What is ROAS, and how do you measure it in Metabase?

ROAS — return on ad spend — is conversion value divided by ad spend, per campaign, channel, or period. It's the efficiency read for paid media:CAC tells you what a customer costs, ROAS tells you what a dollar of ads brings back. Measure it in Metabase from daily performance data synced from Google Ads, Meta Ads, and the rest of your paid stack.

What ROAS measures

It measures how much conversion value a unit of spend generates, which makes it the day-to-day steering metric for budget allocation within a channel. It does not measure profit — a 5x ROAS on a 15% margin product loses money — and it does not measure incrementality, because the platform reporting the number also grades its own homework. Segment it by campaign objective: prospecting, retargeting, and brand campaigns play different roles and should be judged against different bars.

ROAS vs. MER

Per-platform ROAS answers "which campaign should get the next dollar?" Blended MER — total revenue ÷ total marketing spend — answers "is the whole program working?" Because each platform self-reports conversions under its own attribution window, summed platform ROAS routinely exceeds what your revenue table can support. The fix isn't picking a winner; it's charting both and treating a widening gap as the signal that platforms are double-claiming conversions.

What data does it need?

• An ad_performance_daily table with campaign_name, channel, stat_date, spend, and conversion_value, unioned across platforms.
• An orders (or revenue) table for the blended MER comparison — the one number no ad platform can inflate.

SQL patterns

SELECT
  campaign_name,
  date_trunc('week', stat_date) AS week,
  SUM(spend) AS spend,
  SUM(conversion_value) AS conversion_value,
  ROUND(SUM(conversion_value) / NULLIF(SUM(spend), 0), 2) AS roas
FROM ad_performance_daily
WHERE stat_date >= CURRENT_DATE - INTERVAL '90 days'
GROUP BY 1, 2
ORDER BY 2, 1;

WITH channel_spend AS (
  SELECT
    channel,
    date_trunc('month', stat_date) AS month,
    SUM(spend) AS spend,
    SUM(conversion_value) AS platform_conversion_value
  FROM ad_performance_daily
  GROUP BY 1, 2
),
blended_revenue AS (
  SELECT
    date_trunc('month', order_date) AS month,
    SUM(order_total) AS revenue
  FROM orders
  GROUP BY 1
)
SELECT
  c.month,
  c.channel,
  ROUND(c.platform_conversion_value / NULLIF(c.spend, 0), 2)
    AS platform_roas,
  ROUND(
    r.revenue / NULLIF(SUM(c.spend) OVER (PARTITION BY c.month), 0),
    2
  ) AS blended_mer
FROM channel_spend c
JOIN blended_revenue r ON r.month = c.month
ORDER BY 1, 2;

Pitfalls

Where this metric applies

• Google Ads + Metabase — campaign spend and conversion value reports
• Meta Ads + Metabase — ad set performance with attributed purchase value
• TikTok Ads + Metabase — campaign-level spend and conversions
• LinkedIn Ads + Metabase — B2B campaign spend and conversion tracking

Related

Metrics

Dashboards

FAQ