​

Email & SMS Attribution

​

Special Attribution Rules for Email & SMS

​

Why Email/SMS Attribution Works Differently

1. Existing Relationship Required: Unlike acquisition channels, Email and SMS require a prior relationship with the customer.
2. Skewed Attribution: Without special rules, Email/SMS often receives disproportionate credit in first-touch and linear models.
3. Over-attribution Risk: Including Email/SMS in all attribution models can lead to under-valuing true acquisition channels.

​

Email/SMS Attribution Rules

1. First-Touch Attribution:
   • Email/SMS channels do not receive first-touch attribution credit
   • Credit goes to the earliest non-Email/SMS touch point instead
   • For journeys with only Email/SMS touches, no first-touch attribution is assigned
2. Linear Attribution:
   • Email/SMS touches are excluded from linear attribution
   • Credit is distributed only among non-Email/SMS touches
   • For journeys with only Email/SMS touches, no linear attribution is assigned
3. Last-Touch Attribution:
   • Special customer-specific rules apply
   • By default, Email/SMS is excluded from last-touch attribution
   • For specific customers (currently only Elix Healing), Email/SMS can receive last-touch attribution
   • This configuration is managed via the is_email_sms_last_touch_enabled setting
4. Email/SMS-Only Journeys:
   • Journeys where all touches are Email/SMS are generally not attributable
   • These journeys show as “unattributed” in attribution reports
   • Exception: Last-touch attribution for specific customers

​

Dedicated Email/SMS Dimension

​

Email/SMS Dimension Features

• Message ID Extraction: Identifies specific email or SMS messages using:
  1. utm_id (primary - used by Klaviyo and other ESPs)
  2. utm_content (fallback option)
  3. utm_term (secondary fallback)
• Display Format: [channel]message_id
  • Example: “[email]123456” or “[sms]789012”
• Connection to Message Performance: Links to engagement metrics from rpt_outbound_message_performance_daily

​

Benefits of the Dedicated Dimension

• Acquisition Accuracy: By excluding Email/SMS from first touch and linear models, you get a clearer picture of which channels truly acquire customers
• Email/SMS Visibility: Through the dedicated dimension, you still see the impact of Email/SMS on conversions
• Flexible Customer Configuration: Customer-specific settings allow for business model adaptations

​

Analyzing Email/SMS Performance

​

Attribution Reports

1. Last-Touch Analysis (for enabled customers):
   • See which Email/SMS messages are most effective at closing sales
   • Compare last-touch attribution to Email/SMS performance metrics
2. Email/SMS Dimension Analysis:
   • Track which messages appear in customer journeys
   • Identify the Email/SMS messages most frequently associated with purchases

​

Example Analysis Queries

-- Email messages that appear most frequently in purchase journeys
SELECT
  dimension_value.email_sms as message_id,
  COUNT(DISTINCT purchase_order_id) as journey_count
FROM `your_project.sm_experimental.obt_purchase_journeys_with_mta_models`
WHERE
  sm_store_id = 'your-sm_store_id'
  AND dimension_value.email_sms IS NOT NULL
GROUP BY 1
ORDER BY 2 DESC
LIMIT 20

-- Email performance metrics with last-touch attribution
-- (only for customers with Email/SMS last-touch enabled)
	SELECT
	  r.campaign_name,
	  COUNT(DISTINCT r.message_id) as message_count,
	  SUM(r.message_unique_receives) as deliveries,
	  SUM(r.message_unique_opens) as opens,
	  SUM(r.message_unique_clicks) as clicks,
	  SUM(a.last_touch_revenue) as last_touch_revenue,
	  SAFE_DIVIDE(SUM(a.last_touch_revenue), NULLIF(SUM(r.message_unique_receives), 0)) as revenue_per_delivery
	FROM `your_project.sm_transformed_v2.rpt_outbound_message_performance_daily` r
	LEFT JOIN (
	  SELECT
	    dimension_value.email_sms as message_id,
	    SUM(last_touch_revenue_impact.email_sms) as last_touch_revenue
  FROM `your_project.sm_experimental.obt_purchase_journeys_with_mta_models`
  WHERE
    sm_store_id = 'your-sm_store_id'
    AND sm_event_name = 'purchase'
    AND last_touch_revenue_impact.email_sms > 0
  GROUP BY 1
) a ON r.message_id = a.message_id
WHERE
  r.sm_store_id = 'your-sm_store_id'
  AND r.sm_message_channel IN ('email', 'sms')
  AND r.date BETWEEN DATE_SUB(CURRENT_DATE(), INTERVAL 30 DAY) AND CURRENT_DATE()
GROUP BY 1
ORDER BY 7 DESC

​

Business Impact

1. More Accurate Channel Valuation: Acquisition channels get proper credit for bringing in new customers
2. Better Budget Allocation: Prevents overinvestment in email at the expense of true acquisition channels
3. Clearer Marketing Funnel: Distinguishes between acquisition and retention/conversion channels
4. Adaptable to Business Models: Customer-specific settings accommodate different business needs