Join GA4 Traffic Acquisition CSVs with Ad Spend to Measure ROAS

How to export GA4 Traffic acquisition data for spend blending

Google Analytics 4 imports Google Ads cost automatically when linked, but non-Google channels (Meta Ads, LinkedIn Ads, sponsorships) require manual cost imports via Campaign data import (utm_source and utm_medium) or an external join.

1. In Google Analytics 4, open Reports > Acquisition > Traffic acquisition.
2. Switch the primary dimension dropdown from Session primary channel group (Default Channel Group) to Session source / medium or Session campaign, and add Date or Month as a secondary dimension by clicking the + button.
3. Click Share this report > Download File > Download CSV (exports up to 100,000 rows as of October 2026).

Key attribution formulas and SQL equivalents

1. Channel Customer Acquisition Cost (CAC) and ROAS

Once your GA4 CSV with cost data is open in the xmetric app, compute effective CAC and ROAS side by side as calculated fields. (Joining a separate ad spend CSV on normalized campaign and month is coming.)

xmetric / Tableau formula (CAC):

SUM([Ad Spend]) / NULLIF(SUM([Key events]), 0)

xmetric / Tableau formula (ROAS):

SUM([Total revenue]) / NULLIF(SUM([Ad Spend]), 0)

SQL equivalent:

SELECT
  g."Session source / medium" AS source_medium,
  g."Session campaign" AS campaign,
  SUM(s.spend_usd) AS ad_spend,
  SUM(g."Sessions") AS sessions,
  SUM(g."Key events") AS key_events,
  SUM(g."Total revenue") AS revenue,
  SUM(s.spend_usd) / NULLIF(SUM(g."Key events"), 0) AS cac,
  SUM(g."Total revenue") / NULLIF(SUM(s.spend_usd), 0) AS roas
FROM ga4_traffic_acquisition g
LEFT JOIN ad_spend_csv s
  ON LOWER(TRIM(g."Session campaign")) = LOWER(TRIM(s.campaign_name))
GROUP BY 1, 2
ORDER BY ad_spend DESC;

2. Share of Spend vs. Share of Key Events (FIXED LOD)

Spot channels that consume a large share of paid budget while delivering a small share of total conversions.

xmetric / Tableau formula:

(SUM([Key events]) / MIN({ FIXED : SUM([Key events]) }))
- (SUM([Ad Spend]) / MIN({ FIXED : SUM([Ad Spend]) }))

SQL equivalent:

SELECT
  "Session primary channel group (Default Channel Group)" AS channel_group,
  SUM("Key events") * 1.0 / NULLIF(SUM(SUM("Key events")) OVER (), 0) AS share_of_key_events,
  SUM(ad_spend) * 1.0 / NULLIF(SUM(SUM(ad_spend)) OVER (), 0) AS share_of_spend
FROM joined_channel_data
GROUP BY 1;

3. Cost per Engaged Session (Quality-Adjusted CPC)

Ad platforms report cost per link click, which ignores visitors who bounce before the page loads. Dividing spend by GA4 Engaged sessions reveals true cost per qualified visit.

xmetric / Tableau formula:

SUM([Ad Spend]) / NULLIF(SUM([Engaged sessions]), 0)

Where GA4 native attribution and paid connectors differ

• Session-scoped vs. data-driven attribution: The GA4 Traffic acquisition report uses the session-scoped Paid and organic channels last click attribution model for Session source / medium, whereas GA4's Advertising > Attribution reports default to data-driven attribution across touchpoints. If you want multi-touch credit weights from GA4, export the Model comparison CSV from the Advertising workspace instead.
• Data Studio (formerly Looker Studio) 3rd-party connector costs: Data Studio (renamed from Looker Studio in April 2026; free self-service tier, Pro at $9 per user per project per month as of October 2026) includes free built-in connectors for GA4 and Google Ads, but blending Meta Ads, LinkedIn Ads, or TikTok Ads in Data Studio usually requires a paid third-party partner connector. In xmetric, you drop the exported CSV into your browser for free (blending several CSVs is coming).