. Paid search, paid social, display, email, SMS, affiliate and retail-media sponsored-product placements are excluded, even when they carry a promotion_id. Known bots, internal and agency IP ranges, and promotions whose promotion_id uses the staging or preview prefix are also filtered out. In GA4 the traffic_type parameter comes from the property's internal-traffic rules. In Adobe the site sets page.trafficType (external, internal, agency or developer) in the data layer, and it is mapped to eVar38; internal, agency and developer traffic is dropped.,5️⃣ Sessions end after 30 minutes of inactivity. An impression and a click must fall in the same session to pair, and the day boundary is the property reporting time zone. GA4 BigQuery daily tables settle within 72 hours, so the prior three days are restated on each load and a period is frozen after that.,6️⃣ Clicks earned in the hero slot are not comparable with clicks from a footer tile. Rank creatives only within the same creative_slot and location_id, and only when the cell has at least 1,000 deduplicated impressions. For A/B decisions, test the difference in proportions at 95% confidence, not by eyeballing rank.,7️⃣ The GA4 standard report metric 'Item promotion click-through rate' is item-scoped (items clicked in promotion / items viewed in promotion) and will differ from this event-scoped definition when promotions carry several items[]. Name the variant explicitly in any dashboard, and use the item-scoped figure only for product-level promotion analysis.

Technical Details

Measure TypePercentage

Aggregation WindowDaily, Weekly, Monthly

Events

Platform	Event
Google Analytics 4	select_promotion
Adobe	promotionClick

Data Mappings

W3 Data Layer

{
  "page": {
    "pageInfo": {
      "pageID": "home",
      "pageName": "home",
      "destinationURL": "https://www.example.com/"
    },
    "category": {
      "primaryCategory": "home",
      "pageType": "home"
    },
    "attributes": {
      "trafficType": "external"
    }
  },
  "promotion": [
    {
      "promotionInfo": {
        "promotionID": "PROMO_SUMMER_2026_HERO",
        "promotionName": "Summer Sale 2026",
        "creativeName": "summer_sale_hero_v2",
        "creativeSlot": "hero_banner_1",
        "locationID": "homepage_top",
        "promotionType": "internal"
      },
      "attributes": {
        "viewable": true,
        "impressionTimestamp": "2026-09-16T14:05:11Z",
        "clicked": true,
        "clickTimestamp": "2026-09-16T14:05:19Z"
      }
    }
  ],
  "event": [
    {
      "eventInfo": {
        "eventName": "select_promotion",
        "eventAction": "promotionClick",
        "type": "internal_promotion",
        "timeStamp": "2026-09-16T14:05:19Z"
      },
      "category": {
        "primaryCategory": "merchandising"
      },
      "attributes": {
        "promotionID": "PROMO_SUMMER_2026_HERO",
        "creativeSlot": "hero_banner_1",
        "locationID": "homepage_top"
      }
    }
  ]
}

GA4 Data Layer

{
  "event": "select_promotion",
  "page_type": "home",
  "promotion_type": "internal",
  "viewable": true,
  "ecommerce": {
    "promotion_id": "PROMO_SUMMER_2026_HERO",
    "promotion_name": "Summer Sale 2026",
    "creative_name": "summer_sale_hero_v2",
    "creative_slot": "hero_banner_1",
    "location_id": "homepage_top",
    "items": [
      {
        "item_id": "SKU_10452",
        "item_name": "Linen Short-Sleeve Shirt",
        "item_brand": "Coastline",
        "item_category": "Apparel",
        "item_category2": "Men",
        "item_list_id": "homepage_hero",
        "item_list_name": "Homepage Hero",
        "promotion_id": "PROMO_SUMMER_2026_HERO",
        "promotion_name": "Summer Sale 2026",
        "creative_name": "summer_sale_hero_v2",
        "creative_slot": "hero_banner_1",
        "location_id": "homepage_top",
        "index": 0,
        "price": 39.99,
        "quantity": 1
      }
    ]
  }
}

Adobe Client Data Layer

{
  "event": "promotionClick",
  "eventInfo": {
    "path": "promotion.PROMO_SUMMER_2026_HERO",
    "ga4EventEquivalent": "select_promotion",
    "impressionEvent": "promotionView"
  },
  "page": {
    "pageType": "home",
    "pageName": "home",
    "trafficType": "external"
  },
  "promotion": {
    "PROMO_SUMMER_2026_HERO": {
      "promotionID": "PROMO_SUMMER_2026_HERO",
      "promotionName": "Summer Sale 2026",
      "creativeName": "summer_sale_hero_v2",
      "creativeSlot": "hero_banner_1",
      "locationID": "homepage_top",
      "promotionType": "internal",
      "viewable": true,
      "linkedProductSKUs": [
        "SKU_10452"
      ]
    }
  }
}

XDM Mapping

{
  "eventType": "web.webinteraction.linkClicks",
  "timestamp": "2026-09-16T14:05:19Z",
  "web": {
    "webPageDetails": {
      "name": "home"
    },
    "webInteraction": {
      "name": "promotionClick",
      "type": "other",
      "linkClicks": {
        "value": 1
      }
    }
  },
  "productListItems": [
    {
      "SKU": "SKU_10452",
      "name": "Linen Short-Sleeve Shirt",
      "priceTotal": 39.99,
      "quantity": 1
    }
  ],
  "_experience": {
    "analytics": {
      "customDimensions": {
        "eVars": {
          "eVar30": "PROMO_SUMMER_2026_HERO",
          "eVar31": "Summer Sale 2026",
          "eVar32": "summer_sale_hero_v2",
          "eVar33": "hero_banner_1",
          "eVar34": "homepage_top",
          "eVar35": "internal",
          "eVar36": "true",
          "eVar37": "home",
          "eVar38": "external"
        }
      },
      "event1to100": {
        "event30": {
          "value": 0
        },
        "event31": {
          "value": 1
        }
      }
    }
  }
}

SQL Queries

Google Analytics 4 (BigQuery)

-- ==========================================
-- 1. GOOGLE BIGQUERY SQL
-- ==========================================
-- Promotion Click-Through Rate: deduplicated select_promotion / deduplicated viewable view_promotion,
-- one per session x promotion_id x creative_slot x location_id; clicks count only when paired with an impression in the same session.
-- Promotion fields live in items[] in the GA4 export; location_id falls back to the event-level parameter.
WITH base AS (
  SELECT
    PARSE_DATE('%Y%m%d', event_date) AS event_date,
    event_name,
    user_pseudo_id,
    device.category AS device_category,
    items,
    (SELECT MAX(IF(ep.key = 'ga_session_id', ep.value.int_value, NULL)) FROM UNNEST(event_params) AS ep) AS ga_session_id,
    (SELECT MAX(IF(ep.key = 'page_type', ep.value.string_value, NULL)) FROM UNNEST(event_params) AS ep) AS page_type,
    (SELECT MAX(IF(ep.key = 'location_id', ep.value.string_value, NULL)) FROM UNNEST(event_params) AS ep) AS location_id_param,
    (SELECT MAX(IF(ep.key = 'promotion_type', ep.value.string_value, NULL)) FROM UNNEST(event_params) AS ep) AS promotion_type,
    (SELECT MAX(IF(ep.key = 'traffic_type', ep.value.string_value, NULL)) FROM UNNEST(event_params) AS ep) AS traffic_type,
    (SELECT MAX(IF(ep.key = 'viewable', COALESCE(ep.value.string_value, CAST(ep.value.int_value AS STRING)), NULL)) FROM UNNEST(event_params) AS ep) AS viewable
  FROM `your-project.analytics_123456789.events_*`
  WHERE _TABLE_SUFFIX BETWEEN FORMAT_DATE('%Y%m%d', DATE_SUB(CURRENT_DATE(), INTERVAL 30 DAY))
                          AND FORMAT_DATE('%Y%m%d', DATE_SUB(CURRENT_DATE(), INTERVAL 1 DAY))
    AND event_name IN ('view_promotion', 'select_promotion')
),
promo_events AS (
  SELECT
    b.event_date,
    b.event_name,
    CONCAT(b.user_pseudo_id, '.', CAST(b.ga_session_id AS STRING)) AS session_key,
    b.device_category,
    COALESCE(b.page_type, '(not set)') AS page_type,
    it.promotion_id,
    it.promotion_name,
    it.creative_name,
    COALESCE(NULLIF(it.creative_slot, ''), '(not set)') AS creative_slot,
    COALESCE(NULLIF(it.location_id, ''), NULLIF(b.location_id_param, ''), '(not set)') AS location_id,
    b.viewable
  FROM base AS b, UNNEST(b.items) AS it
  WHERE it.promotion_id IS NOT NULL
    AND it.promotion_id != ''
    AND b.ga_session_id IS NOT NULL
    AND LOWER(COALESCE(b.promotion_type, 'internal')) = 'internal'
    AND LOWER(COALESCE(b.traffic_type, '')) NOT IN ('internal', 'agency', 'developer')
    AND NOT REGEXP_CONTAINS(LOWER(it.promotion_id), r'^(stg|staging|preview)[_-]')
),
impressions AS (
  SELECT
    session_key,
    promotion_id,
    creative_slot,
    location_id,
    MIN(event_date) AS event_date,
    ANY_VALUE(promotion_name) AS promotion_name,
    ANY_VALUE(creative_name) AS creative_name,
    ANY_VALUE(device_category) AS device_category,
    ANY_VALUE(page_type) AS page_type
  FROM promo_events
  WHERE event_name = 'view_promotion'
    AND LOWER(viewable) IN ('true', '1')
  GROUP BY session_key, promotion_id, creative_slot, location_id
),
clicks AS (
  SELECT DISTINCT session_key, promotion_id, creative_slot, location_id
  FROM promo_events
  WHERE event_name = 'select_promotion'
),
paired AS (
  SELECT
    i.*,
    IF(c.session_key IS NOT NULL, 1, 0) AS clicked
  FROM impressions AS i
  LEFT JOIN clicks AS c
    USING (session_key, promotion_id, creative_slot, location_id)
)
SELECT
  event_date,
  promotion_id,
  ANY_VALUE(promotion_name) AS promotion_name,
  ANY_VALUE(creative_name) AS creative_name,
  creative_slot,
  location_id,
  page_type,
  device_category,
  COUNT(*) AS promotion_impressions,
  SUM(clicked) AS promotion_clicks,
  ROUND(LEAST(SAFE_DIVIDE(SUM(clicked), COUNT(*)), 1) * 100, 2) AS promotion_ctr_pct,
  COUNT(*) >= 1000 AS meets_min_volume
FROM paired
GROUP BY event_date, promotion_id, creative_slot, location_id, page_type, device_category
ORDER BY event_date DESC, promotion_impressions DESC;

Adobe Experience Platform (AEP Query Service)

-- ==========================================
-- 2. ADOBE AEP QUERY SERVICE SQL
-- ==========================================
-- Promotion Click-Through Rate: promotionClick (event31) / viewable promotionView (event30),
-- deduplicated per session x promotion ID (eVar30) x creative slot (eVar33) x location ID (eVar34);
-- only viewable (eVar36) internal promotions (eVar35) from non-internal/agency traffic (eVar38, sourced from data layer page.trafficType), broken down by page type (eVar37);
-- sessions use a 30-minute inactivity timeout and clicks count only when paired with an impression in the same session.
WITH sessionized AS (
  SELECT
    identityMap['ECID'][0].id AS ecid,
    timestamp,
    device.type AS device_category,
    _experience.analytics.customDimensions.eVars.eVar37 AS page_type,
    _experience.analytics.customDimensions.eVars.eVar35 AS promotion_type,
    _experience.analytics.customDimensions.eVars.eVar36 AS viewable,
    _experience.analytics.customDimensions.eVars.eVar38 AS traffic_type,
    _experience.analytics.customDimensions.eVars.eVar30 AS promotion_id,
    _experience.analytics.customDimensions.eVars.eVar31 AS promotion_name,
    _experience.analytics.customDimensions.eVars.eVar32 AS creative_name,
    COALESCE(NULLIF(_experience.analytics.customDimensions.eVars.eVar33, ''), '(not set)') AS creative_slot,
    COALESCE(NULLIF(_experience.analytics.customDimensions.eVars.eVar34, ''), '(not set)') AS location_id,
    COALESCE(_experience.analytics.event1to100.event30.value, 0) AS promo_view,
    COALESCE(_experience.analytics.event1to100.event31.value, 0) AS promo_click,
    SESS_TIMEOUT(timestamp, 60 * 30) OVER (
      PARTITION BY identityMap['ECID'][0].id
      ORDER BY timestamp
      ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW
    ) AS session
  FROM aep_web_events_dataset
  WHERE timestamp >= date_sub(current_date(), 30)
    AND timestamp < current_date()
    AND identityMap['ECID'][0].id IS NOT NULL
),
promo_hits AS (
  SELECT
    CONCAT(ecid, '-', CAST(session.num AS STRING)) AS session_key,
    to_date(timestamp) AS event_date,
    COALESCE(device_category, '(not set)') AS device_category,
    COALESCE(NULLIF(page_type, ''), '(not set)') AS page_type,
    viewable,
    promotion_id,
    promotion_name,
    creative_name,
    creative_slot,
    location_id,
    promo_view,
    promo_click
  FROM sessionized
  WHERE promotion_id IS NOT NULL
    AND promotion_id != ''
    AND (promo_view > 0 OR promo_click > 0)
    AND LOWER(COALESCE(promotion_type, 'internal')) = 'internal'
    AND LOWER(COALESCE(traffic_type, '')) NOT IN ('internal', 'agency', 'developer')
    AND NOT (LOWER(promotion_id) RLIKE '^(stg|staging|preview)[_-]')
),
impressions AS (
  SELECT
    session_key,
    promotion_id,
    creative_slot,
    location_id,
    MIN(event_date) AS event_date,
    MAX(promotion_name) AS promotion_name,
    MAX(creative_name) AS creative_name,
    MAX(device_category) AS device_category,
    MIN(page_type) AS page_type
  FROM promo_hits
  WHERE promo_view > 0
    AND LOWER(viewable) IN ('true', '1')
  GROUP BY session_key, promotion_id, creative_slot, location_id
),
clicks AS (
  SELECT DISTINCT session_key, promotion_id, creative_slot, location_id
  FROM promo_hits
  WHERE promo_click > 0
),
paired AS (
  SELECT
    i.event_date,
    i.promotion_id,
    i.promotion_name,
    i.creative_name,
    i.creative_slot,
    i.location_id,
    i.page_type,
    i.device_category,
    CASE WHEN c.session_key IS NOT NULL THEN 1 ELSE 0 END AS clicked
  FROM impressions AS i
  LEFT JOIN clicks AS c
    ON  i.session_key = c.session_key
    AND i.promotion_id = c.promotion_id
    AND i.creative_slot = c.creative_slot
    AND i.location_id = c.location_id
)
SELECT
  event_date,
  promotion_id,
  MAX(promotion_name) AS promotion_name,
  MAX(creative_name) AS creative_name,
  creative_slot,
  location_id,
  page_type,
  device_category,
  COUNT(*) AS promotion_impressions,
  SUM(clicked) AS promotion_clicks,
  ROUND(LEAST(SUM(clicked) / NULLIF(COUNT(*), 0), 1.0) * 100, 2) AS promotion_ctr_pct,
  COUNT(*) >= 1000 AS meets_min_volume
FROM paired
GROUP BY event_date, promotion_id, creative_slot, location_id, page_type, device_category
ORDER BY event_date DESC, promotion_impressions DESC;

Source Data

Digital Analytics (GA4)

Dependencies

Prerequisite: List the dependencies required for this KPI

Events:

view-promotionselect-promotion

Dimensions:

promotion-typepage-typedevice-category

Report Attributes

promotion_id promotion_name creative_name creative_slot location_id page_type device_category

Recommended Reporting Views

Views

Onsite Merchandising PerformanceHomepage & Category Landing OptimisationPromotion Calendar & Placement ReviewVendor-Funded Placement Recap

Segment Eligibility

true

Governance

Data SensitivityInternal

Governance

Created bydevyendarm

Created on9/16/2026

Last modified bydevyendarm

Last modified on9/18/2026

Version1.0.0

Validation stateqa passed

StatusPUBLISHED

GitHub

View related Pull Requestcatalog/kpis/promotion-click-through-rate.yml

Promotion Click-Through Rate

Business Use Case

Business Criticality

Journey / Navigation Area

Owning Function

Scope

Formula

Calculation Notes

Technical Details

Events

Data Mappings

SQL Queries

Source Data

Dependencies

Events:

Dimensions:

Report Attributes

Recommended Reporting Views

Segment Eligibility

Governance

Related KPIs

Governance

GitHub

Community Discussion